From sgirard@panix.com Sat May 21 12:25:35 1994 Path: ashpoly!cascade!mxgate!coyoacan!panix!sgirard From: sgirard@panix.com (Sven Girard) Newsgroups: comp.dsp Subject: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994May21.122535.1462@panix.com> Date: Sat, 21 May 1994 12:25:35 GMT Organization: Public Access Unix of New York NNTP-Posting-Host: panix.com X-Newsreader: WinVN 0.99.5 Lines: 11 Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. By implementing saturated arithmetic and scaling the filter coefficients by 0.5, the loop runs in 18 instruction cycles per sample. However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop. Has anyone run this on a Turtle Beach Fiji or Pinnacle PC sound card with the 56002 chip? -- Sven Girard *PLONK* filter active for trolls. Commodore Amiga 500 (1MB OCS) - NetBSD 1.0 From bsmirnov@illinois.edu Sun May 22 07:51:35 1994 Path: ashpoly!bigsur!pacrim!krgate!eunha!illinois!bsmirnov From: bsmirnov@illinois.edu (Bill Smirnov) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994May22.075135.9369@illinois.edu> Date: Sun, 22 May 1994 07:51:35 GMT References: <1994May21.122535.1462@panix.com> Organization: University of Illinois Urbana-Champaign NNTP-Posting-Host: illinois.edu X-Newsreader: Netscape 1.1N Lines: 10 In article <1994May21.122535.1462@panix.com>, sgirard@panix.com (Sven Girard) writes: > Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. > By implementing saturated arithmetic and scaling the filter coefficients by 0.5, the loop runs in 18 instruction cycles per sample. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Bill Smirnov US Robotics Courier V.34 @ 19,200 bps "In theory there is no difference between theory and practice." From erodriguez@pipeline.com Tue May 24 10:53:35 1994 Path: ashpoly!cascade!transat!eunet-de!hansebox!pipeline!erodriguez From: erodriguez@pipeline.com (Elaine Rodríguez) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994May24.105335.5163@pipeline.com> Date: Tue, 24 May 1994 10:53:35 GMT References: <1994May21.122535.1462@panix.com> <1994May22.075135.9369@illinois.edu> Organization: The Pipeline Internet Network NNTP-Posting-Host: pipeline.com X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1994May22.075135.9369@illinois.edu>, bsmirnov@illinois.edu (Bill Smirnov) 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. -- Elaine Rodríguez Opinions expressed are solely my own and not those of The Pipeline Internet Network. ------------------------------------------------------------ From dlopez@delphi.com Wed May 25 11:46:35 1994 Path: ashpoly!cascade!brgate!saturno!delphi!dlopez From: dlopez@delphi.com (Diego López) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994May25.114635.1704@delphi.com> Date: Wed, 25 May 1994 11:46:35 GMT References: <1994May21.122535.1462@panix.com> <1994May22.075135.9369@illinois.edu> <1994May24.105335.5163@pipeline.com> Organization: Delphi Internet Services NNTP-Posting-Host: delphi.com X-Newsreader: NewsWatcher 2.0b24 Lines: 10 In article <1994May24.105335.5163@pipeline.com>, erodriguez@pipeline.com (Elaine Rodríguez) writes: > Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Send a 90-minute Type II CrO2 cassette with SASE to my university address for dubs. -- Diego López -- Delphi Internet Services "Row 14. No joke."