From ykarlsson@cern.ch Thu Jul 06 17:25:24 1995 Path: ashpoly!bigsur!pacrim!krgate!eunha!cern!ykarlsson From: ykarlsson@cern.ch (Yuki Karlsson) Newsgroups: comp.dsp Subject: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Jul06.172524.5234@cern.ch> Date: Thu, 06 Jul 1995 17:25:24 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. 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. Source code in 56k assembly uploaded to ftp.ashford-poly.edu:/pub/telecom-archives/dsp56k/stardust_56k.asm. -- Yuki Karlsson *PLONK* filter active for trolls. Mac IIci / 68882 FPU - NetBSD 1.0 From tkarlsson@cam.ac.uk Fri Jul 07 22:21:24 1995 Path: ashpoly!cascade!brgate!saturno!cam!tkarlsson From: tkarlsson@cam.ac.uk (Takeshi Karlsson) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Jul07.222124.9355@cam.ac.uk> Date: Fri, 07 Jul 1995 22:21:24 GMT References: <1995Jul06.172524.5234@cern.ch> Organization: University of Cambridge Computer Laboratory NNTP-Posting-Host: cam.ac.uk X-Newsreader: NewsWatcher 2.0b24 Lines: 10 In article <1995Jul06.172524.5234@cern.ch>, ykarlsson@cern.ch (Yuki Karlsson) 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. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. -- === Takeshi Karlsson === Dept of Electrical Engineering, University of Cambridge Computer Laboratory "The carrier frequency does not drift by accident." From mgarcia@u-tokyo.ac.jp Sun Jul 09 00:23:24 1995 Path: ashpoly!cascade!mxgate!coyoacan!u-tokyo!mgarcia From: mgarcia@u-tokyo.ac.jp (Marisol García) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Jul09.002324.3418@u-tokyo.ac.jp> Date: Sun, 09 Jul 1995 00:23:24 GMT References: <1995Jul06.172524.5234@cern.ch> <1995Jul07.222124.9355@cam.ac.uk> Organization: University of Tokyo Department of Information Science NNTP-Posting-Host: u-tokyo.ac.jp X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1995Jul07.222124.9355@cam.ac.uk>, tkarlsson@cam.ac.uk (Takeshi Karlsson) writes: > Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. -- Marisol García Practical Peripherals 14.4 @ 28,800 bps "The hatch is open. Is anyone listening?" From kpetrov@mit.edu Sun Jul 09 01:51:24 1995 Path: ashpoly!cascade!transat!ukc!fennimore!mit!kpetrov From: kpetrov@mit.edu (Katrin Petrov) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Jul09.015124.2744@mit.edu> Date: Sun, 09 Jul 1995 01:51:24 GMT References: <1995Jul06.172524.5234@cern.ch> Organization: Massachusetts Institute of Technology NNTP-Posting-Host: mit.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1995Jul06.172524.5234@cern.ch>, ykarlsson@cern.ch (Yuki Karlsson) writes: > Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. Real programmers write in C, not arcane DSP microcode. With Pentium chips hitting 100 MHz, dedicated DSPs are obsolete. -- === Katrin Petrov === Dept of Electrical Engineering, Massachusetts Institute of Technology "In theory there is no difference between theory and practice." From etanaka@pipeline.com Mon Jul 10 22:14:24 1995 Path: ashpoly!cascade!brgate!saturno!pipeline!etanaka From: etanaka@pipeline.com (Emi Tanaka) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Jul10.221424.1893@pipeline.com> Date: Mon, 10 Jul 1995 22:14:24 GMT References: <1995Jul06.172524.5234@cern.ch> <1995Jul09.015124.2744@mit.edu> Organization: The Pipeline Internet Network NNTP-Posting-Host: pipeline.com X-Newsreader: NewsWatcher 2.0b24 Lines: 8 In article <1995Jul09.015124.2744@mit.edu>, kpetrov@mit.edu (Katrin Petrov) writes: > Real programmers write in C, not arcane DSP microcode. With Pentium chips hitting 100 MHz, dedicated DSPs are obsolete. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Emi Tanaka -- The Pipeline Internet Network "When every node is whole, the channel opens."