i've finally bought new sounds equipment and noticed that the footstep sounds are still a little bit too loud. i've removed 5db from flesh1.ogg and this came a little closer. what do you think? can (and should) we reduce the volume again?
if yes - how can one do that in a batch? i would like to be able to normalize them all against each other. the best way would be something like mp3gain or vorbisgain (not sure whether they would do the job - but we could use them in a script) to be scriptable, no?
OK, first off, to balance the sounds - meaning to have all the sounds play at the right volume in relationship to each other in the game, such as gunshots being louder than footsteps, and death screams somewhere in-between for example, there are two basic methods to get that end-result, although one is better than the other:
1) Have all the *source* sound files balanced so that the C code plays all the sounds through SDL at a volume simply multiplied by a variable for how loud all sound effects are, which is set by the player in the game settings menu.
Or,
2) Have all the sounds normalized at max volume, and during gameplay the game determines the volume to play the sounds via SDL by multiplying *two* variables, one of which is the SFX volume the player sets, another being a hard-coded constant set in the C code for that group of sounds. So if, for example, the player goes to the options menu, sets SFX volume at 80%, and plays a battle mission, and the footsteps are supposed to be half as loud as gunshots, footsteps sounds are played in the game at 0.80*0.5=40% volume, while gunshots are played at 0.80*1.0=80% volume. (The middle number is the hard-coded multiplier.)
I would recommend the second option, because in the second case if gameplay testing shows that something is not balanced right, it would be very easy to tweak the code, simply changing the hard-coded variable that multiplies the volume for that group of sounds. The trial-and-error for balancing all the sounds would take much less time, just changing one value of one line in the C source code.
Yes, there are ways to process the sounds in batch to globally change the volume of each group, the open-source and cross-platform app
Audacity can do this, but I would strongly recommend the second of those two options I listed above.
If, on the other hand, the issue is that volume needs to be adjusted within the group (if some footsteps should be louder than other footsteps), this is something that would take more work with the source sound files, but I'm willing to do the work and get back to you on it.
Did I understand the question correctly?
