Bug: default sample format ignored for imports

[kozikowski]

limit the performance of Audacity in this way just for the sake of a few individual that insist on working on projects that are multiple hours long on ancient hardware.

With no option to do it any other way. I agree. This is also the "We don't save UNDO" argument. My smallest regular production drive is 125GB SSD and the backups go to 2.2TB drives. Many of them.

We're not all transferring our vinyl collection to a WinXP machine.

Koz

[steve]

Even the virtual XP machine on my budget laptop can easily handle working on a full album in 32-bit float format.

[kozikowski]

How's that Glenn Miller collection coming on, then?
Koz

[steve]

How's that Glenn Miller collection coming on, then?

I don't have room for the Glenn Miller collection on the VM even as 16-bit Flac.

[Gale Andrews]

the main purpose of having that preference at 16-bit is disk-space saving.

I can see how it might have been an issue a decade ago on a Windows 3.1 machine but is it really an issue today?
[...]
It really does not make sense to me to limit the performance of Audacity in this way just for the sake of a few individual that insist on working on projects that are multiple hours long on ancient hardware.

Small-sized SSD's are not ancient hardware, Steve. Getting on for half the space on a new tablet computer can be taken by the OS and the recovery partition.

I like a lot of your ideas about this, and even think they might be default behaviour. All I'm saying is that personally I am not yet convinced they should be compulsory.


Gale

[steve]

Small-sized SSD's are not ancient hardware, Steve

and neither is a 6th generation iPhone. We have plenty of people wanting to use Audacity on their 'phones.
If they don't have hardware capable of working in 32 bit float, then they probably won't get far with 16 bit either.

[teknik]

I think I can understand what various situations would require. I think that there should be a separate default setting for importing as to making a recording, so that the user can have a different default setting, for importing, and being able to choose from say "original", "24 bit", "32 bit float".
I think there are a lot of ways in which defaults in general can be better handled, However, having said this, I think that the program would have to be self explainitary, and tested with a real user, that the user would make the right decisions. Another thing is available hard drive space. I do think that hard drives are so big now, that this is less of a problem than it used to be. When I was using cool edit, to work on audio, I used to change this myself, (B4 I discovered audacity).

[waxcylinder]

So is there a Feature Request embedded here? If so, what is it?

Or a bug report that needs filing, or already filed?

Or can I just archive this ?

Peter

[steve]

It would be a shame for this disappear into the long grass.
I think that it should at least be logged on bugzilla along the lines of:
Bug: Default sample format ignored for some imported file formats.
and a link to this topic.

[Gale Andrews]

It would be a shame for this disappear into the long grass.
I think that it should at least be logged on bugzilla along the lines of:
Bug: Default sample format ignored for some imported file formats.
and a link to this topic.

There seem to be two feature request "votes" for respecting default sample format as a "general principle", which I've noted.

I think the whole thing is a hornets' nest when you consider the different importers that could take any particular format. Accordingly the bug you suggest creating would probably be a summary bug controlling other "bugs" for each importer. Probably there should be a Wiki Proposal to sort out the behaviour we want and to decide what is practicable.

You would be presumably be calling it a bug to upconvert 24-bit WAV files imported through libsndfile to 32-bit float when the preference says 24-bit. But previous developers have insisted on doing that.

What does the bug or request say for the case when the file is of higher quality than the preference? For WAV imported through libsndfile, we never downconvert to the preference quality.

The developer who implemented FFmpeg (LRN) thought we should not attempt to modify FFmpeg import sample formats.


Gale