Discussion:
[Audacity-devel] Planning Ahead for 2.1.4
James Crook
2016-09-04 10:09:17 UTC
Permalink
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).


Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.

Possibilities for a 2.2.0 designation include:

- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)

With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.

Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant. It
would be nice to do a "dot zero" release, but we don't have to. There
will be plenty enough without.

We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.

The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.

Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.

--James.











------------------------------------------------------------------------------
Mark Young
2016-09-04 10:15:10 UTC
Permalink
Hi,

I have a further Timer Recording enhancement I'm working on that will allow
processes to be executed at different stages of the Timer Recording. It
would for example allow a user to launch a Web stream and then kill it
after the recording. I'd like to submit it when ready for Audacity vNext.

Mark
Post by James Crook
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant. It
would be nice to do a "dot zero" release, but we don't have to. There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
James Crook
2016-09-04 10:39:55 UTC
Permalink
Post by Mark Young
Hi,
I have a further Timer Recording enhancement I'm working on that will allow
processes to be executed at different stages of the Timer Recording. It
would for example allow a user to launch a Web stream and then kill it
after the recording. I'd like to submit it when ready for Audacity vNext.
Please say more about it. We currently have a policy of not opening
TCP/IP connections from Audacity itself (opening a browser is OK),
because we don't have the developer bandwidth to audit TCP/IP code
(including within wxWidgets) to the levels that would be needed. So I
assume this is wxShell()'ing a script. Please say a bit more. Where
will the script itself be stored?

--James.


------------------------------------------------------------------------------
Vaughan Johnson
2016-09-04 12:30:10 UTC
Permalink
James: "...we should be thinking about what we want to have in that
release. We should be doing some thinking right now,"



Writing class? "should be ... thinking" twice in 2 consecutive sentences?
yet ur comfortable complaining about my using 'u'.

===

James: "...and maybe
some coding, in readiness for that release cycle"



Hum, as we've done for years, w/ #pragmas. ur not suggesting anything new
, James.

===

James: "Possibilities for a 2.2.0 designation include:

- Vaughan's 2-in-1 (touch screen mode)
"



It's spelled 2n1. i'm certainly not motivated these days , either by
morale (ongoing attacks at me) or $.

-- but repeated thx to Peter & Gale for defending me.

===

Thx,
Vaughan
Post by James Crook
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant. It
would be nice to do a "dot zero" release, but we don't have to. There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
David Engebretson Jr
2016-09-04 16:29:09 UTC
Permalink
As a daily user of Audacity, and an aspiring midi user, and also an aspiring
Audacity developer, I vote for Roger's midi. Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.

If there is a path to getting ASIO built in, I will be happy to use my
gentle demeanor to approach Steinberg regarding licensing. Please, to those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness. I don't want to hear
the same old negative "general" commentary regarding ASIO ish's. Maybe some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.

Peace,
David
***@comcast.net



-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4

We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).


Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.

Possibilities for a 2.2.0 designation include:

- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)

With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.

Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant. It
would be nice to do a "dot zero" release, but we don't have to. There
will be plenty enough without.

We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.

The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.

Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.

--James.
Peter Sampson
2016-09-05 17:07:12 UTC
Permalink
Scrubbing Phase-3 - further improve the user interface for Audacity's
Scrubbing
http://wiki.audacityteam.org/wiki/Proposal:_Improvements_to_Scrubbing_-_Phase-3

Could benefit from being on this 2.1.4 planning agenda.

Peter

On Sun, Sep 4, 2016 at 5:29 PM, David Engebretson Jr <
Post by David Engebretson Jr
As a daily user of Audacity, and an aspiring midi user, and also an aspiring
Audacity developer, I vote for Roger's midi. Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.
If there is a path to getting ASIO built in, I will be happy to use my
gentle demeanor to approach Steinberg regarding licensing. Please, to those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness. I don't want to hear
the same old negative "general" commentary regarding ASIO ish's. Maybe some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.
Peace,
David
-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant. It
would be nice to do a "dot zero" release, but we don't have to. There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Paul Licameli
2016-09-05 18:16:23 UTC
Permalink
I have done more than I wanted to do in 2.1.3 for scrubbing, flying without
a spec, trying too many experiments, and failing to please everybody. I
gave in to demands for it, but I wanted to work on quality not features, as
I thought we had agreed in winter, and finish my sweep of memory allocation
code instead. But as it is, some improvement in scrubbing got accelerated
to 2.1.3, and my sweep is in an unfinished state yet.

I would happily do nothing for scrubbing for the next release. If others
feel competent to implement the Phase 3 ideas, if that's mostly about
redefining the buttons and clicks and appearance of the ruler, I will let
them. If redoing speed control is involved, or yet more engine
improvements, that requires more serious thought and I would want to be
involved. I would not be happy to see the disctinction of pinned and
unpinned scrubbing retracted.

My own opinion about unfinished scrubbing priorities is that I would rather
focus on the problem of making a decent keyboard-only alternative
interface, and making that work with jog wheel peripherals like that
Contour ShuttlePro. That could be orthogonal to the other questions about
improving the graphical version.

But instead of doing even that: I hear much from people who use -- or
disdain to use -- Audacity for recording narration. The most insistent
thing I hear is that it needs a puch and roll recording feature to be a
better tool for that use.

I would like to have that feature in 2.1.4 in at least a crude form.
(Which in fact I have implemented privately, over a year ago. I regret
sitting on it whenever I hear the reiterated demands for it.)

A more sophisticated form would not be a simple destructive re-recording,
but include some way of remembering alternative takes and alternating among
them. That opens the door to the big question of layering of clips, which
would have implications beyond just this recording feature. I don't think
I want to be that ambitious in 2.1.4, but it's a thing to think about for
yet further future.

PRL




On Mon, Sep 5, 2016 at 1:07 PM, Peter Sampson <
Post by Peter Sampson
Scrubbing Phase-3 - further improve the user interface for Audacity's
Scrubbing
http://wiki.audacityteam.org/wiki/Proposal:_Improvements_
to_Scrubbing_-_Phase-3
Could benefit from being on this 2.1.4 planning agenda.
Peter
On Sun, Sep 4, 2016 at 5:29 PM, David Engebretson Jr <
Post by David Engebretson Jr
As a daily user of Audacity, and an aspiring midi user, and also an aspiring
Audacity developer, I vote for Roger's midi. Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.
If there is a path to getting ASIO built in, I will be happy to use my
gentle demeanor to approach Steinberg regarding licensing. Please, to those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness. I don't want to hear
the same old negative "general" commentary regarding ASIO ish's. Maybe some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.
Peace,
David
-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant. It
would be nice to do a "dot zero" release, but we don't have to. There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Gale Andrews
2016-09-06 01:25:34 UTC
Permalink
Thanks, Paul.

I think punch-in fits in well with changes James is pondering
to make default record "same track" rather than "new track".

I think there was a general spec for scrubbing, but many
different models that had that same general aim. That
spec did not come until after we abandoned CTRL-click
scrub.

Would your destructive scrub (as I think your current
rough idea is) lay any sort of base or have reusability for
layered-non destructive scrub? My impression of
scrubbing was that we had no clear specification at
the outset and we have at the moment a number of
related features with different interfaces.



Gale
Post by Paul Licameli
I have done more than I wanted to do in 2.1.3 for scrubbing, flying without
a spec, trying too many experiments, and failing to please everybody. I
gave in to demands for it, but I wanted to work on quality not features, as
I thought we had agreed in winter, and finish my sweep of memory allocation
code instead. But as it is, some improvement in scrubbing got accelerated
to 2.1.3, and my sweep is in an unfinished state yet.
I would happily do nothing for scrubbing for the next release. If others
feel competent to implement the Phase 3 ideas, if that's mostly about
redefining the buttons and clicks and appearance of the ruler, I will let
them. If redoing speed control is involved, or yet more engine
improvements, that requires more serious thought and I would want to be
involved. I would not be happy to see the disctinction of pinned and
unpinned scrubbing retracted.
My own opinion about unfinished scrubbing priorities is that I would rather
focus on the problem of making a decent keyboard-only alternative interface,
and making that work with jog wheel peripherals like that Contour
ShuttlePro. That could be orthogonal to the other questions about improving
the graphical version.
But instead of doing even that: I hear much from people who use -- or
disdain to use -- Audacity for recording narration. The most insistent
thing I hear is that it needs a puch and roll recording feature to be a
better tool for that use.
I would like to have that feature in 2.1.4 in at least a crude form. (Which
in fact I have implemented privately, over a year ago. I regret sitting on
it whenever I hear the reiterated demands for it.)
A more sophisticated form would not be a simple destructive re-recording,
but include some way of remembering alternative takes and alternating among
them. That opens the door to the big question of layering of clips, which
would have implications beyond just this recording feature. I don't think I
want to be that ambitious in 2.1.4, but it's a thing to think about for yet
further future.
PRL
On Mon, Sep 5, 2016 at 1:07 PM, Peter Sampson
Post by Peter Sampson
Scrubbing Phase-3 - further improve the user interface for Audacity's
Scrubbing
http://wiki.audacityteam.org/wiki/Proposal:_Improvements_to_Scrubbing_-_Phase-3
Could benefit from being on this 2.1.4 planning agenda.
Peter
On Sun, Sep 4, 2016 at 5:29 PM, David Engebretson Jr
Post by David Engebretson Jr
As a daily user of Audacity, and an aspiring midi user, and also an aspiring
Audacity developer, I vote for Roger's midi. Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.
If there is a path to getting ASIO built in, I will be happy to use my
gentle demeanor to approach Steinberg regarding licensing. Please, to those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness. I don't want to hear
the same old negative "general" commentary regarding ASIO ish's. Maybe some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.
Peace,
David
-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant. It
would be nice to do a "dot zero" release, but we don't have to. There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
Paul Licameli
2016-09-06 01:35:07 UTC
Permalink
Post by Gale Andrews
Thanks, Paul.
I think punch-in fits in well with changes James is pondering
to make default record "same track" rather than "new track".
I think there was a general spec for scrubbing, but many
different models that had that same general aim. That
spec did not come until after we abandoned CTRL-click
scrub.
Would your destructive scrub (as I think your current
rough idea is) lay any sort of base or have reusability for
layered-non destructive scrub? My impression of
scrubbing was that we had no clear specification at
the outset and we have at the moment a number of
related features with different interfaces.
I don't understand the question. Punch and roll recording, and scrubbing,
are unrelated.

Rather than pursue the perfect scrubbing feature, I want to add a new and
much demanded feature for the narration niche.

PRL
Post by Gale Andrews
Gale
Post by Paul Licameli
I have done more than I wanted to do in 2.1.3 for scrubbing, flying
without
Post by Paul Licameli
a spec, trying too many experiments, and failing to please everybody. I
gave in to demands for it, but I wanted to work on quality not features,
as
Post by Paul Licameli
I thought we had agreed in winter, and finish my sweep of memory
allocation
Post by Paul Licameli
code instead. But as it is, some improvement in scrubbing got
accelerated
Post by Paul Licameli
to 2.1.3, and my sweep is in an unfinished state yet.
I would happily do nothing for scrubbing for the next release. If others
feel competent to implement the Phase 3 ideas, if that's mostly about
redefining the buttons and clicks and appearance of the ruler, I will let
them. If redoing speed control is involved, or yet more engine
improvements, that requires more serious thought and I would want to be
involved. I would not be happy to see the disctinction of pinned and
unpinned scrubbing retracted.
My own opinion about unfinished scrubbing priorities is that I would
rather
Post by Paul Licameli
focus on the problem of making a decent keyboard-only alternative
interface,
Post by Paul Licameli
and making that work with jog wheel peripherals like that Contour
ShuttlePro. That could be orthogonal to the other questions about
improving
Post by Paul Licameli
the graphical version.
But instead of doing even that: I hear much from people who use -- or
disdain to use -- Audacity for recording narration. The most insistent
thing I hear is that it needs a puch and roll recording feature to be a
better tool for that use.
I would like to have that feature in 2.1.4 in at least a crude form.
(Which
Post by Paul Licameli
in fact I have implemented privately, over a year ago. I regret sitting
on
Post by Paul Licameli
it whenever I hear the reiterated demands for it.)
A more sophisticated form would not be a simple destructive re-recording,
but include some way of remembering alternative takes and alternating
among
Post by Paul Licameli
them. That opens the door to the big question of layering of clips,
which
Post by Paul Licameli
would have implications beyond just this recording feature. I don't
think I
Post by Paul Licameli
want to be that ambitious in 2.1.4, but it's a thing to think about for
yet
Post by Paul Licameli
further future.
PRL
On Mon, Sep 5, 2016 at 1:07 PM, Peter Sampson
Post by Peter Sampson
Scrubbing Phase-3 - further improve the user interface for Audacity's
Scrubbing
http://wiki.audacityteam.org/wiki/Proposal:_Improvements_
to_Scrubbing_-_Phase-3
Post by Paul Licameli
Post by Peter Sampson
Could benefit from being on this 2.1.4 planning agenda.
Peter
On Sun, Sep 4, 2016 at 5:29 PM, David Engebretson Jr
Post by David Engebretson Jr
As a daily user of Audacity, and an aspiring midi user, and also an aspiring
Audacity developer, I vote for Roger's midi. Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.
If there is a path to getting ASIO built in, I will be happy to use my
gentle demeanor to approach Steinberg regarding licensing. Please, to those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness. I don't want to
hear
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
the same old negative "general" commentary regarding ASIO ish's. Maybe some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.
Peace,
David
-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be
good
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
for Audacity publicity, and show that the project is active/vibrant.
It
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
would be nice to do a "dot zero" release, but we don't have to. There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights
better,
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to
try
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest
in
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------
------------------
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by Paul Licameli
Post by Peter Sampson
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by Paul Licameli
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Gale Andrews
2016-09-06 01:45:48 UTC
Permalink
Sorry, Paul. I don't blame you not understanding the question.

I meant:

" Would your destructive punch-in (as I think your current
rough idea is) lay any sort of base or have reusability for
layered-non destructive punch-in?

The rest of it was what I meant to write.

Punch-in isn't niche or narrators only. It's always had the same
level of user requests as "scrubbing". Scroll down a little from
this link:
http://wiki.audacityteam.org/wiki/Feature_Requests#Highest-rated .

There is an element of controversy about destructive scrubbing
of course.


Thanks


Gale
Post by Paul Licameli
Post by Gale Andrews
Thanks, Paul.
I think punch-in fits in well with changes James is pondering
to make default record "same track" rather than "new track".
I think there was a general spec for scrubbing, but many
different models that had that same general aim. That
spec did not come until after we abandoned CTRL-click
scrub.
Would your destructive scrub (as I think your current
rough idea is) lay any sort of base or have reusability for
layered-non destructive scrub? My impression of
scrubbing was that we had no clear specification at
the outset and we have at the moment a number of
related features with different interfaces.
I don't understand the question. Punch and roll recording, and scrubbing,
are unrelated.
Rather than pursue the perfect scrubbing feature, I want to add a new and
much demanded feature for the narration niche.
PRL
Post by Gale Andrews
Gale
Post by Paul Licameli
I have done more than I wanted to do in 2.1.3 for scrubbing, flying without
a spec, trying too many experiments, and failing to please everybody. I
gave in to demands for it, but I wanted to work on quality not features, as
I thought we had agreed in winter, and finish my sweep of memory allocation
code instead. But as it is, some improvement in scrubbing got accelerated
to 2.1.3, and my sweep is in an unfinished state yet.
I would happily do nothing for scrubbing for the next release. If others
feel competent to implement the Phase 3 ideas, if that's mostly about
redefining the buttons and clicks and appearance of the ruler, I will let
them. If redoing speed control is involved, or yet more engine
improvements, that requires more serious thought and I would want to be
involved. I would not be happy to see the disctinction of pinned and
unpinned scrubbing retracted.
My own opinion about unfinished scrubbing priorities is that I would rather
focus on the problem of making a decent keyboard-only alternative interface,
and making that work with jog wheel peripherals like that Contour
ShuttlePro. That could be orthogonal to the other questions about improving
the graphical version.
But instead of doing even that: I hear much from people who use -- or
disdain to use -- Audacity for recording narration. The most insistent
thing I hear is that it needs a puch and roll recording feature to be a
better tool for that use.
I would like to have that feature in 2.1.4 in at least a crude form.
(Which
in fact I have implemented privately, over a year ago. I regret sitting on
it whenever I hear the reiterated demands for it.)
A more sophisticated form would not be a simple destructive
re-recording,
but include some way of remembering alternative takes and alternating among
them. That opens the door to the big question of layering of clips, which
would have implications beyond just this recording feature. I don't think I
want to be that ambitious in 2.1.4, but it's a thing to think about for yet
further future.
PRL
On Mon, Sep 5, 2016 at 1:07 PM, Peter Sampson
Post by Peter Sampson
Scrubbing Phase-3 - further improve the user interface for Audacity's
Scrubbing
http://wiki.audacityteam.org/wiki/Proposal:_Improvements_to_Scrubbing_-_Phase-3
Could benefit from being on this 2.1.4 planning agenda.
Peter
On Sun, Sep 4, 2016 at 5:29 PM, David Engebretson Jr
Post by David Engebretson Jr
As a daily user of Audacity, and an aspiring midi user, and also an aspiring
Audacity developer, I vote for Roger's midi. Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.
If there is a path to getting ASIO built in, I will be happy to use my
gentle demeanor to approach Steinberg regarding licensing. Please, to those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness. I don't want to hear
the same old negative "general" commentary regarding ASIO ish's.
Maybe
some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.
Peace,
David
-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant.
It
would be nice to do a "dot zero" release, but we don't have to. There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
Gale Andrews
2016-09-06 02:22:24 UTC
Permalink
Post by Gale Andrews
Sorry, Paul. I don't blame you not understanding the question.
" Would your destructive punch-in (as I think your current
rough idea is) lay any sort of base or have reusability for
layered-non destructive punch-in?
The rest of it was what I meant to write.
Punch-in isn't niche or narrators only. It's always had the same
level of user requests as "scrubbing". Scroll down a little from
http://wiki.audacityteam.org/wiki/Feature_Requests#Highest-rated .
There is an element of controversy about destructive scrubbing
of course.
You said "destructive scrubbing" again. Is that an error again or do I just
not understand.
Clearly it's way too late and I did not get enough sleep last
night (which is true). Yes, element of controversy about
"destructive punch-in". Sorry.

All I'm getting at really is - some might think destructive punch-in
undesirable, and most of us might agree non-destructive punch-in
is preferable. So if destructive punch-in comes first and
non-destructive follows it, would destructive be removed, and
would both use the same interface to operate them?

Most users seem to want the easy destructive method.



Gale
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Thanks, Paul.
I think punch-in fits in well with changes James is pondering
to make default record "same track" rather than "new track".
I think there was a general spec for scrubbing, but many
different models that had that same general aim. That
spec did not come until after we abandoned CTRL-click
scrub.
Would your destructive scrub (as I think your current
rough idea is) lay any sort of base or have reusability for
layered-non destructive scrub? My impression of
scrubbing was that we had no clear specification at
the outset and we have at the moment a number of
related features with different interfaces.
I don't understand the question. Punch and roll recording, and scrubbing,
are unrelated.
Rather than pursue the perfect scrubbing feature, I want to add a new and
much demanded feature for the narration niche.
PRL
Post by Gale Andrews
Gale
Post by Paul Licameli
I have done more than I wanted to do in 2.1.3 for scrubbing, flying without
a spec, trying too many experiments, and failing to please everybody.
I
gave in to demands for it, but I wanted to work on quality not features,
as
I thought we had agreed in winter, and finish my sweep of memory allocation
code instead. But as it is, some improvement in scrubbing got accelerated
to 2.1.3, and my sweep is in an unfinished state yet.
I would happily do nothing for scrubbing for the next release. If others
feel competent to implement the Phase 3 ideas, if that's mostly about
redefining the buttons and clicks and appearance of the ruler, I will let
them. If redoing speed control is involved, or yet more engine
improvements, that requires more serious thought and I would want to be
involved. I would not be happy to see the disctinction of pinned and
unpinned scrubbing retracted.
My own opinion about unfinished scrubbing priorities is that I would rather
focus on the problem of making a decent keyboard-only alternative interface,
and making that work with jog wheel peripherals like that Contour
ShuttlePro. That could be orthogonal to the other questions about improving
the graphical version.
But instead of doing even that: I hear much from people who use -- or
disdain to use -- Audacity for recording narration. The most insistent
thing I hear is that it needs a puch and roll recording feature to be a
better tool for that use.
I would like to have that feature in 2.1.4 in at least a crude form.
(Which
in fact I have implemented privately, over a year ago. I regret sitting
on
it whenever I hear the reiterated demands for it.)
A more sophisticated form would not be a simple destructive re-recording,
but include some way of remembering alternative takes and alternating among
them. That opens the door to the big question of layering of clips, which
would have implications beyond just this recording feature. I don't think I
want to be that ambitious in 2.1.4, but it's a thing to think about for
yet
further future.
PRL
On Mon, Sep 5, 2016 at 1:07 PM, Peter Sampson
Post by Peter Sampson
Scrubbing Phase-3 - further improve the user interface for Audacity's
Scrubbing
http://wiki.audacityteam.org/wiki/Proposal:_Improvements_to_Scrubbing_-_Phase-3
Could benefit from being on this 2.1.4 planning agenda.
Peter
On Sun, Sep 4, 2016 at 5:29 PM, David Engebretson Jr
Post by David Engebretson Jr
As a daily user of Audacity, and an aspiring midi user, and also an
aspiring
Audacity developer, I vote for Roger's midi. Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.
If there is a path to getting ASIO built in, I will be happy to use my
gentle demeanor to approach Steinberg regarding licensing. Please,
to
those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness. I don't want to
hear
the same old negative "general" commentary regarding ASIO ish's.
Maybe
some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.
Peace,
David
-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is
active/vibrant.
It
would be nice to do a "dot zero" release, but we don't have to.
There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to
try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest
in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite
agreement
we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
Paul Licameli
2016-09-06 10:05:37 UTC
Permalink
What are the opinions about the Snow Leopard question? I would like to
target 10.7 just to raise the common denominator of C++ libraries, but that
means nothing to the user's view and should not decide it. Have we
achieved good-enough?

PRL
Post by Gale Andrews
Post by Gale Andrews
Sorry, Paul. I don't blame you not understanding the question.
" Would your destructive punch-in (as I think your current
rough idea is) lay any sort of base or have reusability for
layered-non destructive punch-in?
The rest of it was what I meant to write.
Punch-in isn't niche or narrators only. It's always had the same
level of user requests as "scrubbing". Scroll down a little from
http://wiki.audacityteam.org/wiki/Feature_Requests#Highest-rated .
There is an element of controversy about destructive scrubbing
of course.
You said "destructive scrubbing" again. Is that an error again or do I
just
not understand.
Clearly it's way too late and I did not get enough sleep last
night (which is true). Yes, element of controversy about
"destructive punch-in". Sorry.
All I'm getting at really is - some might think destructive punch-in
undesirable, and most of us might agree non-destructive punch-in
is preferable. So if destructive punch-in comes first and
non-destructive follows it, would destructive be removed, and
would both use the same interface to operate them?
Most users seem to want the easy destructive method.
Gale
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Thanks, Paul.
I think punch-in fits in well with changes James is pondering
to make default record "same track" rather than "new track".
I think there was a general spec for scrubbing, but many
different models that had that same general aim. That
spec did not come until after we abandoned CTRL-click
scrub.
Would your destructive scrub (as I think your current
rough idea is) lay any sort of base or have reusability for
layered-non destructive scrub? My impression of
scrubbing was that we had no clear specification at
the outset and we have at the moment a number of
related features with different interfaces.
I don't understand the question. Punch and roll recording, and scrubbing,
are unrelated.
Rather than pursue the perfect scrubbing feature, I want to add a new and
much demanded feature for the narration niche.
PRL
Post by Gale Andrews
Gale
Post by Paul Licameli
I have done more than I wanted to do in 2.1.3 for scrubbing, flying without
a spec, trying too many experiments, and failing to please
everybody.
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
I
gave in to demands for it, but I wanted to work on quality not features,
as
I thought we had agreed in winter, and finish my sweep of memory allocation
code instead. But as it is, some improvement in scrubbing got accelerated
to 2.1.3, and my sweep is in an unfinished state yet.
I would happily do nothing for scrubbing for the next release. If others
feel competent to implement the Phase 3 ideas, if that's mostly
about
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
redefining the buttons and clicks and appearance of the ruler, I
will
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
let
them. If redoing speed control is involved, or yet more engine
improvements, that requires more serious thought and I would want
to
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
be
involved. I would not be happy to see the disctinction of pinned
and
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
unpinned scrubbing retracted.
My own opinion about unfinished scrubbing priorities is that I
would
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
rather
focus on the problem of making a decent keyboard-only alternative
interface,
and making that work with jog wheel peripherals like that Contour
ShuttlePro. That could be orthogonal to the other questions about
improving
the graphical version.
But instead of doing even that: I hear much from people who use -- or
disdain to use -- Audacity for recording narration. The most insistent
thing I hear is that it needs a puch and roll recording feature to
be
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
a
better tool for that use.
I would like to have that feature in 2.1.4 in at least a crude
form.
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
(Which
in fact I have implemented privately, over a year ago. I regret sitting
on
it whenever I hear the reiterated demands for it.)
A more sophisticated form would not be a simple destructive re-recording,
but include some way of remembering alternative takes and
alternating
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
among
them. That opens the door to the big question of layering of
clips,
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
which
would have implications beyond just this recording feature. I
don't
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
think I
want to be that ambitious in 2.1.4, but it's a thing to think about for
yet
further future.
PRL
On Mon, Sep 5, 2016 at 1:07 PM, Peter Sampson
Post by Peter Sampson
Scrubbing Phase-3 - further improve the user interface for Audacity's
Scrubbing
http://wiki.audacityteam.org/wiki/Proposal:_Improvements_
to_Scrubbing_-_Phase-3
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Could benefit from being on this 2.1.4 planning agenda.
Peter
On Sun, Sep 4, 2016 at 5:29 PM, David Engebretson Jr
Post by David Engebretson Jr
As a daily user of Audacity, and an aspiring midi user, and also
an
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
aspiring
Audacity developer, I vote for Roger's midi. Especially if it is
something
that will give more options to the blindness community and the
ever
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
expanding availability of accessible DAW's.
If there is a path to getting ASIO built in, I will be happy to
use
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
my
gentle demeanor to approach Steinberg regarding licensing.
Please,
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
to
those
who have been nay sayers in the past, I've heard your repetitive
negativity
regarding ASIO and it's logged in my consciousness. I don't want to
hear
the same old negative "general" commentary regarding ASIO ish's.
Maybe
some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.
Peace,
David
-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4
We're probably coming to a quieter spell in 2.1.3 development as
we
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and
translation
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want
to
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could
perhaps
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in
preferences,
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
working with the developer, so that the new feature is as a
module,
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still
with a
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should
be
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
good
for Audacity publicity, and show that the project is
active/vibrant.
It
would be nice to do a "dot zero" release, but we don't have to.
There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have
code
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
from
Bill Unruh for a spectrogram display that preserves peak heights
better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's
MIDI
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to
try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI
code
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest
in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite
agreement
we
are going for 2.2.0). So with this post I am opening a thread
for
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------
------------------
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
Post by David Engebretson Jr
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
Post by Peter Sampson
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
Post by Paul Licameli
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by Gale Andrews
Post by Paul Licameli
Post by Gale Andrews
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by Gale Andrews
Post by Paul Licameli
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by Gale Andrews
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Vaughan Johnson
2016-09-08 12:34:48 UTC
Permalink
I would happily do nothing..."
One of the best beginnings to a sentence that i've ever seen!

- Vaughan

On Wed, Sep 7, 2016 at 5:00 AM, Peter Sampson <
Can I suggest that those of you who are interested in the details of
punch-in
take the ideas for discussion to a new proposal in the Wiki so that we
don't
lose them (which would be a shame).
This thread should not be getting bogged down in implenentation details,
James
started it as a broad-brush forward planning thread to establish what
might or
might not be in 2.1.4.
Thanks,
Peter
Paul, you understand correctly. I investigated this route because it
seemed it might tie up a lot of shortcuts.
Adrian, I hadn't investigated this deeper yet, but I thought that
ShuttlePro can be programmed by the end user to invoke keystroke
shortcuts. I thought all we needed to do was make a keys-only alternative
user interface for controlling scrubbing. I did not think that the
application needs to cooperate in any special way with ShuttlePro. Was I
wrong? Or does this only enable some extra capabilities?
PRL
Hi Paul & others
I had been trying to persuade Contour to let us have the SDK for the
Shuttle Pro and Express as configuring it through their app would tie up a
lot of keyboard short-cuts. Your email prompted me to try again and they
have sent me the SDKs for Mac and Win. I had a quick look at the Windows
one. It uses a specific dll that isn't open source (shuttlesdk.dll) so I
think this makes it a no-go for Audacity, but I would like others opinions.
https://drive.google.com/open?id=0B1cZefvMiXtmb19ybldvaXRTUFU
Adrian
Post by Paul Licameli
I have done more than I wanted to do in 2.1.3 for scrubbing, flying
without a spec, trying too many experiments, and failing to please
everybody. I gave in to demands for it, but I wanted to work on quality
not features, as I thought we had agreed in winter, and finish my sweep of
memory allocation code instead. But as it is, some improvement in
scrubbing got accelerated to 2.1.3, and my sweep is in an unfinished state
yet.
I would happily do nothing for scrubbing for the next release. If
others feel competent to implement the Phase 3 ideas, if that's mostly
about redefining the buttons and clicks and appearance of the ruler, I will
let them. If redoing speed control is involved, or yet more engine
improvements, that requires more serious thought and I would want to be
involved. I would not be happy to see the disctinction of pinned and
unpinned scrubbing retracted.
My own opinion about unfinished scrubbing priorities is that I would
rather focus on the problem of making a decent keyboard-only alternative
interface, and making that work with jog wheel peripherals like that
Contour ShuttlePro. That could be orthogonal to the other questions about
improving the graphical version.
But instead of doing even that: I hear much from people who use -- or
disdain to use -- Audacity for recording narration. The most insistent
thing I hear is that it needs a puch and roll recording feature to be a
better tool for that use.
I would like to have that feature in 2.1.4 in at least a crude form.
(Which in fact I have implemented privately, over a year ago. I regret
sitting on it whenever I hear the reiterated demands for it.)
A more sophisticated form would not be a simple destructive
re-recording, but include some way of remembering alternative takes and
alternating among them. That opens the door to the big question of
layering of clips, which would have implications beyond just this recording
feature. I don't think I want to be that ambitious in 2.1.4, but it's a
thing to think about for yet further future.
PRL
On Mon, Sep 5, 2016 at 1:07 PM, Peter Sampson <
Post by Peter Sampson
Scrubbing Phase-3 - further improve the user interface for Audacity's
Scrubbing
http://wiki.audacityteam.org/wiki/Proposal:_Improvements_to_
Scrubbing_-_Phase-3
Could benefit from being on this 2.1.4 planning agenda.
Peter
On Sun, Sep 4, 2016 at 5:29 PM, David Engebretson Jr <
Post by David Engebretson Jr
As a daily user of Audacity, and an aspiring midi user, and also an aspiring
Audacity developer, I vote for Roger's midi. Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.
If there is a path to getting ASIO built in, I will be happy to use my
gentle demeanor to approach Steinberg regarding licensing. Please, to those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness. I don't want to hear
the same old negative "general" commentary regarding ASIO ish's.
Maybe some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.
Peace,
David
-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is
active/vibrant. It
would be nice to do a "dot zero" release, but we don't have to.
There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Chris Share
2016-09-07 11:43:09 UTC
Permalink
Just wanted to add my 2c to the discussion. I use Logic Pro X for music-related projects. It has a feature called Quick Swipe Comping which I've found to be excellent. See here <https://www.gadgetdaily.xyz/how-to-use-quick-swipe-comping-in-logic-pro-x/>. If there are plans to implement punch-in/punch-out/layering/etc. then I'd definitely have a look at this implementation.
Cheers,
Chris

From: Adrian Wadey <***@gmail.com>
To: "audacity-***@lists.sourceforge.net" <audacity-***@lists.sourceforge.net>
Sent: Wednesday, 7 September 2016, 21:01
Subject: Re: [Audacity-devel] Planning Ahead for 2.1.4

Paul, you understand correctly. I investigated this route because it seemed it might tie up a lot of shortcuts.
On Wed, Sep 7, 2016 at 11:19 AM, Paul Licameli <***@gmail.com> wrote:

Adrian, I hadn't investigated this deeper yet, but I thought that ShuttlePro can be programmed by the end user to invoke keystroke shortcuts.  I thought all we needed to do was make a keys-only alternative user interface for controlling scrubbing.  I did not think that the application needs to cooperate in any special way with ShuttlePro.  Was I wrong?  Or does this only enable some extra capabilities?
PRL

On Wed, Sep 7, 2016 at 5:13 AM, Adrian Wadey <***@gmail.com> wrote:

Hi Paul & others
I had been trying to persuade Contour to let us have the SDK for the Shuttle Pro and Express as configuring it through their app would tie up a lot of keyboard short-cuts. Your email prompted me to try again and they have sent me the SDKs for Mac and Win. I had a quick look at the Windows one. It uses a specific dll that isn't open source (shuttlesdk.dll) so I think this makes it a no-go for Audacity, but I would like others opinions.
The SDK is here:
https://drive.google.com/open? id=0B1cZefvMiXtmb19ybldvaXRTUF U

Adrian

On Mon, Sep 5, 2016 at 7:16 PM, Paul Licameli <***@gmail.com> wrote:

I have done more than I wanted to do in 2.1.3 for scrubbing, flying without a spec, trying too many experiments, and failing to please everybody.  I gave in to demands for it, but I wanted to work on quality not features, as I thought we had agreed in winter, and finish my sweep of memory allocation code instead.  But as it is, some improvement in scrubbing got accelerated to 2.1.3, and my sweep is in an unfinished state yet.
I would happily do nothing for scrubbing for the next release.  If others feel competent to implement the Phase 3 ideas, if that's mostly about redefining the buttons and clicks and appearance of the ruler, I will let them.  If redoing speed control is involved, or yet more engine improvements, that requires more serious thought and I would want to be involved.  I would not be happy to see the disctinction of pinned and unpinned scrubbing retracted.
My own opinion about unfinished scrubbing priorities is that I would rather focus on the problem of making a decent keyboard-only alternative interface, and making that work with jog wheel peripherals like that Contour ShuttlePro.  That could be orthogonal to the other questions about improving the graphical version.
But instead of doing even that:  I hear much from people who use -- or disdain to use -- Audacity for recording narration.  The most insistent thing I hear is that it needs a puch and roll recording feature to be a better tool for that use.
I would like to have that feature in 2.1.4 in at least a crude form.  (Which in fact I have implemented privately, over a year ago.  I regret sitting on it whenever I hear the reiterated demands for it.)
A more sophisticated form would not be a simple destructive re-recording, but include some way of remembering alternative takes and alternating among them.  That opens the door to the big question of layering of clips, which would have implications beyond just this recording feature.  I don't think I want to be that ambitious in 2.1.4, but it's a thing to think about for yet further future.
PRL



On Mon, Sep 5, 2016 at 1:07 PM, Peter Sampson <***@gmail.co m> wrote:

Scrubbing Phase-3 - further improve the user interface for Audacity's Scrubbing
http://wiki.audacityteam.org/w iki/Proposal:_Improvements_to_ Scrubbing_-_Phase-3

Could benefit from being on this 2.1.4 planning agenda.

Peter

On Sun, Sep 4, 2016 at 5:29 PM, David Engebretson Jr <***@comcast.net> wrote:

As a daily user of Audacity, and an aspiring midi user, and also an aspiring
Audacity developer, I vote for Roger's midi.  Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.

If there is a path to getting ASIO built in, I will be happy to use my
gentle demeanor to approach Steinberg regarding licensing.  Please, to those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness.  I don't want to hear
the same old negative "general" commentary regarding ASIO ish's.  Maybe some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.

Peace,
David
***@comcast.net



-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4

We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).


Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release.  We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.

Possibilities for a 2.2.0 designation include:

- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)

With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.

Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features.  I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant.  It
would be nice to do a "dot zero" release, but we don't have to.  There
will be plenty enough without.

We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated.  I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.

The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users.  That gets the code to some users, and maybe
encourages Audacity to use the ideas.  I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.

Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0).  So with this post I am opening a thread for
discussing what we do/plan with 2.1.4.  It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.

--James.











------------------------------ ------------------------------ ------------------
______________________________ _________________
audacity-devel mailing list
audacity-***@lists.sourcefor ge.net
https://lists.sourceforge.net/ lists/listinfo/audacity-devel


------------------------------ ------------------------------ ------------------
______________________________ _________________
audacity-devel mailing list
audacity-***@lists.sourcefor ge.net
https://lists.sourceforge.net/ lists/listinfo/audacity-devel



------------------------------ ------------------------------ ------------------

______________________________ _________________
audacity-devel mailing list
audacity-***@lists.sourcefor ge.net
https://lists.sourceforge.net/ lists/listinfo/audacity-devel




------------------------------ ------------------------------ ------------------

______________________________ _________________
audacity-devel mailing list
audacity-***@lists.sourcefor ge.net
https://lists.sourceforge.net/ lists/listinfo/audacity-devel




------------------------------ ------------------------------ ------------------

______________________________ _________________
audacity-devel mailing list
audacity-***@lists.sourcefor ge.net
https://lists.sourceforge.net/ lists/listinfo/audacity-devel




------------------------------ ------------------------------ ------------------

______________________________ _________________
audacity-devel mailing list
audacity-***@lists. sourceforge.net
https://lists.sourceforge.net/ lists/listinfo/audacity-devel




------------------------------------------------------------------------------
Vaughan Johnson
2016-09-06 05:37:46 UTC
Permalink
"
Planning Ahead for 2.1.4"


I think "planning " is always "ahead". ;-)))




On Mon, Sep 5, 2016 at 10:07 AM, Peter Sampson <
Post by Peter Sampson
Scrubbing Phase-3 - further improve the user interface for Audacity's
Scrubbing
http://wiki.audacityteam.org/wiki/Proposal:_Improvements_
to_Scrubbing_-_Phase-3
Could benefit from being on this 2.1.4 planning agenda.
Peter
On Sun, Sep 4, 2016 at 5:29 PM, David Engebretson Jr <
Post by David Engebretson Jr
As a daily user of Audacity, and an aspiring midi user, and also an aspiring
Audacity developer, I vote for Roger's midi. Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.
If there is a path to getting ASIO built in, I will be happy to use my
gentle demeanor to approach Steinberg regarding licensing. Please, to those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness. I don't want to hear
the same old negative "general" commentary regarding ASIO ish's. Maybe some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.
Peace,
David
-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant. It
would be nice to do a "dot zero" release, but we don't have to. There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Steve the Fiddle
2016-09-06 09:15:37 UTC
Permalink
About "Punch-In Recording":

This feature request has been made _many_ times on the forum. Most of
these requests are from users that wish to produce their first
audiobook which they can then distribute via ACX / Amazon et al. The
request is based on the belied that a "Punch-In / Punch-Out" feature
will help them achieve their goal more quickly and more easily.

The production standards required by ACX and other commercial outlets
are fairly high. In essence they will only accept an audiobook that is
of "professional quality".

Destructive punch-in / punch-out recording is certainly quicker than
recording multiple tracks and then seamlessly editing the tracks
together, but I wonder how many successful audiobooks have been
produced using this technique, My guess would be "hardly any".

On the other hand, "non-destructive punch-in / punch-out" is a feature
found in many professional audio production tools, that combines the
benefits of speed and convenience with the flexibility of precise
editing. I would guess that "most" successful audiobooks use this
technique.

Rather than answering the question:
"Many users have requested (destructive) punch-in / punch-out. Should
we develop that feature in Audacity?"
I think the question should be:
"Many users want to provide high quality audiobooks. Should we develop
features in Audacity that will help them?"

My view is that while destructive punch-in / punch-out recording may
be popular with novice audiobook producers, such a feature will hinder
them in their goal of producing professional quality work, and is
therefore against their interests. I am very much in favour of
developing as a feature of Audacity, non-destructive "multiple takes"
(including punch-in / punch-out). I'd be very interested in
collaborating on a project to introduce "virtual tracks" into
Audacity.

Steve
Post by Vaughan Johnson
"
Planning Ahead for 2.1.4"
I think "planning " is always "ahead". ;-)))
On Mon, Sep 5, 2016 at 10:07 AM, Peter Sampson
Post by Peter Sampson
Scrubbing Phase-3 - further improve the user interface for Audacity's
Scrubbing
http://wiki.audacityteam.org/wiki/Proposal:_Improvements_to_Scrubbing_-_Phase-3
Could benefit from being on this 2.1.4 planning agenda.
Peter
On Sun, Sep 4, 2016 at 5:29 PM, David Engebretson Jr
Post by David Engebretson Jr
As a daily user of Audacity, and an aspiring midi user, and also an aspiring
Audacity developer, I vote for Roger's midi. Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.
If there is a path to getting ASIO built in, I will be happy to use my
gentle demeanor to approach Steinberg regarding licensing. Please, to those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness. I don't want to hear
the same old negative "general" commentary regarding ASIO ish's. Maybe some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.
Peace,
David
-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant. It
would be nice to do a "dot zero" release, but we don't have to. There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
James Crook
2016-09-07 16:13:38 UTC
Permalink
Try this: enable cutlines, generate some sound, then select within it and
cut. Then select around the cut line and cut again.
Then left click the cut line. It expands, and reveals again the first cut
line.
Um. Windows 10 debug builds - 100% repeatable debug assertion in
MSVCP120D.dll
"Expression: vector erase iterator outside range" on doing this last step.

Something tells me not a lot of people are using this feature and that
the existing code for it is not well thought out.
So Undo, and you see, you are now in a state where a WaveClip owns another
WaveClip (the outer cut), which in turn owns yet another.
As I said, I was wondering if the thing to do might be to extend this
mechanism for the purpose of rotating alternative clips into place.
The current cut-line interface seem a bit clunky to me, but I guess it
could work. And if clips are just nodes and tracks are just nodes, it
is not materially different to a tree of tracks approach in the code for
doing so. Is there a way to toggle to close a cut line again, other
then using 'cut' again?

The GUI problem with cut lines is that being thin it isn't that
practical to add UI flexibility to them. If there is a track for the
alternative clips to live in then any enhancements for
stacking/selecting/collapsing/expanding have a visible place to live.
My own style of working with voice I often do a retake of a phrase and
then subsequently choose where in it to switch from the old take to the
new. I'll even do that mid word in some rare cases, on a
mid-word-silence, 'fff' or 'sss', which is definitely not punch-in as
you couldn't do that accurately enough 'in real time'. I would find
working with hidden takes and treating them as atomic a bit too
restrictive for the way I do retakes, if there was no way to do that.
Cut lines UI looks like it would struggle to show the alternative takes,
particularly to show more than one take at a time.

--James.


------------------------------------------------------------------------------
Paul Licameli
2016-09-07 16:53:37 UTC
Permalink
Post by James Crook
Try this: enable cutlines, generate some sound, then select within it and
cut. Then select around the cut line and cut again.
Then left click the cut line. It expands, and reveals again the first
cut
line.
Um. Windows 10 debug builds - 100% repeatable debug assertion in
MSVCP120D.dll
"Expression: vector erase iterator outside range" on doing this last step.
Aha. Fixed that. It might be I what broke it.

PRL
Post by James Crook
Something tells me not a lot of people are using this feature and that
the existing code for it is not well thought out.
So Undo, and you see, you are now in a state where a WaveClip owns
another
WaveClip (the outer cut), which in turn owns yet another.
As I said, I was wondering if the thing to do might be to extend this
mechanism for the purpose of rotating alternative clips into place.
The current cut-line interface seem a bit clunky to me, but I guess it
could work. And if clips are just nodes and tracks are just nodes, it
is not materially different to a tree of tracks approach in the code for
doing so. Is there a way to toggle to close a cut line again, other
then using 'cut' again?
The GUI problem with cut lines is that being thin it isn't that
practical to add UI flexibility to them. If there is a track for the
alternative clips to live in then any enhancements for
stacking/selecting/collapsing/expanding have a visible place to live.
My own style of working with voice I often do a retake of a phrase and
then subsequently choose where in it to switch from the old take to the
new. I'll even do that mid word in some rare cases, on a
mid-word-silence, 'fff' or 'sss', which is definitely not punch-in as
you couldn't do that accurately enough 'in real time'. I would find
working with hidden takes and treating them as atomic a bit too
restrictive for the way I do retakes, if there was no way to do that.
Cut lines UI looks like it would struggle to show the alternative takes,
particularly to show more than one take at a time.
--James.
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Steve the Fiddle
2016-09-07 01:59:25 UTC
Permalink
A "less work" non-destructive way could be developed as a track
plug-in / module in which pre-roll recording occurs on a special 2
channel mono track, and automatically creates envelopes to mute the
parts that are assumed to be not needed.

Steve
We just gotta do better than juggling six interfaces by iteration and
deciding on none.
I have a suggestion for the UI, which is essentially to replace our list
of tracks with a tree of tracks.
People will not notice any difference if they use it 'the normal way'.
Did you attempt something like this in AU14?
a) We can expand/collapse subtrees as in a file/folder browser. So if
we have a clip with 20 takes, we can view them as 20 stacked tracks, or
as one (collapsed).
b) Muting and soloing works at folder level. We are soloing our
different takes, only playing one of them at a time. The advantage here
is that we are reusing soling rather than inventing a new mechanism.
True, not a new mechanism. But would it work conveniently enough?
I imagine that if there is a place with alternative takes, I would want an
easy way to cycle through them, like Alt+Tab. I would like to time-shift
the take to adjust the timing. I would like automatic cross-fading of the
overlap in the rendering, however long I might make that overlap. I might
also want takes that are not exactly of the same length, especially if I am
doing spoken word which isn't precisely measured like music. So I might
even want everything that exists in a single take, to the right of the
alternatives, to time-shift as I switch among takes that do not agree in
length.
This is getting far ahead of the simpler idea of just reusing the mute and
solo feature.
PRL
Another aspect is using label track to mark the pre-roll and replacement
zones. Attributes of the labels can mark whether a zone is squashy (the
takes do not have to be the same length) or rigid. Putting those aspects
on the labels makes the 'zone customisation' interface more reusable and
less special case. Potentially it also allows a user to more easily mix
and max between different takes at label boundaries rather than at clip
boundaries.
Analogous to GIMP you would be able to 'flatten' (some of) these
layers. It perhaps is the same thing as 'mix' (respecting soloing, and
preserving clip boundaries where possible).
I think the UI development should happen in a branch. It's complex and
unpredictable how long it would take. A possible problem is that when
the branch is 'ready' we might not accept it into Audacity. And then we
are back to iteration.
That's for non-destructive punch in.
------------------------
Paul knows what he is doing with destructive punch in, and there is much
less risk to him of doing that in a branch and then Audacity not
accepting it (or asking for 'just a few small tweaks' to make it ready).
--James.
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
Steve the Fiddle
2016-09-06 19:00:19 UTC
Permalink
Post by Steve the Fiddle
This feature request has been made _many_ times on the forum. Most of
these requests are from users that wish to produce their first
audiobook which they can then distribute via ACX / Amazon et al. The
request is based on the belied that a "Punch-In / Punch-Out" feature
will help them achieve their goal more quickly and more easily.
The production standards required by ACX and other commercial outlets
are fairly high. In essence they will only accept an audiobook that is
of "professional quality".
Destructive punch-in / punch-out recording is certainly quicker than
recording multiple tracks and then seamlessly editing the tracks
together, but I wonder how many successful audiobooks have been
produced using this technique, My guess would be "hardly any".
On the other hand, "non-destructive punch-in / punch-out" is a feature
found in many professional audio production tools, that combines the
benefits of speed and convenience with the flexibility of precise
editing. I would guess that "most" successful audiobooks use this
technique.
"Many users have requested (destructive) punch-in / punch-out. Should
we develop that feature in Audacity?"
"Many users want to provide high quality audiobooks. Should we develop
features in Audacity that will help them?"
My view is that while destructive punch-in / punch-out recording may
be popular with novice audiobook producers, such a feature will hinder
them in their goal of producing professional quality work, and is
therefore against their interests.
That is a reasonable line to take, except that it's not correct to
say these requests are only from audiobook produders.
That'll be why I didn't say that.
Voiceover
artists producing dozens of 20 second soundbytes for submission
every day ask for punch-in too, and so do musicians.
Have you ever had the misfortune of working with music on a recording
system where destructive punch editing was the only option? That was
the case in the bad old days of tape based machines. Thank goodness
for non-linear editing,
My question is that if we want non-destructive punch-in in Audacity,
should we not go straight there, by passing destructive punch-in?
Figure out what we want first.
Post by Steve the Fiddle
I am very much in favour of developing as a feature of Audacity,
non-destructive "multiple takes" (including punch-in / punch-out).
I'd be very interested in collaborating on a project to introduce
"virtual tracks" into Audacity.
OK. But destructive or not, what users object to in the current
scheme is the clutter of extra tracks. Even if you get your
correction spot on first time, there may be dozens or hundreds
of corrections. You can't work efficiently with dozens or hundreds
of tracks and you can't work efficiently if you have to keep closing
those extra tracks by hand.
There should be no need for dozens or hundreds of tracks. For a simple
voice recording job you can already easily do non-destructive pre-roll
recording with two tracks. The problem is that novice users don't know
how, so perhaps we should make that easier for them.

Steve
Gale
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
Paul Licameli
2016-09-06 22:39:07 UTC
Permalink
Post by James Crook
On Tue, Sep 6, 2016 at 3:00 PM, Steve the Fiddle <
Post by Steve the Fiddle
On 6 September 2016 at 10:15, Steve the Fiddle
Post by Steve the Fiddle
This feature request has been made _many_ times on the forum. Most of
these requests are from users that wish to produce their first
audiobook which they can then distribute via ACX / Amazon et al. The
request is based on the belied that a "Punch-In / Punch-Out" feature
will help them achieve their goal more quickly and more easily.
The production standards required by ACX and other commercial outlets
are fairly high. In essence they will only accept an audiobook that
is
Post by Steve the Fiddle
Post by Steve the Fiddle
of "professional quality".
Destructive punch-in / punch-out recording is certainly quicker than
recording multiple tracks and then seamlessly editing the tracks
together, but I wonder how many successful audiobooks have been
produced using this technique, My guess would be "hardly any".
On the other hand, "non-destructive punch-in / punch-out" is a
feature
Post by Steve the Fiddle
Post by Steve the Fiddle
found in many professional audio production tools, that combines the
benefits of speed and convenience with the flexibility of precise
editing. I would guess that "most" successful audiobooks use this
technique.
"Many users have requested (destructive) punch-in / punch-out. Should
we develop that feature in Audacity?"
"Many users want to provide high quality audiobooks. Should we
develop
Post by Steve the Fiddle
Post by Steve the Fiddle
features in Audacity that will help them?"
My view is that while destructive punch-in / punch-out recording may
be popular with novice audiobook producers, such a feature will
hinder
Post by Steve the Fiddle
Post by Steve the Fiddle
them in their goal of producing professional quality work, and is
therefore against their interests.
That is a reasonable line to take, except that it's not correct to
say these requests are only from audiobook produders.
That'll be why I didn't say that.
Voiceover
artists producing dozens of 20 second soundbytes for submission
every day ask for punch-in too, and so do musicians.
Have you ever had the misfortune of working with music on a recording
system where destructive punch editing was the only option? That was
the case in the bad old days of tape based machines. Thank goodness
for non-linear editing,
My question is that if we want non-destructive punch-in in Audacity,
should we not go straight there, by passing destructive punch-in?
Figure out what we want first.
Post by Steve the Fiddle
I am very much in favour of developing as a feature of Audacity,
non-destructive "multiple takes" (including punch-in / punch-out).
I'd be very interested in collaborating on a project to introduce
"virtual tracks" into Audacity.
OK. But destructive or not, what users object to in the current
scheme is the clutter of extra tracks. Even if you get your
correction spot on first time, there may be dozens or hundreds
of corrections. You can't work efficiently with dozens or hundreds
of tracks and you can't work efficiently if you have to keep closing
those extra tracks by hand.
There should be no need for dozens or hundreds of tracks. For a simple
voice recording job you can already easily do non-destructive pre-roll
recording with two tracks. The problem is that novice users don't know
how, so perhaps we should make that easier for them.
Explain the procedure that you mean here.
1) Start recording.
2) Make a blooper and stop.
3) Delete the blooper and click a couple of seconds before the end of the
track.
4) Start recording (track 2).
That's the first drop in with pre-roll. There are now 2 tracks.
5) Make another blooper and stop.
6) Delete the blooper and click a couple of seconds before the end of
track 2 IN TRACK 1.
7) Shift + R (append record in track 1).
That's the second drop-in with pre-roll. There are still only 2 tracks.
Repeat as necessary, alternating between tracks 1 and 2.
If you're happy with a "rough and ready", Export and done.
At the end there will probably be a bit of cleaning up to do (isn't
there always?), but basically you have all of the bits in more or less
the right places, in just two tracks, and the only destructive step is
that you've deleted the "known to be bad" bits (the bloopers).
Everything else is adjustable in post production (should you wish to
do post production).
Steve
Okay, but of course it's only an approximation of punch and roll at the
best.

Deleting is not quite so simple as just picking the spot you want, and
simplifying this repetitive action to the extent possible would be
important.

Are you recording with overdub so that you hear the pre-roll cue? Then
presumably you must wear headphones. But I know some narrators have a
preference not to do that if they can.

Do you want to speak in sync with the cue to get in rhythm, and continue
past the punch-in point? But then there is work to eliminate that from the
mix.

Whereas a better feature for narrators' purposes would address all of this.

You ask about punching both in and out. That's if you want to replace a
measure of music? Not the concern for narrators, but that is not to say it
shouldn't also be considered. Even musicians use Audacity too :-)

PRL
Post by James Crook
PRL
Post by Steve the Fiddle
Steve
Gale
------------------------------------------------------------
------------------
Post by Steve the Fiddle
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by Steve the Fiddle
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Paul Licameli
2016-09-07 00:18:46 UTC
Permalink
Post by James Crook
On Tue, Sep 6, 2016 at 6:16 PM, Steve the Fiddle <
Post by James Crook
On Tue, Sep 6, 2016 at 3:00 PM, Steve the Fiddle
Post by Steve the Fiddle
On 6 September 2016 at 10:15, Steve the Fiddle
Post by Steve the Fiddle
This feature request has been made _many_ times on the forum. Most of
these requests are from users that wish to produce their first
audiobook which they can then distribute via ACX / Amazon et al.
The
Post by James Crook
Post by Steve the Fiddle
Post by Steve the Fiddle
request is based on the belied that a "Punch-In / Punch-Out"
feature
Post by James Crook
Post by Steve the Fiddle
Post by Steve the Fiddle
will help them achieve their goal more quickly and more easily.
The production standards required by ACX and other commercial outlets
are fairly high. In essence they will only accept an audiobook
that
Post by James Crook
Post by Steve the Fiddle
Post by Steve the Fiddle
is
of "professional quality".
Destructive punch-in / punch-out recording is certainly quicker
than
Post by James Crook
Post by Steve the Fiddle
Post by Steve the Fiddle
recording multiple tracks and then seamlessly editing the tracks
together, but I wonder how many successful audiobooks have been
produced using this technique, My guess would be "hardly any".
On the other hand, "non-destructive punch-in / punch-out" is a feature
found in many professional audio production tools, that combines
the
Post by James Crook
Post by Steve the Fiddle
Post by Steve the Fiddle
benefits of speed and convenience with the flexibility of precise
editing. I would guess that "most" successful audiobooks use this
technique.
"Many users have requested (destructive) punch-in / punch-out. Should
we develop that feature in Audacity?"
"Many users want to provide high quality audiobooks. Should we develop
features in Audacity that will help them?"
My view is that while destructive punch-in / punch-out recording
may
Post by James Crook
Post by Steve the Fiddle
Post by Steve the Fiddle
be popular with novice audiobook producers, such a feature will hinder
them in their goal of producing professional quality work, and is
therefore against their interests.
That is a reasonable line to take, except that it's not correct to
say these requests are only from audiobook produders.
That'll be why I didn't say that.
Voiceover
artists producing dozens of 20 second soundbytes for submission
every day ask for punch-in too, and so do musicians.
Have you ever had the misfortune of working with music on a recording
system where destructive punch editing was the only option? That was
the case in the bad old days of tape based machines. Thank goodness
for non-linear editing,
My question is that if we want non-destructive punch-in in
Audacity,
Post by James Crook
Post by Steve the Fiddle
should we not go straight there, by passing destructive punch-in?
Figure out what we want first.
Post by Steve the Fiddle
I am very much in favour of developing as a feature of Audacity,
non-destructive "multiple takes" (including punch-in /
punch-out).
Post by James Crook
Post by Steve the Fiddle
Post by Steve the Fiddle
I'd be very interested in collaborating on a project to introduce
"virtual tracks" into Audacity.
OK. But destructive or not, what users object to in the current
scheme is the clutter of extra tracks. Even if you get your
correction spot on first time, there may be dozens or hundreds
of corrections. You can't work efficiently with dozens or hundreds
of tracks and you can't work efficiently if you have to keep
closing
Post by James Crook
Post by Steve the Fiddle
those extra tracks by hand.
There should be no need for dozens or hundreds of tracks. For a
simple
Post by James Crook
Post by Steve the Fiddle
voice recording job you can already easily do non-destructive
pre-roll
Post by James Crook
Post by Steve the Fiddle
recording with two tracks. The problem is that novice users don't
know
Post by James Crook
Post by Steve the Fiddle
how, so perhaps we should make that easier for them.
Explain the procedure that you mean here.
1) Start recording.
2) Make a blooper and stop.
3) Delete the blooper and click a couple of seconds before the end of
the
Post by James Crook
track.
4) Start recording (track 2).
That's the first drop in with pre-roll. There are now 2 tracks.
5) Make another blooper and stop.
6) Delete the blooper and click a couple of seconds before the end of
track 2 IN TRACK 1.
7) Shift + R (append record in track 1).
That's the second drop-in with pre-roll. There are still only 2 tracks.
Repeat as necessary, alternating between tracks 1 and 2.
If you're happy with a "rough and ready", Export and done.
At the end there will probably be a bit of cleaning up to do (isn't
there always?), but basically you have all of the bits in more or less
the right places, in just two tracks, and the only destructive step is
that you've deleted the "known to be bad" bits (the bloopers).
Everything else is adjustable in post production (should you wish to
do post production).
Steve
Okay, but of course it's only an approximation of punch and roll at the
best.
Sure it could be made easier and more convenient. My point was that it
does not require "dozens or hundreds of tracks".
Point taken.
Post by James Crook
Deleting is not quite so simple as just picking the spot you want,
I don't see that it's "less simple". I agree that deleting is not a
one click action, but it's not difficult or complicated.
and
simplifying this repetitive action to the extent possible would be
important.
Yes, that's what I mean by making it easier and more convenient.
Are you recording with overdub so that you hear the pre-roll cue? Then
presumably you must wear headphones. But I know some narrators have a
preference not to do that if they can.
Because it's non-destructive, it would not be necessary to record with
overdub. They have the option of adjusting the position in post.
If it were destructive, how would a narrator know precisely when to
start talking without an audible cue? Or do you propose some clever
mechanism that does not require them to start talking at just the
right time?
What I said below.
Post by James Crook
Do you want to speak in sync with the cue to get in rhythm, and continue
past the punch-in point? But then there is work to eliminate that from
the
mix.
Yes there would then post production work to do.
Work that would be better automated, as you record. You speak with the
overdub to get in rhythm, and the software takes care of muting that part
and keeping only what follows the break.

Though even in a simple nondestructive version you would likely want some
crossfading applied too at the join.
Post by James Crook
What happens if the new take is slightly ahead (early)? If it's
non-destructive you can slide it to the right, If it's destructive you
may have clipped the first word.
True. A good reason why the nondestructive version would be a very
desirable enhancement. But, much added work.

Would the feature be too poor to be useful without nondestructiveness?
Experiment must tell.
Post by James Crook
Whereas a better feature for narrators' purposes would address all of
this.
You ask about punching both in and out. That's if you want to replace a
measure of music? Not the concern for narrators,
It could be if they find that they've clipped a word 20 minutes into
a 90 minute voice over piece.
Correcting misreads is part of the procedure, but I don't think punch in
and out is so highly valued for that task. Recording, cutting, and pasting
as needed may be adequate. Whereas punch and roll recording is useful for
accomplishing the rough editing as you go, in the booth, without losing the
flow, and leaving less to do in post.

PRL
Post by James Crook
but that is not to say it
shouldn't also be considered. Even musicians use Audacity too :-)
Audacity is rarely used for producing electronic music, primarily
because of it's lack of MIDI support, but that's about the only type
of music where destructive punch in/out works satisfactorily. For
recording people singing or playing instruments, destructive punch
in/out is a horrible way of editing. It was horrible back in the days
of tape and it's still horrible. There are much better ways to do the
job.
Steve
PRL
Post by James Crook
PRL
Post by Steve the Fiddle
Steve
Gale
------------------------------------------------------------
------------------
Post by James Crook
Post by Steve the Fiddle
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by James Crook
Post by Steve the Fiddle
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by James Crook
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by James Crook
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
Robert Hänggi
2016-09-07 13:41:02 UTC
Permalink
Sorry, the handles I've mentioned are for the take boundaries, not the
playback, i.e. the line should be one higher in the list.

I should mention that edits are possible on the take level but they
would always act on the whole clip/take length. For instance, delete
would remove a take or a clip. Same for effects.
The time boxes could perhaps be relative, thus a negative value would
indicate an overlap. Also, there could be radio buttons next to the
times, indicating the transitions:
- cross-fade
- foreground (meaning previous or next clip is muted during overlap)
- background (meaning previous or next clip is not trimmed)

Robert
So it seems you anticipated many of the tasks I imagine one would want.
But I am still not yet sold on a tree of tracks as necessarily best
alternative, or on any involvement of labels with this. I had been
thinking instead of building on the existing organization of a track as a
list of clips.
If the purpose is only to enable you an easy rotation among alternative
takes at some place in the track, then I don't see the use of the
generality of muting and soloing any subset of the takes.
PRL
+1
I like James idea of a tree structure. I could imagine the following
We have a track with different takes
- The track can be expanded by a shortcut (e.g. alt-enter)
- All clips will be listed
- focus is on the clip where the selection start is
- if the clip has more than one take, this will be indicated by either
appending "collapsed" or giving the number of takes itself.
- the selection is temporarily set to the clip boundaries
- you can play this selection
- you can switch to another clip with arrow up/down keys and mouse
(temp cursor follows)
- right arrow expands a clip with multiple takes
- return or mouse click determines which take will be active for
playback mix-down etc.
- tabulator jumps to numeric fields with start time, end time,
"Rigidness" etc, settable for each take individually.
- the transitions can be played with the existing shortcuts
Ctrl+Shift+F5/F7 or e.g Quick-Play
- also available as handles in the waveform.
- left arrow collapses the takes, pressed twice, the track is also
collapsed.
- original selection/cursor position is restored.
I would detailed edits (such as deleting a part within a take) only
allow on the track level and the activated take, in this way the arrow
keys are free to navigate in the expanded tree without the usual
cursor positioning task.
It is not absolutely necessarily to expand to the take level, the
active one could also be chosen/rotated by pressing right arrow or a
mouse click on the take number.
Robert
------------------------------------------------------------------------------
Gale Andrews
2016-09-06 17:56:00 UTC
Permalink
On 6 September 2016 at 16:13, Paul Licameli <***@gmail.com> wrote:
[...]
I do not understand this objection to providing a destructive punch and roll
recording. Such a developent is not exclusive of a nondestructive version
later. An Audacity with this feature is more capable than one without it.
I don't object, if destructive punch-in is indeed not exclusive of
a non-destructive version. People don't always have time to
do multiple takes and crossfades, so it must be easy, without
clutter of extra tracks do punch-in on one track.

The "virtual tracks" sound wonderful but IMO it should be made
easy to keep the layers or virtual tracks in one real track and
crossfade in one track.
"I've recorded 90 minutes of my audiobook and now I've found there's
some bits of words missing. Help me!"
Answer: Don't do destructive punch-in recording.
Listen to each correction first then Edit > Undo if it is not as you
want it.


Gale



Steve wrote:

"Virtual tracks" is a feature common many hardware multi-track recorders.

A hardware multi-track recorder has a fixed number of physical input
and output channels, and usually a fixed number of available "tracks".
The number of tracks is usually a multiple of the number of physical
channels. For example, a recorder may have 2 physical inputs, 4
physical outputs and 8 available "tracks" to record into and play
from. The input and output channels can be mapped by the user to
specific tracks. For example it may be possible to record from 2
physical inputs to tracks 1 & 2, 3 & 4, 5 & 6 or 7 & 8. In this
example, the user is limited to a maximum of 2 input channels at a
time, but can record up to 8 separate tracks.

"Virtual tracks" is the ability to associate more than one set of
audio data to a "track". In the above example, there could be data for
one active track and data for a further 7 virtual tracks associated
with each of the 8 tracks. Thus up to 8 separate "takes" can exist
simultaneously on each track.

Hardware implementations don't generally take the concept much further
than that, but in software the idea may be expanded much further. In
software there can potentially be an unlimited number of virtual
tracks, and multiple audio clips in each of them. In software it is
possible to "explode" a track so that each virtual track becomes a new
(real) track, or to stack multiple real tracks into a single

Perhaps this is what you mean by "layers"?

Another way of looking at it would be as an array, in which each row
(of clips) is a virtual track and we can switch between rows on the
fly. The "current row" being the visible track. Thus if an audio clip
is dragged so that it "collides" with another audio clip, one of the
audio clips is mapped to a new row and we can crossfade between rows.
Of course the actual audio data does not need to be moved, we're just
mapping time regions within the clip's audio data to different virtual
tracks. You can think of it as a stack of tracks, where we can switch
from one to another within the stack during playback. Alternatively
you could think of it as "groups" of tracks, in which the visible
waveform is taken from one or more tracks in the group.

I think the idea becomes very exciting when you think that it could
give us a really nice GUI means of creating crossfades, and (almost)
non-destructive effects in that the processed data can be written to a
new virtual track rather than overwriting the original data, and
non-destructive punch-in / punch-out, and "undoable" mixdowns, and
probably more.

The initial problem would be in structuring track data in such a was
as to make it possible, and non-destructive punch recording could be a
good use case to focus on.

------------------------------------------------------------------------------
Steve the Fiddle
2016-09-06 19:04:10 UTC
Permalink
Post by Gale Andrews
[...]
I do not understand this objection to providing a destructive punch and roll
recording. Such a developent is not exclusive of a nondestructive version
later. An Audacity with this feature is more capable than one without it.
I don't object, if destructive punch-in is indeed not exclusive of
a non-destructive version. People don't always have time to
do multiple takes and crossfades, so it must be easy, without
clutter of extra tracks do punch-in on one track.
The "virtual tracks" sound wonderful but IMO it should be made
easy to keep the layers or virtual tracks in one real track and
crossfade in one track.
"I've recorded 90 minutes of my audiobook and now I've found there's
some bits of words missing. Help me!"
Answer: Don't do destructive punch-in recording.
Listen to each correction first then Edit > Undo if it is not as you
want it.
and that is exactly what they do NOT want to do. They want punch
editing so that they can continue without losing their flow. I know
that because that's what they repeatedly say.

Steve
Post by Gale Andrews
Gale
"Virtual tracks" is a feature common many hardware multi-track recorders.
A hardware multi-track recorder has a fixed number of physical input
and output channels, and usually a fixed number of available "tracks".
The number of tracks is usually a multiple of the number of physical
channels. For example, a recorder may have 2 physical inputs, 4
physical outputs and 8 available "tracks" to record into and play
from. The input and output channels can be mapped by the user to
specific tracks. For example it may be possible to record from 2
physical inputs to tracks 1 & 2, 3 & 4, 5 & 6 or 7 & 8. In this
example, the user is limited to a maximum of 2 input channels at a
time, but can record up to 8 separate tracks.
"Virtual tracks" is the ability to associate more than one set of
audio data to a "track". In the above example, there could be data for
one active track and data for a further 7 virtual tracks associated
with each of the 8 tracks. Thus up to 8 separate "takes" can exist
simultaneously on each track.
Hardware implementations don't generally take the concept much further
than that, but in software the idea may be expanded much further. In
software there can potentially be an unlimited number of virtual
tracks, and multiple audio clips in each of them. In software it is
possible to "explode" a track so that each virtual track becomes a new
(real) track, or to stack multiple real tracks into a single
Perhaps this is what you mean by "layers"?
Another way of looking at it would be as an array, in which each row
(of clips) is a virtual track and we can switch between rows on the
fly. The "current row" being the visible track. Thus if an audio clip
is dragged so that it "collides" with another audio clip, one of the
audio clips is mapped to a new row and we can crossfade between rows.
Of course the actual audio data does not need to be moved, we're just
mapping time regions within the clip's audio data to different virtual
tracks. You can think of it as a stack of tracks, where we can switch
from one to another within the stack during playback. Alternatively
you could think of it as "groups" of tracks, in which the visible
waveform is taken from one or more tracks in the group.
I think the idea becomes very exciting when you think that it could
give us a really nice GUI means of creating crossfades, and (almost)
non-destructive effects in that the processed data can be written to a
new virtual track rather than overwriting the original data, and
non-destructive punch-in / punch-out, and "undoable" mixdowns, and
probably more.
The initial problem would be in structuring track data in such a was
as to make it possible, and non-destructive punch recording could be a
good use case to focus on.
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
Gale Andrews
2016-09-08 22:53:39 UTC
Permalink
Post by Steve the Fiddle
Post by Gale Andrews
[...]
I do not understand this objection to providing a destructive punch and roll
recording. Such a developent is not exclusive of a nondestructive version
later. An Audacity with this feature is more capable than one without it.
I don't object, if destructive punch-in is indeed not exclusive of
a non-destructive version. People don't always have time to
do multiple takes and crossfades, so it must be easy, without
clutter of extra tracks do punch-in on one track.
The "virtual tracks" sound wonderful but IMO it should be made
easy to keep the layers or virtual tracks in one real track and
crossfade in one track.
"I've recorded 90 minutes of my audiobook and now I've found there's
some bits of words missing. Help me!"
Answer: Don't do destructive punch-in recording.
Listen to each correction first then Edit > Undo if it is not as you
want it.
and that is exactly what they do NOT want to do. They want punch
editing so that they can continue without losing their flow. I know
that because that's what they repeatedly say.
Of course, but right now Undo that is what they have to do, IMO,
if they want to be quick enough not to lose the flow.


Gale
Post by Steve the Fiddle
Post by Gale Andrews
"Virtual tracks" is a feature common many hardware multi-track recorders.
A hardware multi-track recorder has a fixed number of physical input
and output channels, and usually a fixed number of available "tracks".
The number of tracks is usually a multiple of the number of physical
channels. For example, a recorder may have 2 physical inputs, 4
physical outputs and 8 available "tracks" to record into and play
from. The input and output channels can be mapped by the user to
specific tracks. For example it may be possible to record from 2
physical inputs to tracks 1 & 2, 3 & 4, 5 & 6 or 7 & 8. In this
example, the user is limited to a maximum of 2 input channels at a
time, but can record up to 8 separate tracks.
"Virtual tracks" is the ability to associate more than one set of
audio data to a "track". In the above example, there could be data for
one active track and data for a further 7 virtual tracks associated
with each of the 8 tracks. Thus up to 8 separate "takes" can exist
simultaneously on each track.
Hardware implementations don't generally take the concept much further
than that, but in software the idea may be expanded much further. In
software there can potentially be an unlimited number of virtual
tracks, and multiple audio clips in each of them. In software it is
possible to "explode" a track so that each virtual track becomes a new
(real) track, or to stack multiple real tracks into a single
Perhaps this is what you mean by "layers"?
Another way of looking at it would be as an array, in which each row
(of clips) is a virtual track and we can switch between rows on the
fly. The "current row" being the visible track. Thus if an audio clip
is dragged so that it "collides" with another audio clip, one of the
audio clips is mapped to a new row and we can crossfade between rows.
Of course the actual audio data does not need to be moved, we're just
mapping time regions within the clip's audio data to different virtual
tracks. You can think of it as a stack of tracks, where we can switch
from one to another within the stack during playback. Alternatively
you could think of it as "groups" of tracks, in which the visible
waveform is taken from one or more tracks in the group.
I think the idea becomes very exciting when you think that it could
give us a really nice GUI means of creating crossfades, and (almost)
non-destructive effects in that the processed data can be written to a
new virtual track rather than overwriting the original data, and
non-destructive punch-in / punch-out, and "undoable" mixdowns, and
probably more.
The initial problem would be in structuring track data in such a was
as to make it possible, and non-destructive punch recording could be a
good use case to focus on.
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
Robert Hänggi
2016-09-06 14:15:49 UTC
Permalink
Post by Steve the Fiddle
This feature request has been made _many_ times on the forum. Most of
these requests are from users that wish to produce their first
audiobook which they can then distribute via ACX / Amazon et al. The
request is based on the belied that a "Punch-In / Punch-Out" feature
will help them achieve their goal more quickly and more easily.
The production standards required by ACX and other commercial outlets
are fairly high. In essence they will only accept an audiobook that is
of "professional quality".
Destructive punch-in / punch-out recording is certainly quicker than
recording multiple tracks and then seamlessly editing the tracks
together, but I wonder how many successful audiobooks have been
produced using this technique, My guess would be "hardly any".
On the other hand, "non-destructive punch-in / punch-out" is a feature
found in many professional audio production tools, that combines the
benefits of speed and convenience with the flexibility of precise
editing. I would guess that "most" successful audiobooks use this
technique.
"Many users have requested (destructive) punch-in / punch-out. Should
we develop that feature in Audacity?"
"Many users want to provide high quality audiobooks. Should we develop
features in Audacity that will help them?"
My view is that while destructive punch-in / punch-out recording may
be popular with novice audiobook producers, such a feature will hinder
them in their goal of producing professional quality work, and is
therefore against their interests. I am very much in favour of
developing as a feature of Audacity, non-destructive "multiple takes"
(including punch-in / punch-out). I'd be very interested in
collaborating on a project to introduce "virtual tracks" into
Audacity.
Steve
+1
Exactly what I'm thinking.
I used to work with Cakewalk Guitar tracks and it gave me the ability
to do just that.
My current work flow is somewhat similar to a non-destructive punch-in
(although it lacks the actual "punch" feature).
For instance, I select a measure and record a guitar solo or a vocal
line over it. This ends up in about 4 or five takes that I can choose
from eventually.
What I'm missing:
- looped recording, gives you the possibility of first listening to
the passage and giving you time for preparation. Guitar track did
actually some sort of sound-activated recording which allowed to
correct a single tone or word while keeping the rest.
- easy insertion of one of the clips into the actual instrument track
(pushing it to the top, if you will). Currently, I have to select the
recorded track, cut it (Ctrl-x), deselect the track while switching to
the target, paste the content and delete the superfluous empty track.
IMHO, Audacity should be so intelligent to realize that the cut region
is actually the only content of the take-track and that the track can
automatically be removed after pasting.
- Choosing one take for listening and a later mix-down
(non-destructive variant as opposed to above) demands muting the other
tracks manually and selecting it (for rendering).
- The mix-down of a named and a unnamed track results in "Mixed
Track", rather than the already existing name.
- I don't understand why the recorded track isn't automatically
selected. Isn't it most likely that this is the one that will be
edited next?

In general, I prefer when I can keep a selection of my best takes
instead of overwriting the audio directly. I'm therefore certainly not
in favour of switching from "record into new track" to "append
record".

I see the benefit of a destructive punch-in for narration tracks
though-they do not have to take tempo into account.
But I would approach it in a "hold and record" fashion:
- you select the word/phrase/sentence that you want to overwrite (optionally)
- you press and hold R or click and hold the record button
- you get a chime sound to indicate that the "insert" recording mode
has been activated.
- You speak your text.
- you release the key or button.
- another out-take sound

I think this is fairly common behaviour for many voice recorders.

Robert
Post by Steve the Fiddle
Post by Vaughan Johnson
"
Planning Ahead for 2.1.4"
I think "planning " is always "ahead". ;-)))
On Mon, Sep 5, 2016 at 10:07 AM, Peter Sampson
Post by Peter Sampson
Scrubbing Phase-3 - further improve the user interface for Audacity's
Scrubbing
http://wiki.audacityteam.org/wiki/Proposal:_Improvements_to_Scrubbing_-_Phase-3
Could benefit from being on this 2.1.4 planning agenda.
Peter
On Sun, Sep 4, 2016 at 5:29 PM, David Engebretson Jr
Post by David Engebretson Jr
As a daily user of Audacity, and an aspiring midi user, and also an aspiring
Audacity developer, I vote for Roger's midi. Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.
If there is a path to getting ASIO built in, I will be happy to use my
gentle demeanor to approach Steinberg regarding licensing. Please, to those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness. I don't want to hear
the same old negative "general" commentary regarding ASIO ish's. Maybe some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.
Peace,
David
-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant.
It
would be nice to do a "dot zero" release, but we don't have to. There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
Steve the Fiddle
2016-09-06 20:34:07 UTC
Permalink
On Tue, Sep 6, 2016 at 1:43 PM, Peter Sampson
My question is that if we want non-destructive punch-in in Audacity,
should we not go straight there, by passing destructive punch-in?
+1
I see no point in providing a sub-optimal destructive punch-in if we are
going to replace it with a pukka non-destructive version later.
Peter
Difficulty, my pukka sahib.
Getting the destructive punch-in right first entails much less work -- which
I have already invested in -- requiring solution of the necessary, common
problems.
Making the non-destructive punch-in entails a lot of hard thought about how
to restructure our saved sound data, as Steve said, and may be fraught with
much peril. And virtual tracks, or layers, call them what you will, would
require additional user interface decisions, which we will wrangle six ways
from Sunday and still disagree on. It's a much more major undertaking.
So the question is whether the easier feature is worth it by itself, if the
more elaborate feature can't be achieved in the release.
Steve seems to be suggesting that the destructive punch and roll might be
useless and worse than nothing. I'm not persuaded of that.
Could you clarify whether you mean "punch in/out" or "recording with pre-roll".
It's the "punch out" part of "punch in/out" recording that I
particularly dislike because it is so much more difficult to get a
good result than when using separate tracks.

Steve
PRL
Post by Steve the Fiddle
This feature request has been made _many_ times on the forum. Most of
these requests are from users that wish to produce their first
audiobook which they can then distribute via ACX / Amazon et al. The
request is based on the belied that a "Punch-In / Punch-Out" feature
will help them achieve their goal more quickly and more easily.
The production standards required by ACX and other commercial outlets
are fairly high. In essence they will only accept an audiobook that is
of "professional quality".
Destructive punch-in / punch-out recording is certainly quicker than
recording multiple tracks and then seamlessly editing the tracks
together, but I wonder how many successful audiobooks have been
produced using this technique, My guess would be "hardly any".
On the other hand, "non-destructive punch-in / punch-out" is a feature
found in many professional audio production tools, that combines the
benefits of speed and convenience with the flexibility of precise
editing. I would guess that "most" successful audiobooks use this
technique.
"Many users have requested (destructive) punch-in / punch-out. Should
we develop that feature in Audacity?"
"Many users want to provide high quality audiobooks. Should we develop
features in Audacity that will help them?"
My view is that while destructive punch-in / punch-out recording may
be popular with novice audiobook producers, such a feature will hinder
them in their goal of producing professional quality work, and is
therefore against their interests.
That is a reasonable line to take, except that it's not correct to
say these requests are only from audiobook produders. Voiceover
artists producing dozens of 20 second soundbytes for submission
every day ask for punch-in too, and so do musicians.
My question is that if we want non-destructive punch-in in Audacity,
should we not go straight there, by passing destructive punch-in?
Figure out what we want first.
Post by Steve the Fiddle
I am very much in favour of developing as a feature of Audacity,
non-destructive "multiple takes" (including punch-in / punch-out).
I'd be very interested in collaborating on a project to introduce
"virtual tracks" into Audacity.
OK. But destructive or not, what users object to in the current
scheme is the clutter of extra tracks. Even if you get your
correction spot on first time, there may be dozens or hundreds
of corrections. You can't work efficiently with dozens or hundreds
of tracks and you can't work efficiently if you have to keep closing
those extra tracks by hand.
Gale
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
Steve the Fiddle
2016-09-06 16:47:33 UTC
Permalink
Post by Steve the Fiddle
This feature request has been made _many_ times on the forum. Most of
these requests are from users that wish to produce their first
audiobook which they can then distribute via ACX / Amazon et al. The
request is based on the belied that a "Punch-In / Punch-Out" feature
will help them achieve their goal more quickly and more easily.
The production standards required by ACX and other commercial outlets
are fairly high. In essence they will only accept an audiobook that is
of "professional quality".
Destructive punch-in / punch-out recording is certainly quicker than
recording multiple tracks and then seamlessly editing the tracks
together, but I wonder how many successful audiobooks have been
produced using this technique, My guess would be "hardly any".
On the other hand, "non-destructive punch-in / punch-out" is a feature
found in many professional audio production tools, that combines the
benefits of speed and convenience with the flexibility of precise
editing. I would guess that "most" successful audiobooks use this
technique.
"Many users have requested (destructive) punch-in / punch-out. Should
we develop that feature in Audacity?"
"Many users want to provide high quality audiobooks. Should we develop
features in Audacity that will help them?"
My view is that while destructive punch-in / punch-out recording may
be popular with novice audiobook producers, such a feature will hinder
them in their goal of producing professional quality work, and is
therefore against their interests. I am very much in favour of
developing as a feature of Audacity, non-destructive "multiple takes"
(including punch-in / punch-out). I'd be very interested in
collaborating on a project to introduce "virtual tracks" into
Audacity.
Please explain "virtual tracks". I have mentioned "layering."
"Virtual tracks" is a feature common many hardware multi-track recorders.

A hardware multi-track recorder has a fixed number of physical input
and output channels, and usually a fixed number of available "tracks".
The number of tracks is usually a multiple of the number of physical
channels. For example, a recorder may have 2 physical inputs, 4
physical outputs and 8 available "tracks" to record into and play
from. The input and output channels can be mapped by the user to
specific tracks. For example it may be possible to record from 2
physical inputs to tracks 1 & 2, 3 & 4, 5 & 6 or 7 & 8. In this
example, the user is limited to a maximum of 2 input channels at a
time, but can record up to 8 separate tracks.

"Virtual tracks" is the ability to associate more than one set of
audio data to a "track". In the above example, there could be data for
one active track and data for a further 7 virtual tracks associated
with each of the 8 tracks. Thus up to 8 separate "takes" can exist
simultaneously on each track.

Hardware implementations don't generally take the concept much further
than that, but in software the idea may be expanded much further. In
software there can potentially be an unlimited number of virtual
tracks, and multiple audio clips in each of them. In software it is
possible to "explode" a track so that each virtual track becomes a new
(real) track, or to stack multiple real tracks into a single

Perhaps this is what you mean by "layers"?

Another way of looking at it would be as an array, in which each row
(of clips) is a virtual track and we can switch between rows on the
fly. The "current row" being the visible track. Thus if an audio clip
is dragged so that it "collides" with another audio clip, one of the
audio clips is mapped to a new row and we can crossfade between rows.
Of course the actual audio data does not need to be moved, we're just
mapping time regions within the clip's audio data to different virtual
tracks. You can think of it as a stack of tracks, where we can switch
from one to another within the stack during playback. Alternatively
you could think of it as "groups" of tracks, in which the visible
waveform is taken from one or more tracks in the group.

I think the idea becomes very exciting when you think that it could
give us a really nice GUI means of creating crossfades, and (almost)
non-destructive effects in that the processed data can be written to a
new virtual track rather than overwriting the original data, and
non-destructive punch-in / punch-out, and "undoable" mixdowns, and
probably more.

The initial problem would be in structuring track data in such a was
as to make it possible, and non-destructive punch recording could be a
good use case to focus on.
I do not understand this objection to providing a destructive punch and roll
recording.
Which part did you not understand?
Such a developent is not exclusive of a nondestructive version
later. An Audacity with this feature is more capable than one without it.
As it is now, if you want to record narration in Audacity, your alternative
is the tedious work of straight recording and then editing out the wrong
http://www.stevenjaycohen.com/2014/01/19/punch-and-roll-editing-in-audacity-on-mac-osx/
... neither of which really let you easily do what even the destructive
punch and roll would do: splice your corrected read in while you make your
own precise timing judgments in your performance.
Making non-destructiveness work, so one could rotate among alternative
takes, would introduce much complication -- I see in the bug database how
"multiclip" was a category, how that development was considered a big
destabilizer. Well, I think layer is only more so.
But getting the basic punch and roll right -- having playback that switches
instantaneously to recording at a selected time (more precisely than
Steven's scripts ever could, some delay will happen with them) -- this I
have accomplished. It's a smaller project.
I don't see a problem with "punch and roll", but why make it destructive?

I just don't look forward to the:
"I've recorded 90 minutes of my audiobook and now I've found there's
some bits of words missing. Help me!"
Answer: Don't do destructive punch-in recording.

Steve
PRL
Post by Steve the Fiddle
Steve
Post by Vaughan Johnson
"
Planning Ahead for 2.1.4"
I think "planning " is always "ahead". ;-)))
On Mon, Sep 5, 2016 at 10:07 AM, Peter Sampson
Post by Peter Sampson
Scrubbing Phase-3 - further improve the user interface for Audacity's
Scrubbing
http://wiki.audacityteam.org/wiki/Proposal:_Improvements_to_Scrubbing_-_Phase-3
Could benefit from being on this 2.1.4 planning agenda.
Peter
On Sun, Sep 4, 2016 at 5:29 PM, David Engebretson Jr
Post by David Engebretson Jr
As a daily user of Audacity, and an aspiring midi user, and also an aspiring
Audacity developer, I vote for Roger's midi. Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.
If there is a path to getting ASIO built in, I will be happy to use my
gentle demeanor to approach Steinberg regarding licensing. Please, to those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness. I don't want to hear
the same old negative "general" commentary regarding ASIO ish's.
Maybe
some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.
Peace,
David
-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant.
It
would be nice to do a "dot zero" release, but we don't have to. There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
Paul Licameli
2016-09-06 18:11:38 UTC
Permalink
Post by Gale Andrews
On Tue, Sep 6, 2016 at 5:15 AM, Steve the Fiddle <
Post by Steve the Fiddle
This feature request has been made _many_ times on the forum. Most of
these requests are from users that wish to produce their first
audiobook which they can then distribute via ACX / Amazon et al. The
request is based on the belied that a "Punch-In / Punch-Out" feature
will help them achieve their goal more quickly and more easily.
The production standards required by ACX and other commercial outlets
are fairly high. In essence they will only accept an audiobook that is
of "professional quality".
Destructive punch-in / punch-out recording is certainly quicker than
recording multiple tracks and then seamlessly editing the tracks
together, but I wonder how many successful audiobooks have been
produced using this technique, My guess would be "hardly any".
On the other hand, "non-destructive punch-in / punch-out" is a feature
found in many professional audio production tools, that combines the
benefits of speed and convenience with the flexibility of precise
editing. I would guess that "most" successful audiobooks use this
technique.
"Many users have requested (destructive) punch-in / punch-out. Should
we develop that feature in Audacity?"
"Many users want to provide high quality audiobooks. Should we develop
features in Audacity that will help them?"
My view is that while destructive punch-in / punch-out recording may
be popular with novice audiobook producers, such a feature will hinder
them in their goal of producing professional quality work, and is
therefore against their interests. I am very much in favour of
developing as a feature of Audacity, non-destructive "multiple takes"
(including punch-in / punch-out). I'd be very interested in
collaborating on a project to introduce "virtual tracks" into
Audacity.
Please explain "virtual tracks". I have mentioned "layering."
"Virtual tracks" is a feature common many hardware multi-track recorders.
A hardware multi-track recorder has a fixed number of physical input
and output channels, and usually a fixed number of available "tracks".
The number of tracks is usually a multiple of the number of physical
channels. For example, a recorder may have 2 physical inputs, 4
physical outputs and 8 available "tracks" to record into and play
from. The input and output channels can be mapped by the user to
specific tracks. For example it may be possible to record from 2
physical inputs to tracks 1 & 2, 3 & 4, 5 & 6 or 7 & 8. In this
example, the user is limited to a maximum of 2 input channels at a
time, but can record up to 8 separate tracks.
"Virtual tracks" is the ability to associate more than one set of
audio data to a "track". In the above example, there could be data for
one active track and data for a further 7 virtual tracks associated
with each of the 8 tracks. Thus up to 8 separate "takes" can exist
simultaneously on each track.
Hardware implementations don't generally take the concept much further
than that, but in software the idea may be expanded much further. In
software there can potentially be an unlimited number of virtual
tracks, and multiple audio clips in each of them. In software it is
possible to "explode" a track so that each virtual track becomes a new
(real) track, or to stack multiple real tracks into a single
Perhaps this is what you mean by "layers"?
Yes, you describe something like what I imagine. I use the term "layers"
by analogy with what's in GIMP or some CAD systems, or even as in that
sound editor, Sony SpectrLayers.
Post by Gale Andrews
Another way of looking at it would be as an array, in which each row
(of clips) is a virtual track and we can switch between rows on the
fly. The "current row" being the visible track. Thus if an audio clip
is dragged so that it "collides" with another audio clip, one of the
audio clips is mapped to a new row and we can crossfade between rows.
Of course the actual audio data does not need to be moved, we're just
mapping time regions within the clip's audio data to different virtual
tracks. You can think of it as a stack of tracks, where we can switch
from one to another within the stack during playback. Alternatively
you could think of it as "groups" of tracks, in which the visible
waveform is taken from one or more tracks in the group.
I think the idea becomes very exciting when you think that it could
give us a really nice GUI means of creating crossfades, and (almost)
non-destructive effects in that the processed data can be written to a
new virtual track rather than overwriting the original data, and
non-destructive punch-in / punch-out, and "undoable" mixdowns, and
probably more.
The initial problem would be in structuring track data in such a was
as to make it possible, and non-destructive punch recording could be a
good use case to focus on.
Yes, devil in the details there.

Recall what we have now: A WaveTrack holds a sequence of WaveClips, which
are supposed to be non-overlapping in time, but there have been bugs, some
of them that I fixed, which allowed violation of that constraint.

Also a WaveClip can have one or more cutlines associated with it, which are
other WaveClips, and so recursively in a tree structure. This feature
might be little used, but it's there.

A cutline is made with Ctrl+X, if you have checked the appopriate box in
preferences. Then you can left-click to re-expand the cutline, or right
click to delete it completely, and in either case, there is no longer a
cutline.

I thought a beginning toward layers might be to extend the cutline
capability so that you can alternate between the expanded and contracted
state without losing the cutline, and then extend that alternation to a
rotation among alternative clips.

Or maybe not.

We would also, as you said, want to allow a clip to overlap a clip
partially, and have means of specifying what cross-fading happens there. I
suppose that is somewhat analogous to transparency in image editing.

You see how intricate this becomes once you begin entertaining the idea.

PRL
Post by Gale Andrews
I do not understand this objection to providing a destructive punch and
roll
recording.
Which part did you not understand?
Such a developent is not exclusive of a nondestructive version
later. An Audacity with this feature is more capable than one without
it.
As it is now, if you want to record narration in Audacity, your
alternative
is the tedious work of straight recording and then editing out the wrong
http://www.stevenjaycohen.com/2014/01/19/punch-and-roll-
editing-in-audacity-on-mac-osx/
... neither of which really let you easily do what even the destructive
punch and roll would do: splice your corrected read in while you make
your
own precise timing judgments in your performance.
Making non-destructiveness work, so one could rotate among alternative
takes, would introduce much complication -- I see in the bug database how
"multiclip" was a category, how that development was considered a big
destabilizer. Well, I think layer is only more so.
But getting the basic punch and roll right -- having playback that
switches
instantaneously to recording at a selected time (more precisely than
Steven's scripts ever could, some delay will happen with them) -- this I
have accomplished. It's a smaller project.
I don't see a problem with "punch and roll", but why make it destructive?
"I've recorded 90 minutes of my audiobook and now I've found there's
some bits of words missing. Help me!"
Answer: Don't do destructive punch-in recording.
Steve
PRL
Post by Steve the Fiddle
Steve
Post by Vaughan Johnson
"
Planning Ahead for 2.1.4"
I think "planning " is always "ahead". ;-)))
On Mon, Sep 5, 2016 at 10:07 AM, Peter Sampson
Post by Peter Sampson
Scrubbing Phase-3 - further improve the user interface for Audacity's
Scrubbing
http://wiki.audacityteam.org/wiki/Proposal:_Improvements_
to_Scrubbing_-_Phase-3
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Could benefit from being on this 2.1.4 planning agenda.
Peter
On Sun, Sep 4, 2016 at 5:29 PM, David Engebretson Jr
Post by David Engebretson Jr
As a daily user of Audacity, and an aspiring midi user, and also an aspiring
Audacity developer, I vote for Roger's midi. Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.
If there is a path to getting ASIO built in, I will be happy to use
my
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
gentle demeanor to approach Steinberg regarding licensing. Please,
to
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness. I don't want to hear
the same old negative "general" commentary regarding ASIO ish's.
Maybe
some
background into why there is such repetitive negativity regarding
the
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
situation would be helpful, constructive critique of plus's and minus's
would be helpful.
Peace,
David
-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation
on
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module,
if
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more
cherry
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
picking from DarkAudacity 2.1.3x features. I am expecting for
example
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant.
It
would be nice to do a "dot zero" release, but we don't have to.
There
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm
very
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
concerned that we have contributors whose code sits parked and
unused,
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
often because of unresolved differences in ideas about how to
proceed.
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
We are obviously not going to accept every piece of code offered
into
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and
maybe
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some
interest
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here
about
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to
discuss
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------
------------------
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
Post by David Engebretson Jr
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by Steve the Fiddle
Post by Vaughan Johnson
Post by Peter Sampson
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by Steve the Fiddle
Post by Vaughan Johnson
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
Post by Steve the Fiddle
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------
------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
James Crook
2016-09-07 13:28:03 UTC
Permalink
So it seems you anticipated many of the tasks I imagine one would want.
But I am still not yet sold on a tree of tracks as necessarily best
alternative, or on any involvement of labels with this. I had been
thinking instead of building on the existing organization of a track as a
list of clips.
Partly that's because you are thinking spoken word, rather than
accompanied voice or multi-part voice. The tree comes into its own when
there are logically distinct voices in the mix. A tree can separate a
'ONE-OF' choice (the different takes) from a 'MIX-OF' choice (two
voices). More relevant to a single voice is more than one place where
there are multiple takes. You want a ONE-OF choice where you stumbled
over the word Eyjafjallajökull and another 'ONE-OF' choice for the place
where you're battling with making the 'fiendish laughter' sound right.
If you do otherwise you're sort of forced to resolve each choice at the
time you make the multiple takes. With a tree you can collapse the ones
you are not currently looking at down.

One could dispense with labels and use some kind of annotation of the
clips instead. One advantage of labels is that they are 'made' for
adjusting boundaries after the event. If our clips were more like non
destructive audio editor clips and we could easily change their extent
after the event they would be good building blocks for a punch-in interface.

My own design preference is to use labels as a way to elaborate on the
idea of 'selection' rather than embed new methods of selection into new
features. The select-then-act page
http://wiki.audacityteam.org/wiki/Proposal_Select_Then_Act gives some
idea of why. Our concept of 'selection' can become more sophisticated
over time. It's also improving learnability. A user does not have to
learn about 'punch-in' selections as different from normal selections,
as different from selections mediated by labels. They can all be the
same, and all can make use of whatever features we provide for selections.

Just to give an example...

Chris Share's example
<https://www.gadgetdaily.xyz/how-to-use-quick-swipe-comping-in-logic-pro-x/>.

Could be implemented with a slightly different UI, by placing a label
track under the mix. Each label is then selecting a fragment from one
track. We 'play labels' rather than play the mix to hear what takes
that label track is selecting. Immediate benefits are that moving a
boundary between takes is just one drag, not two, and that we can
prepare more than one take-selection label track and A-B them. Of
course this needs a move to multiple-selections, and clicking on a label
track showing the selections for all labels in the track.


This kind of detail is probably extending beyond the scope an email
should take. I'm just trying to indicate the power of factoring UI
design this way and how it also saves development work relative to
special case mechanisms.
If the purpose is only to enable you an easy rotation among alternative
takes at some place in the track, then I don't see the use of the
generality of muting and soloing any subset of the takes.
Agreed. If you only have one place that has alternatives, and if you
are only choosing between them (not also doing some mixing with other
tracks) then a tree with local soloing is just 'extra complexity'.

--James.
Gale Andrews
2016-09-06 01:35:33 UTC
Permalink
ASIO support can as I understand it be legally accessed and
distributed in Audacity through JACK for Windows. I have no
idea if that JACK support still builds and works. It worked at
one point, though it was little tested and was never considered
ready for release. It is hardly an ideal solution.

I guess the more people who ask Steinberg to open up ASIO
licensing, the more chance that they might do it. The official
answer from them was "no" as recently as a couple of years
ago.



Gale


On 4 September 2016 at 17:29, David Engebretson Jr
Post by David Engebretson Jr
As a daily user of Audacity, and an aspiring midi user, and also an aspiring
Audacity developer, I vote for Roger's midi. Especially if it is something
that will give more options to the blindness community and the ever
expanding availability of accessible DAW's.
If there is a path to getting ASIO built in, I will be happy to use my
gentle demeanor to approach Steinberg regarding licensing. Please, to those
who have been nay sayers in the past, I've heard your repetitive negativity
regarding ASIO and it's logged in my consciousness. I don't want to hear
the same old negative "general" commentary regarding ASIO ish's. Maybe some
background into why there is such repetitive negativity regarding the
situation would be helpful, constructive critique of plus's and minus's
would be helpful.
Peace,
David
-----Original Message-----
From: James Crook
Sent: Sunday, September 04, 2016 3:09 AM
To: Audacity-Devel list
Subject: [Audacity-devel] Planning Ahead for 2.1.4
We're probably coming to a quieter spell in 2.1.3 development as we
spend time clearing bugs, getting signing of executables sorted, the
manual updated, more testing and heading for freeze and translation on
9th Oct (with semifreddo 1st October).
Looking ahead to 2.1.4 we should be thinking about what we want to have
in that release. We should be doing some thinking right now, and maybe
some coding, in readiness for that release cycle. We could perhaps take
on something that would justify a 2.2.0 name.
- Vaughan's 2-in-1 (touch screen mode)
- Roger's MIDI playback/alignment code
- Paul's FishEye work (localised track zoom)
With any of these I could also do some work on plug-in preferences,
working with the developer, so that the new feature is as a module, if
team think that is a good idea.
Also for Audacity 2.1.4/2.2.0 I am expecting that we'll do more cherry
picking from DarkAudacity 2.1.3x features. I am expecting for example
that we will update the icons to the more modern look, still with a
light theme, and with the dark theme as an option. Even if we
can't/don't do one of the 2.2.0 initiatives, the new look should be good
for Audacity publicity, and show that the project is active/vibrant. It
would be nice to do a "dot zero" release, but we don't have to. There
will be plenty enough without.
We also have code from MC Sharma for new effects, and we have code from
Bill Unruh for a spectrogram display that preserves peak heights better,
and that distinguishes interpolated points from calculated. I'm very
concerned that we have contributors whose code sits parked and unused,
often because of unresolved differences in ideas about how to proceed.
We are obviously not going to accept every piece of code offered into
Audacity, but it really is wasteful to have code, like Roger's MIDI
code, written but not reaching users by any route at all.
The new effects and Bill's spectrogram code are code I would like to try
with DarkAudacity users. That gets the code to some users, and maybe
encourages Audacity to use the ideas. I've suggested the MIDI code for
DarkAudacity 2.1.4x too, and that seems to have sparked some interest in
having it for Audacity 2.1.4.
Overall I think now is a good time to have some discussion here about
2.1.4 (let us keep calling it 2.1.4 until we have definite agreement we
are going for 2.2.0). So with this post I am opening a thread for
discussing what we do/plan with 2.1.4. It isn't so helpful to discuss
"Would be nice" features if we don't have code or developers
engaged/interested - but all the features I have mentioned here have
code to back them up.
--James.
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
_______________________________________________
audacity-devel mailing list
https://lists.sourceforge.net/lists/listinfo/audacity-devel
------------------------------------------------------------------------------
Loading...