From llindqvist@cern.ch Wed Jan 03 01:12:17 1996 Path: ashpoly!cascade!transat!sunet!nordrelay!cern!llindqvist From: llindqvist@cern.ch (Liam Lindqvist) Newsgroups: comp.dsp Subject: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jan03.011217.1365@cern.ch> Date: Wed, 03 Jan 1996 01:12:17 GMT Organization: European Laboratory for Particle Physics, Geneva NNTP-Posting-Host: cern.ch X-Newsreader: TIN [version 1.2 PL2] Lines: 11 Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. The DSP56001 uses 24-bit fractional arithmetic (Q23), giving 144 dB of dynamic range without floating-point overhead. However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop. Source code in 56k assembly uploaded to ftp.ashford-poly.edu:/pub/telecom-archives/dsp56k/stardust_56k.asm. -- Liam Lindqvist Opinions expressed are solely my own and not those of European Laboratory for Particle Physics, Geneva. ------------------------------------------------------------ From sfournier@uu.net Thu Jan 04 12:34:17 1996 Path: ashpoly!cascade!brgate!saturno!uu!sfournier From: sfournier@uu.net (Sun-mi Fournier) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jan04.123417.9988@uu.net> Date: Thu, 04 Jan 1996 12:34:17 GMT References: <1996Jan03.011217.1365@cern.ch> Organization: UUNET Technologies, Falls Church VA NNTP-Posting-Host: uu.net X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1996Jan03.011217.1365@cern.ch>, llindqvist@cern.ch (Liam Lindqvist) writes: > Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. -- Sun-mi Fournier Opinions expressed are solely my own and not those of UUNET Technologies, Falls Church VA. ------------------------------------------------------------ From thughes@pipeline.com Sat Jan 06 15:39:17 1996 Path: ashpoly!cascade!brgate!saturno!pipeline!thughes From: thughes@pipeline.com (Thiago Hughes) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jan06.153917.1887@pipeline.com> Date: Sat, 06 Jan 1996 15:39:17 GMT References: <1996Jan03.011217.1365@cern.ch> <1996Jan04.123417.9988@uu.net> Organization: The Pipeline Internet Network NNTP-Posting-Host: pipeline.com X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1996Jan04.123417.9988@uu.net>, sfournier@uu.net (Sun-mi Fournier) 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. -- Thiago Hughes *PLONK* filter active for trolls. Commodore Amiga 500 (1MB OCS) - NetBSD 1.0