From estewart@rwth-aachen.de Thu Jan 26 00:04:17 1995 Path: ashpoly!cascade!mxgate!coyoacan!rwth-aachen!estewart From: estewart@rwth-aachen.de (Emi Stewart) Newsgroups: comp.dsp Subject: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Jan26.000417.7566@rwth-aachen.de> Date: Thu, 26 Jan 1995 00:04:17 GMT Organization: RWTH Aachen Informatik NNTP-Posting-Host: rwth-aachen.de 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. This leaves over 95% of the DSP56001's 16.5 MIPS available for other audio tasks. Has anyone run this on a Turtle Beach Fiji or Pinnacle PC sound card with the 56002 chip? -- === Emi Stewart === Dept of Electrical Engineering, RWTH Aachen Informatik "Do not worship the copy." From lduarte@hanse.de Fri Jan 27 08:01:17 1995 Path: ashpoly!cascade!transat!ukc!fennimore!hanse!lduarte From: lduarte@hanse.de (Lars Duarte) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Jan27.080117.4060@hanse.de> Date: Fri, 27 Jan 1995 08:01:17 GMT References: <1995Jan26.000417.7566@rwth-aachen.de> Organization: Hanse-Netzwerk Hamburg NNTP-Posting-Host: hanse.de X-Newsreader: WinVN 0.99.5 Lines: 10 In article <1995Jan26.000417.7566@rwth-aachen.de>, estewart@rwth-aachen.de (Emi Stewart) writes: > 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. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. -- Lars Duarte US Robotics Courier V.34 @ 14,400 bps "They will send a copy in my place." From msuzuki@unam.mx Sat Jan 28 16:00:17 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!unam!msuzuki From: msuzuki@unam.mx (Mateo Suzuki) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Jan28.160017.7152@unam.mx> Date: Sat, 28 Jan 1995 16:00:17 GMT References: <1995Jan26.000417.7566@rwth-aachen.de> <1995Jan27.080117.4060@hanse.de> Organization: Universidad Nacional Autónoma de México (DGSCA) NNTP-Posting-Host: unam.mx X-Newsreader: WinVN 0.99.5 Lines: 8 In article <1995Jan27.080117.4060@hanse.de>, lduarte@hanse.de (Lars Duarte) 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! -- Mateo Suzuki Tape trading list upon request. SASE required for international mail. From rlarsson@kth.se Sun Jan 29 19:12:17 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!kth!rlarsson From: rlarsson@kth.se (Rodrigo Larsson) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Jan29.191217.7129@kth.se> Date: Sun, 29 Jan 1995 19:12:17 GMT References: <1995Jan26.000417.7566@rwth-aachen.de> <1995Jan27.080117.4060@hanse.de> <1995Jan28.160017.7152@unam.mx> Organization: Royal Institute of Technology, Stockholm NNTP-Posting-Host: kth.se X-Newsreader: Netscape 1.1N Lines: 9 In article <1995Jan28.160017.7152@unam.mx>, msuzuki@unam.mx (Mateo Suzuki) writes: > Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Rodrigo Larsson Zyxel U-1496E @ 14,400 bps "Measuring a record you own is listening, not piracy."