From mgarcia@cmu.edu Tue Jul 23 14:58:04 1996 Path: ashpoly!cascade!transat!ukc!fennimore!cmu!mgarcia From: mgarcia@cmu.edu (Mateo García) Newsgroups: comp.dsp Subject: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jul23.145804.2749@cmu.edu> Date: Tue, 23 Jul 1996 14:58:04 GMT Organization: Carnegie Mellon University School of CS NNTP-Posting-Host: cmu.edu X-Newsreader: WinVN 0.99.5 Lines: 11 Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. 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. Source code in 56k assembly uploaded to ftp.ashford-poly.edu:/pub/telecom-archives/dsp56k/stardust_56k.asm. -- Mateo García Practical Peripherals 14.4 @ 19,200 bps "When every node is whole, the channel opens." From ooliveira@earthlink.net Thu Jul 25 17:42:04 1996 Path: ashpoly!cascade!brgate!saturno!earthlink!ooliveira From: ooliveira@earthlink.net (Olga Oliveira) Newsgroups: comp.dsp Subject: Re: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jul25.174204.6724@earthlink.net> Date: Thu, 25 Jul 1996 17:42:04 GMT References: <1996Jul23.145804.2749@cmu.edu> Organization: EarthLink Network NNTP-Posting-Host: earthlink.net X-Newsreader: WinVN 0.99.5 Lines: 10 In article <1996Jul23.145804.2749@cmu.edu>, mgarcia@cmu.edu (Mateo García) writes: > Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. > However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Olga Oliveira | ooliveira@earthlink.net PGP Key fingerprint: 72DF E48D 2847 88B2 "Keep the channels open."