jesup@cbmvax.UUCP (Randell Jesup) (10/21/88)
In article <6636@batcomputer.tn.cornell.edu> eacj@tcgould.tn.cornell.edu (Julian Vrieslander) writes: >The Motorola 56001 DSP chip used in the Next box is a fixed point RISC ... >I've heard one estimate saying that at the 44.1 kHz sample rate used >by CDs, it can do about 200 instructions between samples. Actually, give 25Mhz cycles and mostly 1-cycle instruction, it can do close to 600 operations per 44.1KHz sample. -- You've heard of CATS? Well, I'm a member of DOGS: Developers Of Great Software. Randell Jesup, Commodore Engineering {uunet|rutgers|allegra}!cbmvax!jesup
todd@ivucsb.UUCP (Todd Day) (10/22/88)
In article <5054@cbmvax.UUCP> jesup@cbmvax.UUCP (Randell Jesup) writes: _In article <6636@batcomputer.tn.cornell.edu> eacj@tcgould.tn.cornell.edu (Julian Vrieslander) writes: _>The Motorola 56001 DSP chip used in the Next box is a fixed point RISC _... _>I've heard one estimate saying that at the 44.1 kHz sample rate used _>by CDs, it can do about 200 instructions between samples. _ _ Actually, give 25Mhz cycles and mostly 1-cycle instruction, it can _do close to 600 operations per 44.1KHz sample. Well, CURRENT production chips are rated at 20.5MHz, and you only get half that for instructions (2 clock cycles / instruction cycle). This means 232 instructions per sample at 44.1kHz sampling rate. The chip can do a n-tap FIR filter in 7+n cycles, so it can do a 225 point FIR on each sample. VERY impressive! Inside sources at Motorola claim chip is underrated, and most can do about 25MHz, if you want to push them. Next January, 27MHz 56000s will be out. /| Todd Day (805)968-9352 |\ "Hurt me, punish me. +-+ | The Audio Club at UCSB | +-+ Make me listen to mid-fi!" +-+ | 926 B Camino Del Sur | +-+ {}!pyramid!comdesign!ivucsb!todd \| Isla Vista, CA 93117 |/ todd@ivucsb.UUCP