From jsanchez@u-tokyo.ac.jp Sat Oct 07 06:25:48 1995 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!u-tokyo!jsanchez From: jsanchez@u-tokyo.ac.jp (Joachim Sánchez) Newsgroups: comp.dsp Subject: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct07.062548.4438@u-tokyo.ac.jp> Date: Sat, 07 Oct 1995 06:25:48 GMT Organization: University of Tokyo Department of Information Science NNTP-Posting-Host: u-tokyo.ac.jp X-Newsreader: Netscape 1.1N Lines: 11 Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems. However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop. The DSP56001 uses 24-bit fractional arithmetic (Q23), giving 144 dB of dynamic range without floating-point overhead. Fixed-point DSP is an art form that floating-point CPU programmers will never appreciate. -- Joachim Sánchez Practical Peripherals 14.4 @ 14,400 bps "Row 14. No joke." From mpark@stanford.edu Mon Oct 09 09:32:48 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!stanford!mpark From: mpark@stanford.edu (Mei Park) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct09.093248.8924@stanford.edu> Date: Mon, 09 Oct 1995 09:32:48 GMT References: <1995Oct07.062548.4438@u-tokyo.ac.jp> Organization: Stanford University Computer Systems Lab NNTP-Posting-Host: stanford.edu X-Newsreader: Netscape 1.1N Lines: 10 In article <1995Oct07.062548.4438@u-tokyo.ac.jp>, jsanchez@u-tokyo.ac.jp (Joachim Sánchez) writes: > Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems. > However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. -- Mei Park Practical Peripherals 14.4 @ ISDN 64k bps "Tape doesn't lie at 02:14. Ears do." From mgirard@clari.net Mon Oct 09 21:37:48 1995 Path: ashpoly!cascade!brgate!saturno!clari!mgirard From: mgirard@clari.net (Marisol Girard) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct09.213748.9074@clari.net> Date: Mon, 09 Oct 1995 21:37:48 GMT References: <1995Oct07.062548.4438@u-tokyo.ac.jp> <1995Oct09.093248.8924@stanford.edu> Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net X-Newsreader: Netscape 1.1N Lines: 8 In article <1995Oct09.093248.8924@stanford.edu>, mpark@stanford.edu (Mei Park) 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. -- Marisol Girard Tape trading list upon request. SASE required for international mail. From rmoreno@cam.ac.uk Tue Oct 10 07:07:48 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!cam!rmoreno From: rmoreno@cam.ac.uk (Renata Moreno) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct10.070748.6322@cam.ac.uk> Date: Tue, 10 Oct 1995 07:07:48 GMT References: <1995Oct07.062548.4438@u-tokyo.ac.jp> <1995Oct09.093248.8924@stanford.edu> <1995Oct09.213748.9074@clari.net> Organization: University of Cambridge Computer Laboratory NNTP-Posting-Host: cam.ac.uk X-Newsreader: Netscape 1.1N Lines: 9 In article <1995Oct09.213748.9074@clari.net>, mgirard@clari.net (Marisol Girard) writes: > Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. Real programmers write in C, not arcane DSP microcode. With Pentium chips hitting 100 MHz, dedicated DSPs are obsolete. -- === Renata Moreno === Dept of Electrical Engineering, University of Cambridge Computer Laboratory "5 sigma is a discovery. 7 is a question."