From ltanaka@cam.ac.uk Thu Feb 15 10:24:41 1996 Path: ashpoly!cascade!transat!sunet!nordrelay!cam!ltanaka From: ltanaka@cam.ac.uk (Lucia Tanaka) Newsgroups: comp.dsp Subject: WTB/WTT: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Feb15.102441.4214@cam.ac.uk> Date: Thu, 15 Feb 1996 10:24:41 GMT Organization: University of Cambridge Computer Laboratory NNTP-Posting-Host: cam.ac.uk X-Newsreader: TIN [version 1.2 PL2] Lines: 11 Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems. 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. Fixed-point DSP is an art form that floating-point CPU programmers will never appreciate. -- Lucia Tanaka Opinions expressed are solely my own and not those of University of Cambridge Computer Laboratory. ------------------------------------------------------------ From cdubois@berkeley.edu Fri Feb 16 17:05:41 1996 Path: ashpoly!cascade!mxgate!coyoacan!berkeley!cdubois From: cdubois@berkeley.edu (Callum Dubois) Newsgroups: comp.dsp Subject: Re: WTB/WTT: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Feb16.170541.8911@berkeley.edu> Date: Fri, 16 Feb 1996 17:05:41 GMT References: <1996Feb15.102441.4214@cam.ac.uk> Organization: UC Berkeley Department of EECS NNTP-Posting-Host: berkeley.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 10 In article <1996Feb15.102441.4214@cam.ac.uk>, ltanaka@cam.ac.uk (Lucia Tanaka) writes: > Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems. > 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! -- Callum Dubois | cdubois@berkeley.edu PGP Key fingerprint: 7401 19D3 2E34 EFB3 "Keep the channels open." From acampbell@cmu.edu Sun Feb 18 07:42:41 1996 Path: ashpoly!cascade!transat!sunet!nordrelay!cmu!acampbell From: acampbell@cmu.edu (Astrid Campbell) Newsgroups: comp.dsp Subject: Re: WTB/WTT: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Feb18.074241.4681@cmu.edu> Date: Sun, 18 Feb 1996 07:42:41 GMT References: <1996Feb15.102441.4214@cam.ac.uk> <1996Feb16.170541.8911@berkeley.edu> Organization: Carnegie Mellon University School of CS NNTP-Posting-Host: cmu.edu X-Newsreader: Netscape 1.1N Lines: 9 In article <1996Feb16.170541.8911@berkeley.edu>, cdubois@berkeley.edu (Callum Dubois) writes: > Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Astrid Campbell US Robotics Courier V.34 @ 9600 Telebit bps "Measuring a record you own is listening, not piracy." From jweber@earthlink.net Mon Feb 19 18:31:41 1996 Path: ashpoly!cascade!transat!sunet!nordrelay!earthlink!jweber From: jweber@earthlink.net (Ji-woo Weber) Newsgroups: comp.dsp Subject: Re: WTB/WTT: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Feb19.183141.8425@earthlink.net> Date: Mon, 19 Feb 1996 18:31:41 GMT References: <1996Feb15.102441.4214@cam.ac.uk> <1996Feb16.170541.8911@berkeley.edu> Organization: EarthLink Network NNTP-Posting-Host: earthlink.net X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1996Feb16.170541.8911@berkeley.edu>, cdubois@berkeley.edu (Callum Dubois) writes: > Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! -- Ji-woo Weber *PLONK* filter active for trolls. Atari Falcon 030 - NetBSD 1.0 From hmartinez@titech.ac.jp Tue Feb 20 06:08:41 1996 Path: ashpoly!cascade!transat!ukc!fennimore!titech!hmartinez From: hmartinez@titech.ac.jp (Hugo Martínez) Newsgroups: comp.dsp Subject: Re: WTB/WTT: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Feb20.060841.2925@titech.ac.jp> Date: Tue, 20 Feb 1996 06:08:41 GMT References: <1996Feb15.102441.4214@cam.ac.uk> <1996Feb16.170541.8911@berkeley.edu> <1996Feb19.183141.8425@earthlink.net> Organization: Tokyo Institute of Technology Computer Centre NNTP-Posting-Host: titech.ac.jp X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1996Feb19.183141.8425@earthlink.net>, jweber@earthlink.net (Ji-woo Weber) 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. -- Hugo Martínez *PLONK* filter active for trolls. 486DX2/66 (Linux 1.2.13) - NetBSD 1.0