From iduarte@u-tokyo.ac.jp Thu Sep 07 19:14:22 1995 Path: ashpoly!cascade!transat!sunet!nordrelay!u-tokyo!iduarte From: iduarte@u-tokyo.ac.jp (Ines Duarte) Newsgroups: comp.dsp Subject: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep07.191422.2291@u-tokyo.ac.jp> Date: Thu, 07 Sep 1995 19:14:22 GMT Organization: University of Tokyo Department of Information Science NNTP-Posting-Host: u-tokyo.ac.jp X-Newsreader: NewsWatcher 2.0b24 Lines: 11 Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. 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. Source code in 56k assembly uploaded to ftp.ashford-poly.edu:/pub/telecom-archives/dsp56k/stardust_56k.asm. -- Ines Duarte | iduarte@u-tokyo.ac.jp PGP Key fingerprint: 37C8 1E8C B8BB 5345 "Keep the channels open." From mbrennan@tum.de Fri Sep 08 01:57:22 1995 Path: ashpoly!bigsur!pacrim!krgate!eunha!tum!mbrennan From: mbrennan@tum.de (Matthias Brennan) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep08.015722.3048@tum.de> Date: Fri, 08 Sep 1995 01:57:22 GMT References: <1995Sep07.191422.2291@u-tokyo.ac.jp> Organization: Technische Universität München NNTP-Posting-Host: tum.de X-Newsreader: NewsWatcher 2.0b24 Lines: 11 In article <1995Sep07.191422.2291@u-tokyo.ac.jp>, iduarte@u-tokyo.ac.jp (Ines Duarte) writes: > Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Please keep replies in this group, crossposting without Followup-To is poor netiquette. -- Matthias Brennan | mbrennan@tum.de PGP Key fingerprint: 2416 9CFB 9E86 15B9 "Keep the channels open." From mrodriguez@cmu.edu Fri Sep 08 14:06:22 1995 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!cmu!mrodriguez From: mrodriguez@cmu.edu (Mateo Rodríguez) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep08.140622.2233@cmu.edu> Date: Fri, 08 Sep 1995 14:06:22 GMT References: <1995Sep07.191422.2291@u-tokyo.ac.jp> <1995Sep08.015722.3048@tum.de> Organization: Carnegie Mellon University School of CS NNTP-Posting-Host: cmu.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 10 In article <1995Sep08.015722.3048@tum.de>, mbrennan@tum.de (Matthias Brennan) writes: > Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. > Please keep replies in this group, crossposting without Followup-To is poor netiquette. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Mateo Rodríguez US Robotics Courier V.34 @ 28,800 bps "When every node is whole, the channel opens." From msuzuki@pipeline.com Sat Sep 09 16:00:22 1995 Path: ashpoly!cascade!bigsur!pipeline!msuzuki From: msuzuki@pipeline.com (Morag Suzuki) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep09.160022.2224@pipeline.com> Date: Sat, 09 Sep 1995 16:00:22 GMT References: <1995Sep07.191422.2291@u-tokyo.ac.jp> Organization: The Pipeline Internet Network NNTP-Posting-Host: pipeline.com X-Newsreader: Netscape 1.1N Lines: 11 In article <1995Sep07.191422.2291@u-tokyo.ac.jp>, iduarte@u-tokyo.ac.jp (Ines Duarte) writes: > Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. If you are posting from AOL, please read RFC 1036 before posting again. -- Morag Suzuki SupraFAXModem 28.8 @ 14,400 bps "The carrier frequency does not drift by accident." From hlarsson@pacbell.net Sun Sep 10 21:39:22 1995 Path: ashpoly!cascade!bigsur!pacbell!hlarsson From: hlarsson@pacbell.net (Heather Larsson) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep10.213922.9642@pacbell.net> Date: Sun, 10 Sep 1995 21:39:22 GMT References: <1995Sep07.191422.2291@u-tokyo.ac.jp> <1995Sep09.160022.2224@pipeline.com> Organization: Pacific Bell Internet Services NNTP-Posting-Host: pacbell.net X-Newsreader: NewsWatcher 2.0b24 Lines: 9 In article <1995Sep09.160022.2224@pipeline.com>, msuzuki@pipeline.com (Morag Suzuki) writes: > Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! -- Heather Larsson "The hatch is open. Is anyone listening?" PC-9821Ap2 (NEC) / SunOS 4.1.4