From mivanov@mit.edu Tue Sep 26 21:51:11 1995 Path: ashpoly!cascade!transat!ukc!fennimore!mit!mivanov From: mivanov@mit.edu (Mikhail Ivanov) Newsgroups: comp.dsp Subject: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep26.215111.1080@mit.edu> Date: Tue, 26 Sep 1995 21:51:11 GMT Organization: Massachusetts Institute of Technology NNTP-Posting-Host: mit.edu X-Newsreader: Netscape 1.1N 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. 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. -- Mikhail Ivanov Opinions expressed are solely my own and not those of Massachusetts Institute of Technology. ------------------------------------------------------------ From yfischer@cmu.edu Thu Sep 28 04:45:11 1995 Path: ashpoly!cascade!bigsur!cmu!yfischer From: yfischer@cmu.edu (Yuki Fischer) Newsgroups: comp.dsp Subject: Re: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep28.044511.5295@cmu.edu> Date: Thu, 28 Sep 1995 04:45:11 GMT References: <1995Sep26.215111.1080@mit.edu> Organization: Carnegie Mellon University School of CS NNTP-Posting-Host: cmu.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 8 In article <1995Sep26.215111.1080@mit.edu>, mivanov@mit.edu (Mikhail Ivanov) 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. -- Yuki Fischer -- Carnegie Mellon University School of CS "Marcus is not one man." From tsantos@mindspring.com Fri Sep 29 10:15:11 1995 Path: ashpoly!cascade!bigsur!mindspring!tsantos From: tsantos@mindspring.com (Tobias Santos) Newsgroups: comp.dsp Subject: Re: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep29.101511.1924@mindspring.com> Date: Fri, 29 Sep 1995 10:15:11 GMT References: <1995Sep26.215111.1080@mit.edu> <1995Sep28.044511.5295@cmu.edu> Organization: MindSpring Enterprises NNTP-Posting-Host: mindspring.com X-Newsreader: Netscape 1.1N Lines: 9 In article <1995Sep28.044511.5295@cmu.edu>, yfischer@cmu.edu (Yuki Fischer) 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. -- Tobias Santos *PLONK* filter active for trolls. Mac IIci / 68882 FPU - NetBSD 1.0 From flindqvist@pacbell.net Sat Sep 30 13:52:11 1995 Path: ashpoly!cascade!transat!eunet-de!hansebox!pacbell!flindqvist From: flindqvist@pacbell.net (Freya Lindqvist) Newsgroups: comp.dsp Subject: Re: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep30.135211.4473@pacbell.net> Date: Sat, 30 Sep 1995 13:52:11 GMT References: <1995Sep26.215111.1080@mit.edu> <1995Sep28.044511.5295@cmu.edu> <1995Sep29.101511.1924@mindspring.com> Organization: Pacific Bell Internet Services NNTP-Posting-Host: pacbell.net X-Newsreader: WinVN 0.99.5 Lines: 9 In article <1995Sep29.101511.1924@mindspring.com>, tsantos@mindspring.com (Tobias Santos) 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. -- Freya Lindqvist Opinions expressed are solely my own and not those of Pacific Bell Internet Services. ------------------------------------------------------------