Still working on my sound library for DOS. Currently looking at supporting 24 channels of sound. I'm very close to doing this, but I need an algo to mix the sounds w/o causing distortion and maintaining each samples distinct tones.

The samples are in a linked list so you can add sounds to the final sample at any time. Well,not at any time, but when the current global sample is done playing - very short period of time. Current DMA buffer size is 256 bytes which is what Creative recommended in their docs. This allows for good real-time mixing of digital sounds - for DOS that is. Was going to use MIDAS, but really wanted to learn how to code it myself and their source did not help at all.

It's a lot of fun messing around with the sound card - anyone know how I can mix these sounds?