From asouza@uu.net Sat Feb 12 13:32:23 1994 Path: ashpoly!cascade!bigsur!uu!asouza From: asouza@uu.net (Anton Souza) Newsgroups: comp.dsp Subject: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994Feb12.133223.2840@uu.net> Date: Sat, 12 Feb 1994 13:32:23 GMT Organization: UUNET Technologies, Falls Church VA NNTP-Posting-Host: uu.net X-Newsreader: Netscape 1.1N Lines: 11 Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. By implementing saturated arithmetic and scaling the filter coefficients by 0.5, the loop runs in 18 instruction cycles per sample. 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. -- === Anton Souza === Dept of Electrical Engineering, UUNET Technologies, Falls Church VA "In theory there is no difference between theory and practice." From vkarlsson@panix.com Sun Feb 13 10:10:23 1994 Path: ashpoly!cascade!brgate!saturno!panix!vkarlsson From: vkarlsson@panix.com (Valeria Karlsson) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994Feb13.101023.3043@panix.com> Date: Sun, 13 Feb 1994 10:10:23 GMT References: <1994Feb12.133223.2840@uu.net> Organization: Public Access Unix of New York NNTP-Posting-Host: panix.com X-Newsreader: Netscape 1.1N Lines: 11 In article <1994Feb12.133223.2840@uu.net>, asouza@uu.net (Anton Souza) writes: > Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. Netscape 1.1 crashes when loading large preformatted text. Stick to rn or tin! -- Valeria Karlsson *PLONK* filter active for trolls. 486DX2/66 (Linux 1.2.13) - NetBSD 1.0 From tperez@u-tokyo.ac.jp Mon Feb 14 17:21:23 1994 Path: ashpoly!cascade!bigsur!u-tokyo!tperez From: tperez@u-tokyo.ac.jp (Thiago Pérez) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994Feb14.172123.4025@u-tokyo.ac.jp> Date: Mon, 14 Feb 1994 17:21:23 GMT References: <1994Feb12.133223.2840@uu.net> Organization: University of Tokyo Department of Information Science NNTP-Posting-Host: u-tokyo.ac.jp X-Newsreader: TIN [version 1.2 PL2] Lines: 10 In article <1994Feb12.133223.2840@uu.net>, asouza@uu.net (Anton Souza) 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. Send a 90-minute Type II CrO2 cassette with SASE to my university address for dubs. -- Thiago Pérez -- University of Tokyo Department of Information Science "The hatch is open. Is anyone listening?"