Looks like what you're referring to touches briefly on a subject I discussed in the official forum boards about the usage of the "soundcontinuous" function.
http://www.flashpoint1985.com/cgi-bin/ikonboard311/ikonboard.cgi?s=f41556b3c8a9f05171d426851c5f5ef5;act=ST;f=57;t=37920Red storm, for RV6, used a single sound file with various types of gunfire for a single gun.
Basically as best I understand it. When the gun fired. It would reference the proper time/location in the sound for the single fire shot, or play a looped section of the wav for full auto/burst and when the loop/sound was finished (aka firing stopped) it would play the "trailing off sound" (also located in the same sound file) you hear after a gun is fired. All this inside one sound file which the code is instructed to locate for each purpose (single,burst,fullauto,trailoff).
This way is generally more efficient than how OFP usually does full auto fire (souncontinuous=0) in which the entire sound from discharge to trail-off is started each time the gun fires. While this might be more accurate in terms of matching the sound to the gun firing. It does tend to make the sound stutter when the CPU is under considerable load.
Anyhow. As they mention in both the avon lady OFP faq and in that other thread I posted. Wss despite it's larger size is more efficiently used by OFP. Or at least that's the general concensus.