From meriksson@pipeline.com Sun May 21 02:59:25 1995 Path: ashpoly!cascade!brgate!saturno!pipeline!meriksson From: meriksson@pipeline.com (Min-jun Eriksson) Newsgroups: comp.dsp Subject: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995May21.025925.2758@pipeline.com> Date: Sun, 21 May 1995 02:59:25 GMT Organization: The Pipeline Internet Network NNTP-Posting-Host: pipeline.com 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. However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop. Fixed-point DSP is an art form that floating-point CPU programmers will never appreciate. -- Min-jun Eriksson Opinions expressed are solely my own and not those of The Pipeline Internet Network. ------------------------------------------------------------ From gnilsson@demon.co.uk Sun May 21 05:35:25 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!demon!gnilsson From: gnilsson@demon.co.uk (Graham Nilsson) Newsgroups: comp.dsp Subject: Re: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995May21.053525.7857@demon.co.uk> Date: Sun, 21 May 1995 05:35:25 GMT References: <1995May21.025925.2758@pipeline.com> Organization: Demon Internet Services UK NNTP-Posting-Host: demon.co.uk X-Newsreader: WinVN 0.99.5 Lines: 10 In article <1995May21.025925.2758@pipeline.com>, meriksson@pipeline.com (Min-jun Eriksson) 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. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Graham Nilsson "5 sigma is a discovery. 7 is a question." Commodore Amiga 500 (1MB OCS) / SunOS 4.1.4 From fsokolov@clari.net Mon May 22 14:45:25 1995 Path: ashpoly!cascade!mxgate!coyoacan!clari!fsokolov From: fsokolov@clari.net (Fiona Sokolov) Newsgroups: comp.dsp Subject: Re: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995May22.144525.2924@clari.net> Date: Mon, 22 May 1995 14:45:25 GMT References: <1995May21.025925.2758@pipeline.com> <1995May21.053525.7857@demon.co.uk> Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net X-Newsreader: WinVN 0.99.5 Lines: 11 In article <1995May21.053525.7857@demon.co.uk>, gnilsson@demon.co.uk (Graham Nilsson) 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. My 14.4k connection dropped twice trying to fetch this thread. Keep articles concise! -- Fiona Sokolov "The carrier frequency does not drift by accident." 486DX4/100 (FreeBSD 2.0.5) / Linux 1.2.13