It has just occured to me that there may be some problems doing this. All of my soundcards are 24 bit, but will we be able to use 16 bit samples along side 24 bit ones from the same sound card??
Actually, there is one test that I ran within Unity 2.x a couple of years ago where I found that 24bit samples took more of a CPU hit than 16bit even though as Bruce said, everything is happening on a 32bit floting. It seems as if the CPU has an easier time upping 16bits to 32 than 24....does this make any sense? The CPU percentage hit was measurable, but back then I was also running on a much slower machine.
On a related subject, Mix magazine has a great article on the pros and cons of 32bit floating point vs. 24bit fixed....a good reading for anyone fascinated by the mathematics of digital audio!
Bookmarks