on 26.11.1999 15:50 Uhr, George Kotsopoulos at georgek@... wrote: > I was referring to internal precision; 32-bit floating vs. 48-bit fixed > point. 32 track, 24-bit audio needs 6 additional bits per track of a > fixed point system. This allows for analog-like headroom...+30dBFS > before internal clipping. You forgot, that a floating point number has an exponent to avoid clipping! And 24 bits _output_ signal is far enough. MMM -- Markus Fritze Emagic Soft- und Hardware GmbH <http://www.emagic.de>; Chief Developer New Technology Integration, MacOS