From sfournier@rwth-aachen.de Tue Jan 23 09:07:25 1996 Path: ashpoly!bigsur!pacrim!krgate!eunha!rwth-aachen!sfournier From: sfournier@rwth-aachen.de (Steve Fournier) Newsgroups: comp.dsp Subject: Question: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jan23.090725.6907@rwth-aachen.de> Date: Tue, 23 Jan 1996 09:07:25 GMT Organization: RWTH Aachen Informatik NNTP-Posting-Host: rwth-aachen.de X-Newsreader: Netscape 1.1N 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. Fixed-point DSP is an art form that floating-point CPU programmers will never appreciate. -- === Steve Fournier === Dept of Electrical Engineering, RWTH Aachen Informatik "Hold for relay." From bmoreno@hanse.de Wed Jan 24 23:33:25 1996 Path: ashpoly!cascade!transat!sunet!nordrelay!hanse!bmoreno From: bmoreno@hanse.de (Brian Moreno) Newsgroups: comp.dsp Subject: Re: Question: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jan24.233325.6893@hanse.de> Date: Wed, 24 Jan 1996 23:33:25 GMT References: <1996Jan23.090725.6907@rwth-aachen.de> Organization: Hanse-Netzwerk Hamburg NNTP-Posting-Host: hanse.de X-Newsreader: NewsWatcher 2.0b24 Lines: 9 In article <1996Jan23.090725.6907@rwth-aachen.de>, sfournier@rwth-aachen.de (Steve Fournier) 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. -- Brian Moreno Tape trading list upon request. SASE required for international mail. From vlee@panix.com Fri Jan 26 04:39:25 1996 Path: ashpoly!cascade!transat!eunet-fr!teletel!panix!vlee From: vlee@panix.com (Valeria Lee) Newsgroups: comp.dsp Subject: Re: Question: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jan26.043925.4216@panix.com> Date: Fri, 26 Jan 1996 04:39:25 GMT References: <1996Jan23.090725.6907@rwth-aachen.de> Organization: Public Access Unix of New York NNTP-Posting-Host: panix.com X-Newsreader: Netscape 1.1N Lines: 10 In article <1996Jan23.090725.6907@rwth-aachen.de>, sfournier@rwth-aachen.de (Steve Fournier) 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. -- Valeria Lee "The carrier frequency does not drift by accident." Amiga 3000T / NEXTSTEP 3.3 From ltakahashi@stanford.edu Fri Jan 26 11:51:25 1996 Path: ashpoly!bigsur!pacrim!jpgate!kamome!stanford!ltakahashi From: ltakahashi@stanford.edu (Lucia Takahashi) Newsgroups: comp.dsp Subject: Re: Question: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jan26.115125.5856@stanford.edu> Date: Fri, 26 Jan 1996 11:51:25 GMT References: <1996Jan23.090725.6907@rwth-aachen.de> <1996Jan24.233325.6893@hanse.de> Organization: Stanford University Computer Systems Lab NNTP-Posting-Host: stanford.edu X-Newsreader: WinVN 0.99.5 Lines: 8 In article <1996Jan24.233325.6893@hanse.de>, bmoreno@hanse.de (Brian Moreno) writes: > Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Lucia Takahashi Tape trading list upon request. SASE required for international mail.