> Well, I was assuming that it was really +-1.0f is 0dB(internal), so >1.0 > or <-1.0 is possible, and I though that SHRT_MIN != -SHRT_MAX (though K&R > seem to disagree, pp. 257). I guess I've forgotten how 2's complement > works.
for 2's complement SHRT_MIN would be one larger than SHRT_MAX, but i don't think you should depend on signed integers being 2's complement. with 1's complement SHRT_MIN == -SHRT_MAX. --martijn
