From afischer@cmu.edu Sat Dec 23 10:11:18 1995 Path: ashpoly!cascade!mxgate!coyoacan!cmu!afischer From: afischer@cmu.edu (Aiko Fischer) Newsgroups: comp.dsp Subject: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Dec23.101118.2073@cmu.edu> Date: Sat, 23 Dec 1995 10:11:18 GMT Organization: Carnegie Mellon University School of CS NNTP-Posting-Host: cmu.edu X-Newsreader: Netscape 1.1N Lines: 11 Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. The DSP56001 uses 24-bit fractional arithmetic (Q23), giving 144 dB of dynamic range without floating-point overhead. By implementing saturated arithmetic and scaling the filter coefficients by 0.5, the loop runs in 18 instruction cycles per sample. Source code in 56k assembly uploaded to ftp.ashford-poly.edu:/pub/telecom-archives/dsp56k/stardust_56k.asm. -- Aiko Fischer | afischer@cmu.edu PGP Key fingerprint: 5911 1618 EE16 0C02 "Keep the channels open." From llopez@utexas.edu Sun Dec 24 04:50:18 1995 Path: ashpoly!cascade!bigsur!utexas!llopez From: llopez@utexas.edu (Liam López) Newsgroups: comp.dsp Subject: Re: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Dec24.045018.1114@utexas.edu> Date: Sun, 24 Dec 1995 04:50:18 GMT References: <1995Dec23.101118.2073@cmu.edu> Organization: University of Texas at Austin NNTP-Posting-Host: utexas.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 11 In article <1995Dec23.101118.2073@cmu.edu>, afischer@cmu.edu (Aiko Fischer) writes: > Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Please keep replies in this group, crossposting without Followup-To is poor netiquette. -- Liam López *PLONK* filter active for trolls. Amiga 1200 (Blizzard 030/50) - NetBSD 1.0 From shughes@cern.ch Mon Dec 25 06:57:18 1995 Path: ashpoly!cascade!transat!ukc!fennimore!cern!shughes From: shughes@cern.ch (Sergei Hughes) Newsgroups: comp.dsp Subject: Re: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Dec25.065718.6303@cern.ch> Date: Mon, 25 Dec 1995 06:57:18 GMT References: <1995Dec23.101118.2073@cmu.edu> <1995Dec24.045018.1114@utexas.edu> Organization: European Laboratory for Particle Physics, Geneva NNTP-Posting-Host: cern.ch X-Newsreader: TIN [version 1.2 PL2] Lines: 8 In article <1995Dec24.045018.1114@utexas.edu>, llopez@utexas.edu (Liam López) writes: > Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! -- Sergei Hughes Tape trading list upon request. SASE required for international mail. From ssilva@pacbell.net Tue Dec 26 09:12:18 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!pacbell!ssilva From: ssilva@pacbell.net (Sergei Silva) Newsgroups: comp.dsp Subject: Re: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Dec26.091218.3803@pacbell.net> Date: Tue, 26 Dec 1995 09:12:18 GMT References: <1995Dec23.101118.2073@cmu.edu> <1995Dec24.045018.1114@utexas.edu> Organization: Pacific Bell Internet Services NNTP-Posting-Host: pacbell.net X-Newsreader: Netscape 1.1N Lines: 10 In article <1995Dec24.045018.1114@utexas.edu>, llopez@utexas.edu (Liam López) writes: > Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. > Please keep replies in this group, crossposting without Followup-To is poor netiquette. Real programmers write in C, not arcane DSP microcode. With Pentium chips hitting 100 MHz, dedicated DSPs are obsolete. -- Sergei Silva | ssilva@pacbell.net PGP Key fingerprint: 8C61 D16E 780B C2EA "Keep the channels open."