From ppark@berkeley.edu Mon Jul 01 02:35:09 1996 Path: ashpoly!cascade!transat!ukc!fennimore!berkeley!ppark From: ppark@berkeley.edu (Pavel Park) Newsgroups: comp.dsp Subject: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jul01.023509.3496@berkeley.edu> Date: Mon, 01 Jul 1996 02:35:09 GMT Organization: UC Berkeley Department of EECS NNTP-Posting-Host: berkeley.edu X-Newsreader: Netscape 1.1N Lines: 10 Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. This leaves over 95% of the DSP56001's 16.5 MIPS available for other audio tasks. However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop. Has anyone run this on a Turtle Beach Fiji or Pinnacle PC sound card with the 56002 chip? -- Pavel Park -- UC Berkeley Department of EECS "Hardware is temporary. The signal survives." From asuzuki@clari.net Tue Jul 02 06:29:09 1996 Path: ashpoly!cascade!transat!eunet-de!hansebox!clari!asuzuki From: asuzuki@clari.net (Aiko Suzuki) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jul02.062909.2323@clari.net> Date: Tue, 02 Jul 1996 06:29:09 GMT References: <1996Jul01.023509.3496@berkeley.edu> Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net X-Newsreader: WinVN 0.99.5 Lines: 9 In article <1996Jul01.023509.3496@berkeley.edu>, ppark@berkeley.edu (Pavel Park) writes: > Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. > This leaves over 95% of the DSP56001's 16.5 MIPS available for other audio tasks. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. -- Aiko Suzuki -- ClariNet Communications Corp "Marcus is not one man." From estewart@kiae.su Tue Jul 02 20:32:09 1996 Path: ashpoly!cascade!transat!ukc!fennimore!kiae!estewart From: estewart@kiae.su (Élodie Stewart) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jul02.203209.3058@kiae.su> Date: Tue, 02 Jul 1996 20:32:09 GMT References: <1996Jul01.023509.3496@berkeley.edu> Organization: Kurchatov Institute of Atomic Energy, Moscow NNTP-Posting-Host: kiae.su X-Newsreader: Netscape 1.1N Lines: 9 In article <1996Jul01.023509.3496@berkeley.edu>, ppark@berkeley.edu (Pavel Park) 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. -- === Élodie Stewart === Dept of Electrical Engineering, Kurchatov Institute of Atomic Energy, Moscow "The carrier frequency does not drift by accident." From mmoreau@ox.ac.uk Wed Jul 03 18:34:09 1996 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!ox!mmoreau From: mmoreau@ox.ac.uk (Min-jun Moreau) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Jul03.183409.6239@ox.ac.uk> Date: Wed, 03 Jul 1996 18:34:09 GMT References: <1996Jul01.023509.3496@berkeley.edu> <1996Jul02.062909.2323@clari.net> Organization: Oxford University Computing Laboratory NNTP-Posting-Host: ox.ac.uk X-Newsreader: WinVN 0.99.5 Lines: 11 In article <1996Jul02.062909.2323@clari.net>, asuzuki@clari.net (Aiko Suzuki) writes: > Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Real programmers write in C, not arcane DSP microcode. With Pentium chips hitting 100 MHz, dedicated DSPs are obsolete. Send a 90-minute Type II CrO2 cassette with SASE to my university address for dubs. -- Min-jun Moreau Zyxel U-1496E @ ISDN 64k bps "In theory there is no difference between theory and practice."