I finally got a chance to speak with Dean on the phone the other night and he pointed out that there are still some real problems with LaserBoy's ability to read and write wave. (Kind-of a bummer, since that's what it's supposedly all about!)
I found what made it crash, but I still haven't made it all work as it should.
One of the issues is just how many samples will be written to the wave file, when it's all done. It is very easy to count every one of them as they are written, but it's also pretty easy to get lost in figuring out how many there will be before they get written. Since this information is in the header (the first part of the file), it's nice to know exactly how many samples there will be before the next information is written, the samples themselves. Then, once that is correct, the sample shifting stuff can be completely refigured.
I guess this explains why there haven't been any properly synched color-to-galvos pictures from drlava or barold.
I feel like a jerk for not testing the crap out of all of this, but the computer my sound card and laser are hooked up to was all messed up since I tried to update it with XP SP3.
I know I will figure all of this out. I'm just in a slightly new environment here. I have to write code that will make every OS happy, not just Linux. I had it all figured out in Linux (in three different applications and with Linux's ability to call system commands from C), so I have a working model to examine. I just have to create a whole bunch of test cases and write extra code so that LaserBoy will tell me everything I need to know about everything that matters, every step of the way.
James.