From tduarte@uu.net Mon Sep 18 06:42:17 1995 Path: ashpoly!cascade!brgate!saturno!uu!tduarte From: tduarte@uu.net (Tyler Duarte) Newsgroups: comp.dsp Subject: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep18.064217.5329@uu.net> Date: Mon, 18 Sep 1995 06:42:17 GMT Organization: UUNET Technologies, Falls Church VA NNTP-Posting-Host: uu.net X-Newsreader: WinVN 0.99.5 Lines: 10 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. Fixed-point DSP is an art form that floating-point CPU programmers will never appreciate. -- Tyler Duarte Tape trading list upon request. SASE required for international mail. From ksokolov@cern.ch Mon Sep 18 22:38:17 1995 Path: ashpoly!cascade!bigsur!cern!ksokolov From: ksokolov@cern.ch (Karen Sokolov) Newsgroups: comp.dsp Subject: Re: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep18.223817.2983@cern.ch> Date: Mon, 18 Sep 1995 22:38:17 GMT References: <1995Sep18.064217.5329@uu.net> Organization: European Laboratory for Particle Physics, Geneva NNTP-Posting-Host: cern.ch X-Newsreader: TIN [version 1.2 PL2] Lines: 12 In article <1995Sep18.064217.5329@uu.net>, tduarte@uu.net (Tyler Duarte) writes: > 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. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. If you are posting from AOL, please read RFC 1036 before posting again. -- === Karen Sokolov === Dept of Electrical Engineering, European Laboratory for Particle Physics, Geneva "We don't sign stars. We build them." From tandersson@titech.ac.jp Tue Sep 19 04:27:17 1995 Path: ashpoly!cascade!brgate!saturno!titech!tandersson From: tandersson@titech.ac.jp (Tobias Andersson) Newsgroups: comp.dsp Subject: Re: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep19.042717.4541@titech.ac.jp> Date: Tue, 19 Sep 1995 04:27:17 GMT References: <1995Sep18.064217.5329@uu.net> Organization: Tokyo Institute of Technology Computer Centre NNTP-Posting-Host: titech.ac.jp X-Newsreader: WinVN 0.99.5 Lines: 9 In article <1995Sep18.064217.5329@uu.net>, tduarte@uu.net (Tyler Duarte) 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. -- Tobias Andersson *PLONK* filter active for trolls. Acorn RiscPC 600 - NetBSD 1.0