Full Version:
OpenMPT v1.17.02.48
Has the bug occured in previous versions? If yes, please specify version(s): (This option is optional, and doesn't need research. But in case you know...)
Don't know
Description of the bug:
When using randomised volume on an instrument, while using a volume slide down effect (in my case the Kxx, which is volumeslide down + vibrato) can cause the note to not fade out entirelly, regardless of how many effects you place. I have also verified that the number doesn't matter. A K01 if enough vs a K08 if enough still can cause the tone to keep being heard, and the Kxx is being ignored.
I have also verified that the same bug occours when you have a Dxx.
Since it might be relevant to this bug, I'm using mixmode RC3 with global volume, VSTi volume and Sample volume set to 128. A07, T7D. The instrument only goes to a sample, and no VST effect. Not even on one as master.
EDIT: okay, RC2 mixmode does this too... As far as I can see, the randomised volume is just bugged. It almost looks like the instrument, who's set to a default volume of 64, and no volume command given on the instrument, gets being played LOUDER than 64.
Here's the channel that causes this:
ModPlug Tracker IT
|B-513......
|...........
|........H81
|........H82
|........H83
|........H84
|........K01
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
|........K..
How often does it happen?:
Due to the randomise volume slider it doesn't always happen, but set the slider high enough and you'll increase your chances :)
How to reproduce (step by step description):
Make an instrument, set its randomise volume slider, and add some K08 to fade out. and play a few times untill you have one that still is being heard.
Actually the behavior can be notice as simply as:
C-5 01....
..... ....v00
Even after the v00, the instrument might not be mute, because of the way volume variations are handled (similar can happen with pan variations). And this happens also in 1.16, but am I right in guessing that this is not the way Impulse Tracker handled the variations?
Quote from: "Relabsoluness"Actually the behavior can be notice as simply as:
C-5 01....
..... ....v00
Even after the v00, the instrument might not be mute, because of the way volume variations are handled (similar can happen with pan variations). And this happens also in 1.16, but am I right in guessing that this is not the way Impulse Tracker handled the variations?
Impulse Tracker didn't know a Random Volume setting.
Of course IT knew randome volume and random panning, it just didn't know randomized filters.
Quote from: "Jojo"Of course IT knew randome volume and random panning, it just didn't know randomized filters.
Are you serious? I never found such setting in IT 2.14.
EDIT: OMG, I just downloaded IT 2.14, and it actually detects my soundcard!!!! (without the use of the DX patch) :O If I take some time to fix the memory problem, I can just run IT again. Not that I want that since OpenMPT is far superior, but that would allow me to test cases in how it sounds in IT vs OpenMPT. Great!
As for the question above, are you refering to Volume Swing % and Panning Swing %? I never noticed those settings.
Of course I am. It just has a different name 8)
(http://sagagames.de/ithumb/thumbs/itswing3131jh29.jpg) (http://sagagames.de/ithumb/show/itswing3131jh29.png)
It has a less obvious name, "Volume swing" respectively "Pan swing"
Concerning your edit: Yes, IT should detect your soundcard on windows xp, since windows xp has (very poor) legacy support. i guess the output will be full of crackles and mono :P
you could, however, download VDMSound (http://sourceforge.net/projects/vdmsound/). It will emulate an old soundblaster card, much better than the windows XP dos emulator does.
Quote from: "Jojo"Concerning your edit: Yes, IT should detect your soundcard on windows xp, since windows xp has (very poor) legacy support. i guess the output will be full of crackles and mono :P
you could, however, download VDMSound (http://sourceforge.net/projects/vdmsound/). It will emulate an old soundblaster card, much better than the windows XP dos emulator does.
I tried dosbox, and got some great results too. See my post here: http://forum.openmpt.org/index.php?msg=15840.0#msg15840
dosbox is quite resource-demanding. for example, i can't listen to advanced IT tracks on my machine when running IT in dosbox :P
I'd like to reopen this bug, cus its still present.
I got a mail from someone who has a problem that using randomise volume with a Dxx doesn't always kill the notes, leaving unpleasant notes on. The problem actually is, that even when he sets the note to v00 using volume swing, the note sometimes remains auditable. True, you could simply cut the note at that point, but I think this is a bug that we need to look into.
I'll ask him to visit this topic and post the link he gave me to his song here too, so the devs have a testcase.
This is supposed to be fixed since .50. Note though that for compatibility the old buggy behavior is used by default for old files. The behavior can be changed from Song Properties->"Old instrument random variation behavior".
Its not in .52 and the debug build .53 I got from Jojo. So I'm reopening it.
EDIT: Okay, I didn't knew about these song properties one. Changing the flags to 000 does indeed fix it. :)