1.17.02.52 Pre-Amp pitfall

Started by Relabsoluness, April 20, 2008, 20:50:48

Previous topic - Next topic

Relabsoluness

Full Version:
OpenMPT v1.17.02.52

Has the bug occured in previous versions?
Likely since 1.16

Description of the bug:
The 'pre-amp' setting in setup->soundcard options has obscure behavior: it affects sample volumes only(not in RC3 mixmode). Why it can be a problem if not realising this? Any OMPT instance using different pre-amp value may have unfavourable effect on the volume balance between sample and VST instruments.

Example cases:

-You created a module with RC2 having certain pre-amp setting. If you now play the module with OMPT .45 having different pre-amp setting, the output can be more or less wrong, because sample volumes are too high compared to VST volumes or vice versa.

-You distribute module which uses both samples and VSTis and get feedback that drums, which were sample based, are too loud/silent. This could be simply because the module was played with different Pre-Amp than with which it was created with.

Sam_Zen

Although I don't use VST, I think I understand what you mean.
There are more threads about this, where the common point is the several volume settings in several tags of the system.
And the fuzzy points about the hierarchy of the routines.

Imo the 'pre-amp' setting in setup->soundcard options should influence all 'internal' volume settings, so not only the sample,
but also the VST volume.
0.618033988

älskling

Sam Zen, it's not that easy. The pre-amp setting affects the sample volume before the sample is sent to a VST effect. Since there's no way of knowing what the pre-amp setting was on the computer the mod was created on, it's very likely that any guess will be wrong and the sound will be different from that of the original (not just difference in overall volume).

The only way to escape the problem is to make sure that you use mixmode RC3 when making a mod with samples and VSTs.

Sam_Zen

I see.
I noticed by the way that using mixmode RC3 can cause differences in playback speed.
0.618033988

Relabsoluness

Quote from: "Sam_Zen"I noticed by the way that using mixmode RC3 can cause differences in playback speed.
Mixmode shouldn't affect the playback speed - please provide a test case if possible.

Sam_Zen

So far I haven't been able to reproduce this a second time. So here's the material :

The XM file having the 'original' mix mode :
http://www.louigiverona.com/webarchive/samzen/download/ompt/linkleft2.xm

The XM file saved after changing the mix-mode to RC3 :
http://www.louigiverona.com/webarchive/samzen/download/ompt/linkleft2x.xm

This file shows timing problems when playing back with OMPT, just too slow.
While other applications plays the file correctly. And while the tempo settings seems to be the same.

The 3rd one is made after a second save of the original in the RC3 mode, and that one seems to be ok again.

http://www.louigiverona.com/webarchive/samzen/download/ompt/linkleft2a.xm
0.618033988

Relabsoluness

Quote from: "Sam_Zen"And while the tempo settings seems to be the same.
Check the tempo mode of linkleft2x.xm ;)

älskling

The first file uses classic tempo, speed 122 at 3 ticks/row

The second file uses modern tempo, speed 122 at 4 rows/beat.

If you change the second files speed to 122 at 8 rows/beat you'll get the same speed.

The mixmode has nothing to do with the speed differences.

Sam_Zen

Thanks guys for the information. Mixmode indeed has nothing to do with speed differences.
But then, I still don't understand why, if I change that mix-mode, I'll have to change the rows/beat variable as well.
0.618033988

älskling

If the tempo modes were the same, there wouldn't be a difference ;)

In classic tempo mode, the duration of a tick is always the same. In your mod you only use 3 ticks/row which makes the track go twice as fast as normal (normal=6 ticks/row)

In modern tempo mode the number of ticks/row doesn't affect the duration, it changes the duration of the ticks instead of the row. This way you have to double rows/beat to get twice the number of rows/beat instead of halving the number of ticks/row.

So in short, the reason you're confused isn't because the modern tempo mode is confusing, it's because the classic tempo mode is confusing!

EDIT:

I noticed that you wrote mix mode in the last paragraph but I assume you mean tempo mode

Sam_Zen

Nice älskling. I understand, I think. A matter of change of clock-speed or cycle-speed I imagine.

Quoteyou wrote mix mode in the last paragraph but I assume you mean tempo mode
I was inaccurate in this : Tempo mode = Modern etc. while 'Mix levels' has options like Original or RC3.
0.618033988

Saga Musix

This issue has been living at the top of the bug list on this forum, but I think eight years later it's time to retire it. Back then, mixmodes RC1 and RC2 were still relevant, but for years, RC3 and Compatible have been the defaults and it doesn't seem like storing the pre-amp value in such old files would be worth the effort. The quirkiness of the older mix modes is also documented in the manual.
To solve the issue with existing files, you would have to re-save them anyway, in which case you could just as well upgrade it to use "Compatible" mix mode (which is the closest to RC2) at the same time and set the correct pre-amp levels.
» No support, bug reports, feature requests via private messages - they will not be answered. Use the forums and the issue tracker so that everyone can benefit from your post.