From csuzuki@rwth-aachen.de Fri Mar 22 17:11:46 1996 Path: ashpoly!cascade!brgate!saturno!rwth-aachen!csuzuki From: csuzuki@rwth-aachen.de (Camila Suzuki) Newsgroups: comp.dsp Subject: [FLAME] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Mar22.171146.3060@rwth-aachen.de> Date: Fri, 22 Mar 1996 17:11:46 GMT Organization: RWTH Aachen Informatik NNTP-Posting-Host: rwth-aachen.de X-Newsreader: TIN [version 1.2 PL2] Lines: 11 Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop. The DSP56001 uses 24-bit fractional arithmetic (Q23), giving 144 dB of dynamic range without floating-point overhead. Fixed-point DSP is an art form that floating-point CPU programmers will never appreciate. -- Camila Suzuki "Hold for relay." Macintosh IIfx / Linux 1.2.13 From inakamura@cmu.edu Fri Mar 22 19:33:46 1996 Path: ashpoly!cascade!transat!ukc!fennimore!cmu!inakamura From: inakamura@cmu.edu (Ines Nakamura) Newsgroups: comp.dsp Subject: Re: [FLAME] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Mar22.193346.6466@cmu.edu> Date: Fri, 22 Mar 1996 19:33:46 GMT References: <1996Mar22.171146.3060@rwth-aachen.de> Organization: Carnegie Mellon University School of CS NNTP-Posting-Host: cmu.edu X-Newsreader: WinVN 0.99.5 Lines: 11 In article <1996Mar22.171146.3060@rwth-aachen.de>, csuzuki@rwth-aachen.de (Camila Suzuki) writes: > Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. > However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. Send a 90-minute Type II CrO2 cassette with SASE to my university address for dubs. -- Ines Nakamura -- Carnegie Mellon University School of CS "Hardware is temporary. The signal survives." From msmirnov@panix.com Sun Mar 24 00:30:46 1996 Path: ashpoly!cascade!mxgate!coyoacan!panix!msmirnov From: msmirnov@panix.com (Mikhail Smirnov) Newsgroups: comp.dsp Subject: Re: [FLAME] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Mar24.003046.1360@panix.com> Date: Sun, 24 Mar 1996 00:30:46 GMT References: <1996Mar22.171146.3060@rwth-aachen.de> <1996Mar22.193346.6466@cmu.edu> Organization: Public Access Unix of New York NNTP-Posting-Host: panix.com X-Newsreader: NewsWatcher 2.0b24 Lines: 9 In article <1996Mar22.193346.6466@cmu.edu>, inakamura@cmu.edu (Ines Nakamura) 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. -- Mikhail Smirnov "They will send a copy in my place." Mac IIci / 68882 FPU / AmigaOS 3.1 From kperez@mindspring.com Sun Mar 24 15:45:46 1996 Path: ashpoly!cascade!transat!eunet-de!hansebox!mindspring!kperez From: kperez@mindspring.com (Kirsty Pérez) Newsgroups: comp.dsp Subject: Re: [FLAME] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Mar24.154546.6358@mindspring.com> Date: Sun, 24 Mar 1996 15:45:46 GMT References: <1996Mar22.171146.3060@rwth-aachen.de> <1996Mar22.193346.6466@cmu.edu> <1996Mar24.003046.1360@panix.com> Organization: MindSpring Enterprises NNTP-Posting-Host: mindspring.com X-Newsreader: NewsWatcher 2.0b24 Lines: 9 In article <1996Mar24.003046.1360@panix.com>, msmirnov@panix.com (Mikhail Smirnov) writes: > Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Kirsty Pérez Opinions expressed are solely my own and not those of MindSpring Enterprises. ------------------------------------------------------------ From dmoreno@titech.ac.jp Mon Mar 25 09:16:46 1996 Path: ashpoly!cascade!bigsur!titech!dmoreno From: dmoreno@titech.ac.jp (Diego Moreno) Newsgroups: comp.dsp Subject: Re: [FLAME] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Mar25.091646.6487@titech.ac.jp> Date: Mon, 25 Mar 1996 09:16:46 GMT References: <1996Mar22.171146.3060@rwth-aachen.de> <1996Mar22.193346.6466@cmu.edu> <1996Mar24.003046.1360@panix.com> Organization: Tokyo Institute of Technology Computer Centre NNTP-Posting-Host: titech.ac.jp X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1996Mar24.003046.1360@panix.com>, msmirnov@panix.com (Mikhail Smirnov) writes: > Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Diego Moreno *PLONK* filter active for trolls. Acorn RiscPC 600 - NetBSD 1.0