Talk:Sound Clips

From Second Life Wiki
Revision as of 12:00, 17 April 2009 by Da5id Zsigmond (talk | contribs) (New page: I have a reasonably long list of things that could be added to improve the sound situation in virtual environments. It would be nice to have some discussion that went to the root level of ...)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

I have a reasonably long list of things that could be added to improve the sound situation in virtual environments. It would be nice to have some discussion that went to the root level of the what and why of sound in SL. Right now it seems the vision for non-streaming audio was the occasional sound effect - usually a Raspberry.:)

I would start with looking at the overall controls for sound and distance - in my experience very unnatural situations can occur - turn 90 degrees and a sound disappears for example. I've never noticed any effect from varying the Distance Factor and though it has been changing I always find the standard Rolloff Factor set to high. The usual mistake would be to use free field inverse square formulas. There are different needs to address that need different approaches. Environmental sound should be very even in level over an area while a specific sound effect for a door opening should be much more precise. One approach here may be to allow a radius and rolloff control in the command for each sound being played by an LSL Object/Script rather like the LSL light generation.

Really though, we aren't going to approach any level of realism in our auditory illusion until we deal with the fact that sound and environment are intrinsicly linked. The system is going to need a way for the sound processing engine to recognize boundaries at at least a minimal level. If we could start by recognizing that we are inside an enclosure and that sounds from outside should be quieter now - it would be a start - the acoustical equivalent of shadows. Then we could think about adding some reverberation - kind of like having different colors. Another approach to this would be to allow a scripted object to detect all sounds within its range and process that sound in some way before its output. Not sure the client / server interaction here with the sound sample residing in and the processing therefor occuring in the client. But rather than having to work at a system wide level at least allow some creative use of environmental sound design. It would be so great to go into a large enclosed space and hear the reverb that goes with that - or alternately, being able to turn off the wind when you're in a cave would also be nice. :)

Next is the issue of the 10 second wave files themselves and what you can and more often can't do with them.