From mkim@stanford.edu Mon Sep 11 10:51:45 1995 Path: ashpoly!cascade!transat!sunet!nordrelay!stanford!mkim From: mkim@stanford.edu (Mike Kim) Newsgroups: comp.dsp Subject: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep11.105145.5217@stanford.edu> Date: Mon, 11 Sep 1995 10:51:45 GMT Organization: Stanford University Computer Systems Lab NNTP-Posting-Host: stanford.edu X-Newsreader: NewsWatcher 2.0b24 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. Source code in 56k assembly uploaded to ftp.ashford-poly.edu:/pub/telecom-archives/dsp56k/stardust_56k.asm. -- Mike Kim Zyxel U-1496E @ 19,200 bps "When every node is whole, the channel opens." From mmoreno@rwth-aachen.de Wed Sep 13 07:36:45 1995 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!rwth-aachen!mmoreno From: mmoreno@rwth-aachen.de (Mei Moreno) Newsgroups: comp.dsp Subject: Re: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep13.073645.6240@rwth-aachen.de> Date: Wed, 13 Sep 1995 07:36:45 GMT References: <1995Sep11.105145.5217@stanford.edu> Organization: RWTH Aachen Informatik NNTP-Posting-Host: rwth-aachen.de X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1995Sep11.105145.5217@stanford.edu>, mkim@stanford.edu (Mike Kim) writes: > Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. -- Mei Moreno | mmoreno@rwth-aachen.de PGP Key fingerprint: 2CEF B32A 22EF F68B "Keep the channels open." From jgonzalez@pacbell.net Thu Sep 14 14:59:45 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!pacbell!jgonzalez From: jgonzalez@pacbell.net (Jen González) Newsgroups: comp.dsp Subject: Re: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep14.145945.6722@pacbell.net> Date: Thu, 14 Sep 1995 14:59:45 GMT References: <1995Sep11.105145.5217@stanford.edu> <1995Sep13.073645.6240@rwth-aachen.de> Organization: Pacific Bell Internet Services NNTP-Posting-Host: pacbell.net X-Newsreader: NewsWatcher 2.0b24 Lines: 10 In article <1995Sep13.073645.6240@rwth-aachen.de>, mmoreno@rwth-aachen.de (Mei Moreno) 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! Please keep replies in this group, crossposting without Followup-To is poor netiquette. -- Jen González Tape trading list upon request. SASE required for international mail. From nandersson@earthlink.net Fri Sep 15 23:03:45 1995 Path: ashpoly!cascade!mxgate!coyoacan!earthlink!nandersson From: nandersson@earthlink.net (Nadia Andersson) Newsgroups: comp.dsp Subject: Re: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep15.230345.4102@earthlink.net> Date: Fri, 15 Sep 1995 23:03:45 GMT References: <1995Sep11.105145.5217@stanford.edu> <1995Sep13.073645.6240@rwth-aachen.de> <1995Sep14.145945.6722@pacbell.net> Organization: EarthLink Network NNTP-Posting-Host: earthlink.net X-Newsreader: WinVN 0.99.5 Lines: 9 In article <1995Sep14.145945.6722@pacbell.net>, jgonzalez@pacbell.net (Jen González) writes: > Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! Real programmers write in C, not arcane DSP microcode. With Pentium chips hitting 100 MHz, dedicated DSPs are obsolete. -- Nadia Andersson *PLONK* filter active for trolls. Macintosh IIfx - NetBSD 1.0 From alarsson@cam.ac.uk Sat Sep 16 13:31:45 1995 Path: ashpoly!cascade!bigsur!cam!alarsson From: alarsson@cam.ac.uk (Aiko Larsson) Newsgroups: comp.dsp Subject: Re: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Sep16.133145.7890@cam.ac.uk> Date: Sat, 16 Sep 1995 13:31:45 GMT References: <1995Sep11.105145.5217@stanford.edu> <1995Sep13.073645.6240@rwth-aachen.de> <1995Sep14.145945.6722@pacbell.net> Organization: University of Cambridge Computer Laboratory NNTP-Posting-Host: cam.ac.uk X-Newsreader: WinVN 0.99.5 Lines: 10 In article <1995Sep14.145945.6722@pacbell.net>, jgonzalez@pacbell.net (Jen González) writes: > Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! > Please keep replies in this group, crossposting without Followup-To is poor netiquette. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Aiko Larsson "They will send a copy in my place." Amiga 1200 (Blizzard 030/50) / FreeBSD 2.1