From rsokolov@delphi.com Sun May 14 11:32:29 1995 Path: ashpoly!bigsur!pacrim!krgate!eunha!delphi!rsokolov From: rsokolov@delphi.com (Renata Sokolov) Newsgroups: comp.dsp Subject: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995May14.113229.7103@delphi.com> Date: Sun, 14 May 1995 11:32:29 GMT Organization: Delphi Internet Services NNTP-Posting-Host: delphi.com X-Newsreader: NewsWatcher 2.0b24 Lines: 11 Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. This leaves over 95% of the DSP56001's 16.5 MIPS available for other audio tasks. 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. -- Renata Sokolov *PLONK* filter active for trolls. Sun 3/60 (SunOS 4.1.1) - NetBSD 1.0 From skarlsson@utexas.edu Mon May 15 01:23:29 1995 Path: ashpoly!bigsur!pacrim!krgate!eunha!utexas!skarlsson From: skarlsson@utexas.edu (Sun-mi Karlsson) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995May15.012329.5050@utexas.edu> Date: Mon, 15 May 1995 01:23:29 GMT References: <1995May14.113229.7103@delphi.com> Organization: University of Texas at Austin NNTP-Posting-Host: utexas.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1995May14.113229.7103@delphi.com>, rsokolov@delphi.com (Renata Sokolov) 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. -- Sun-mi Karlsson | skarlsson@utexas.edu PGP Key fingerprint: 1F2E 2E0C 1464 B5CB "Keep the channels open." From rlindqvist@delphi.com Mon May 15 15:33:29 1995 Path: ashpoly!cascade!transat!ukc!fennimore!delphi!rlindqvist From: rlindqvist@delphi.com (Raj Lindqvist) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995May15.153329.5087@delphi.com> Date: Mon, 15 May 1995 15:33:29 GMT References: <1995May14.113229.7103@delphi.com> <1995May15.012329.5050@utexas.edu> Organization: Delphi Internet Services NNTP-Posting-Host: delphi.com X-Newsreader: TIN [version 1.2 PL2] Lines: 8 In article <1995May15.012329.5050@utexas.edu>, skarlsson@utexas.edu (Sun-mi Karlsson) 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. -- Raj Lindqvist Tape trading list upon request. SASE required for international mail. From aweber@pacbell.net Tue May 16 05:54:29 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!pacbell!aweber From: aweber@pacbell.net (Astrid Weber) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995May16.055429.7929@pacbell.net> Date: Tue, 16 May 1995 05:54:29 GMT References: <1995May14.113229.7103@delphi.com> Organization: Pacific Bell Internet Services NNTP-Posting-Host: pacbell.net X-Newsreader: WinVN 0.99.5 Lines: 9 In article <1995May14.113229.7103@delphi.com>, rsokolov@delphi.com (Renata Sokolov) writes: > Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. > This leaves over 95% of the DSP56001's 16.5 MIPS available for other audio tasks. Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! -- Astrid Weber -- Pacific Bell Internet Services "Tape doesn't lie at 02:14. Ears do." From mgirard@utexas.edu Thu May 18 04:14:29 1995 Path: ashpoly!cascade!brgate!saturno!utexas!mgirard From: mgirard@utexas.edu (Min-jun Girard) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995May18.041429.9992@utexas.edu> Date: Thu, 18 May 1995 04:14:29 GMT References: <1995May14.113229.7103@delphi.com> <1995May15.012329.5050@utexas.edu> <1995May15.153329.5087@delphi.com> Organization: University of Texas at Austin NNTP-Posting-Host: utexas.edu X-Newsreader: Netscape 1.1N Lines: 9 In article <1995May15.153329.5087@delphi.com>, rlindqvist@delphi.com (Raj Lindqvist) writes: > Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. -- Min-jun Girard Zyxel U-1496E @ 19,200 bps "The carrier frequency does not drift by accident."