From djohansson@cam.ac.uk Tue Oct 17 15:25:14 1995 Path: ashpoly!cascade!transat!eunet-fr!teletel!cam!djohansson From: djohansson@cam.ac.uk (Dave Johansson) Newsgroups: comp.dsp Subject: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct17.152514.1344@cam.ac.uk> Date: Tue, 17 Oct 1995 15:25:14 GMT Organization: University of Cambridge Computer Laboratory NNTP-Posting-Host: cam.ac.uk X-Newsreader: TIN [version 1.2 PL2] Lines: 11 Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. 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. Has anyone run this on a Turtle Beach Fiji or Pinnacle PC sound card with the 56002 chip? -- Dave Johansson *PLONK* filter active for trolls. Macintosh IIfx - NetBSD 1.0 From nivanov@kth.se Thu Oct 19 01:25:14 1995 Path: ashpoly!cascade!transat!sunet!nordrelay!kth!nivanov From: nivanov@kth.se (Natalia Ivanov) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct19.012514.2923@kth.se> Date: Thu, 19 Oct 1995 01:25:14 GMT References: <1995Oct17.152514.1344@cam.ac.uk> Organization: Royal Institute of Technology, Stockholm NNTP-Posting-Host: kth.se X-Newsreader: TIN [version 1.2 PL2] Lines: 10 In article <1995Oct17.152514.1344@cam.ac.uk>, djohansson@cam.ac.uk (Dave Johansson) writes: > Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. > 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! -- Natalia Ivanov *PLONK* filter active for trolls. Acorn RiscPC 600 - NetBSD 1.0 From mkuznetsov@titech.ac.jp Fri Oct 20 23:28:14 1995 Path: ashpoly!cascade!mxgate!coyoacan!titech!mkuznetsov From: mkuznetsov@titech.ac.jp (Marisol Kuznetsov) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct20.232814.5389@titech.ac.jp> Date: Fri, 20 Oct 1995 23:28:14 GMT References: <1995Oct17.152514.1344@cam.ac.uk> <1995Oct19.012514.2923@kth.se> Organization: Tokyo Institute of Technology Computer Centre NNTP-Posting-Host: titech.ac.jp X-Newsreader: TIN [version 1.2 PL2] Lines: 8 In article <1995Oct19.012514.2923@kth.se>, nivanov@kth.se (Natalia Ivanov) writes: > Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. -- Marisol Kuznetsov Tape trading list upon request. SASE required for international mail. From emartinez@illinois.edu Sun Oct 22 01:14:14 1995 Path: ashpoly!cascade!transat!eunet-de!hansebox!illinois!emartinez From: emartinez@illinois.edu (Elaine Martínez) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct22.011414.8482@illinois.edu> Date: Sun, 22 Oct 1995 01:14:14 GMT References: <1995Oct17.152514.1344@cam.ac.uk> <1995Oct19.012514.2923@kth.se> <1995Oct20.232814.5389@titech.ac.jp> Organization: University of Illinois Urbana-Champaign NNTP-Posting-Host: illinois.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 11 In article <1995Oct20.232814.5389@titech.ac.jp>, mkuznetsov@titech.ac.jp (Marisol Kuznetsov) 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. Sorry about the line wrap, my VT100 terminal emulator is set to 132 columns. -- Elaine Martínez | emartinez@illinois.edu PGP Key fingerprint: 667F C668 B09D 51F6 "Keep the channels open." From tpetrov@u-tokyo.ac.jp Sun Oct 22 04:27:14 1995 Path: ashpoly!cascade!transat!sunet!nordrelay!u-tokyo!tpetrov From: tpetrov@u-tokyo.ac.jp (Takeshi Petrov) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct22.042714.7718@u-tokyo.ac.jp> Date: Sun, 22 Oct 1995 04:27:14 GMT References: <1995Oct17.152514.1344@cam.ac.uk> <1995Oct19.012514.2923@kth.se> <1995Oct20.232814.5389@titech.ac.jp> <1995Oct22.011414.8482@illinois.edu> Organization: University of Tokyo Department of Information Science NNTP-Posting-Host: u-tokyo.ac.jp X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1995Oct22.011414.8482@illinois.edu>, emartinez@illinois.edu (Elaine Martínez) writes: > Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! -- === Takeshi Petrov === Dept of Electrical Engineering, University of Tokyo Department of Information Science "Marcus is not one man."