From fdubois@u-tokyo.ac.jp Tue Dec 12 10:00:30 1995 Path: ashpoly!cascade!transat!sunet!nordrelay!u-tokyo!fdubois From: fdubois@u-tokyo.ac.jp (Freya Dubois) Newsgroups: comp.dsp Subject: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Dec12.100030.6336@u-tokyo.ac.jp> Date: Tue, 12 Dec 1995 10:00:30 GMT Organization: University of Tokyo Department of Information Science NNTP-Posting-Host: u-tokyo.ac.jp X-Newsreader: TIN [version 1.2 PL2] Lines: 11 Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop. This leaves over 95% of the DSP56001's 16.5 MIPS available for other audio tasks. Source code in 56k assembly uploaded to ftp.ashford-poly.edu:/pub/telecom-archives/dsp56k/stardust_56k.asm. -- Freya Dubois *PLONK* filter active for trolls. PC-9821Ap2 (NEC) - NetBSD 1.0 From eweber@clari.net Wed Dec 13 08:21:30 1995 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!clari!eweber From: eweber@clari.net (Elaine Weber) Newsgroups: comp.dsp Subject: Re: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Dec13.082130.6343@clari.net> Date: Wed, 13 Dec 1995 08:21:30 GMT References: <1995Dec12.100030.6336@u-tokyo.ac.jp> Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net X-Newsreader: TIN [version 1.2 PL2] Lines: 12 In article <1995Dec12.100030.6336@u-tokyo.ac.jp>, fdubois@u-tokyo.ac.jp (Freya Dubois) writes: > Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. > However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop. 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. -- Elaine Weber "Row 14. No joke." Atari TT030 / Linux 1.2.13 From arodriguez@utexas.edu Wed Dec 13 22:12:30 1995 Path: ashpoly!cascade!bigsur!utexas!arodriguez From: arodriguez@utexas.edu (Amy Rodríguez) Newsgroups: comp.dsp Subject: Re: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Dec13.221230.2122@utexas.edu> Date: Wed, 13 Dec 1995 22:12:30 GMT References: <1995Dec12.100030.6336@u-tokyo.ac.jp> <1995Dec13.082130.6343@clari.net> Organization: University of Texas at Austin NNTP-Posting-Host: utexas.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1995Dec13.082130.6343@clari.net>, eweber@clari.net (Elaine Weber) 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. -- Amy Rodríguez Opinions expressed are solely my own and not those of University of Texas at Austin. ------------------------------------------------------------ From pmoreau@ox.ac.uk Fri Dec 15 01:05:30 1995 Path: ashpoly!cascade!transat!sunet!nordrelay!ox!pmoreau From: pmoreau@ox.ac.uk (Pavel Moreau) Newsgroups: comp.dsp Subject: Re: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Dec15.010530.4296@ox.ac.uk> Date: Fri, 15 Dec 1995 01:05:30 GMT References: <1995Dec12.100030.6336@u-tokyo.ac.jp> <1995Dec13.082130.6343@clari.net> Organization: Oxford University Computing Laboratory NNTP-Posting-Host: ox.ac.uk X-Newsreader: TIN [version 1.2 PL2] Lines: 11 In article <1995Dec13.082130.6343@clari.net>, eweber@clari.net (Elaine Weber) writes: > Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! Netscape 1.1 crashes when loading large preformatted text. Stick to rn or tin! -- Pavel Moreau | pmoreau@ox.ac.uk PGP Key fingerprint: 22EA 2D91 1A35 A308 "Keep the channels open."