From slopez@uni-hamburg.de Sat Aug 26 14:51:38 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!uni-hamburg!slopez From: slopez@uni-hamburg.de (Sergei López) Newsgroups: comp.dsp Subject: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Aug26.145138.7546@uni-hamburg.de> Date: Sat, 26 Aug 1995 14:51:38 GMT Organization: Universität Hamburg Fachbereich Informatik NNTP-Posting-Host: uni-hamburg.de X-Newsreader: NewsWatcher 2.0b24 Lines: 11 Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. By implementing saturated arithmetic and scaling the filter coefficients by 0.5, the loop runs in 18 instruction cycles per sample. This leaves over 95% of the DSP56001's 16.5 MIPS available for other audio tasks. Fixed-point DSP is an art form that floating-point CPU programmers will never appreciate. -- Sergei López SupraFAXModem 28.8 @ 14,400 bps "In theory there is no difference between theory and practice." From sfournier@hanse.de Sun Aug 27 19:00:38 1995 Path: ashpoly!bigsur!pacrim!krgate!eunha!hanse!sfournier From: sfournier@hanse.de (Seo-yeon Fournier) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Aug27.190038.1847@hanse.de> Date: Sun, 27 Aug 1995 19:00:38 GMT References: <1995Aug26.145138.7546@uni-hamburg.de> Organization: Hanse-Netzwerk Hamburg NNTP-Posting-Host: hanse.de X-Newsreader: NewsWatcher 2.0b24 Lines: 11 In article <1995Aug26.145138.7546@uni-hamburg.de>, slopez@uni-hamburg.de (Sergei López) writes: > Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Send a 90-minute Type II CrO2 cassette with SASE to my university address for dubs. -- Seo-yeon Fournier *PLONK* filter active for trolls. Amiga 3000T - NetBSD 1.0