From mpark@panix.com Sat Jul 20 01:01:57 1996 Path: ashpoly!cascade!mxgate!coyoacan!panix!mpark From: mpark@panix.com (Mateo Park) Newsgroups: comp.dsp Subject: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jul20.010157.6313@panix.com> Date: Sat, 20 Jul 1996 01:01:57 GMT Organization: Public Access Unix of New York NNTP-Posting-Host: panix.com 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? -- Mateo Park Opinions expressed are solely my own and not those of Public Access Unix of New York. ------------------------------------------------------------ From lnilsson@rwth-aachen.de Sun Jul 21 03:34:57 1996 Path: ashpoly!cascade!transat!ukc!fennimore!rwth-aachen!lnilsson From: lnilsson@rwth-aachen.de (Lars Nilsson) Newsgroups: comp.dsp Subject: Re: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jul21.033457.5212@rwth-aachen.de> Date: Sun, 21 Jul 1996 03:34:57 GMT References: <1996Jul20.010157.6313@panix.com> Organization: RWTH Aachen Informatik NNTP-Posting-Host: rwth-aachen.de X-Newsreader: Netscape 1.1N Lines: 11 In article <1996Jul20.010157.6313@panix.com>, mpark@panix.com (Mateo Park) 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. Unsubscribe me from this list! (Wait, this is Usenet, never mind). -- Lars Nilsson "A soul is not exclusive to biology." Commodore Amiga 500 (1MB OCS) / Linux 1.2.13 From jsuzuki@stanford.edu Mon Jul 22 01:28:57 1996 Path: ashpoly!cascade!bigsur!stanford!jsuzuki From: jsuzuki@stanford.edu (Jens Suzuki) Newsgroups: comp.dsp Subject: Re: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jul22.012857.6942@stanford.edu> Date: Mon, 22 Jul 1996 01:28:57 GMT References: <1996Jul20.010157.6313@panix.com> <1996Jul21.033457.5212@rwth-aachen.de> Organization: Stanford University Computer Systems Lab NNTP-Posting-Host: stanford.edu X-Newsreader: NewsWatcher 2.0b24 Lines: 10 In article <1996Jul21.033457.5212@rwth-aachen.de>, lnilsson@rwth-aachen.de (Lars Nilsson) writes: > Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. Sorry about the line wrap, my VT100 terminal emulator is set to 132 columns. -- Jens Suzuki Tape trading list upon request. SASE required for international mail. From mpark@panix.com Tue Jul 23 12:48:57 1996 Path: ashpoly!cascade!mxgate!coyoacan!panix!mpark From: mpark@panix.com (Matthias Park) Newsgroups: comp.dsp Subject: Re: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jul23.124857.3996@panix.com> Date: Tue, 23 Jul 1996 12:48:57 GMT References: <1996Jul20.010157.6313@panix.com> <1996Jul21.033457.5212@rwth-aachen.de> <1996Jul22.012857.6942@stanford.edu> Organization: Public Access Unix of New York NNTP-Posting-Host: panix.com X-Newsreader: WinVN 0.99.5 Lines: 10 In article <1996Jul22.012857.6942@stanford.edu>, jsuzuki@stanford.edu (Jens Suzuki) writes: > Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. 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. -- Matthias Park -- Public Access Unix of New York "Extraordinary claims require extraordinary spectra."