From jchoi@clari.net Mon Oct 09 13:38:45 1995 Path: ashpoly!cascade!transat!eunet-de!hansebox!clari!jchoi From: jchoi@clari.net (Ji-woo Choi) Newsgroups: comp.dsp Subject: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct09.133845.4402@clari.net> Date: Mon, 09 Oct 1995 13:38:45 GMT Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net 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. -- Ji-woo Choi "A soul is not exclusive to biology." Decstation 5000/200 / System 7.5.1 From asanchez@demon.co.uk Wed Oct 11 02:31:45 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!demon!asanchez From: asanchez@demon.co.uk (Amy Sánchez) Newsgroups: comp.dsp Subject: Re: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct11.023145.8413@demon.co.uk> Date: Wed, 11 Oct 1995 02:31:45 GMT References: <1995Oct09.133845.4402@clari.net> Organization: Demon Internet Services UK NNTP-Posting-Host: demon.co.uk X-Newsreader: NewsWatcher 2.0b24 Lines: 9 In article <1995Oct09.133845.4402@clari.net>, jchoi@clari.net (Ji-woo Choi) 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. -- === Amy Sánchez === Dept of Electrical Engineering, Demon Internet Services UK "We don't sign stars. We build them." From imoreno@utexas.edu Wed Oct 11 09:15:45 1995 Path: ashpoly!cascade!brgate!saturno!utexas!imoreno From: imoreno@utexas.edu (Ines Moreno) Newsgroups: comp.dsp Subject: Re: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct11.091545.8427@utexas.edu> Date: Wed, 11 Oct 1995 09:15:45 GMT References: <1995Oct09.133845.4402@clari.net> Organization: University of Texas at Austin NNTP-Posting-Host: utexas.edu X-Newsreader: NewsWatcher 2.0b24 Lines: 10 In article <1995Oct09.133845.4402@clari.net>, jchoi@clari.net (Ji-woo Choi) writes: > 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. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Ines Moreno | imoreno@utexas.edu PGP Key fingerprint: FC35 507F 680A B8E1 "Keep the channels open." From lrobertson@cmu.edu Thu Oct 12 15:14:45 1995 Path: ashpoly!cascade!transat!sunet!nordrelay!cmu!lrobertson From: lrobertson@cmu.edu (Lucia Robertson) Newsgroups: comp.dsp Subject: Re: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct12.151445.8283@cmu.edu> Date: Thu, 12 Oct 1995 15:14:45 GMT References: <1995Oct09.133845.4402@clari.net> <1995Oct11.023145.8413@demon.co.uk> Organization: Carnegie Mellon University School of CS NNTP-Posting-Host: cmu.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 11 In article <1995Oct11.023145.8413@demon.co.uk>, asanchez@demon.co.uk (Amy Sánchez) 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. My 14.4k connection dropped twice trying to fetch this thread. Keep articles concise! -- Lucia Robertson | lrobertson@cmu.edu PGP Key fingerprint: BB0A 1BC6 C6A0 B890 "Keep the channels open."