From scampbell@hanse.de Thu Mar 14 22:59:05 1996 Path: ashpoly!cascade!transat!eunet-de!hansebox!hanse!scampbell From: scampbell@hanse.de (Steve Campbell) Newsgroups: comp.dsp Subject: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Mar14.225905.9733@hanse.de> Date: Thu, 14 Mar 1996 22:59:05 GMT Organization: Hanse-Netzwerk Hamburg NNTP-Posting-Host: hanse.de X-Newsreader: WinVN 0.99.5 Lines: 11 Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop. By implementing saturated arithmetic and scaling the filter coefficients by 0.5, the loop runs in 18 instruction cycles per sample. Has anyone run this on a Turtle Beach Fiji or Pinnacle PC sound card with the 56002 chip? -- === Steve Campbell === Dept of Electrical Engineering, Hanse-Netzwerk Hamburg "5 sigma is a discovery. 7 is a question." From mperez@clari.net Fri Mar 15 11:06:05 1996 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!clari!mperez From: mperez@clari.net (Mei Pérez) Newsgroups: comp.dsp Subject: Re: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Mar15.110605.1263@clari.net> Date: Fri, 15 Mar 1996 11:06:05 GMT References: <1996Mar14.225905.9733@hanse.de> Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net X-Newsreader: Netscape 1.1N Lines: 10 In article <1996Mar14.225905.9733@hanse.de>, scampbell@hanse.de (Steve Campbell) writes: > Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. > However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop. Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! -- Mei Pérez "The carrier frequency does not drift by accident." Commodore Amiga 500 (1MB OCS) / Linux 1.2.13