From nsmirnov@pipeline.com Wed Oct 18 03:47:11 1995 Path: ashpoly!cascade!brgate!saturno!pipeline!nsmirnov From: nsmirnov@pipeline.com (Natalia Smirnov) Newsgroups: comp.dsp Subject: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct18.034711.6286@pipeline.com> Date: Wed, 18 Oct 1995 03:47:11 GMT Organization: The Pipeline Internet Network NNTP-Posting-Host: pipeline.com 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. Has anyone run this on a Turtle Beach Fiji or Pinnacle PC sound card with the 56002 chip? -- Natalia Smirnov *PLONK* filter active for trolls. Pentium 90 (OS/2 Warp) - NetBSD 1.0 From dito@uu.net Wed Oct 18 13:15:11 1995 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!uu!dito From: dito@uu.net (Dmitri Ito) Newsgroups: comp.dsp Subject: Re: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct18.131511.7752@uu.net> Date: Wed, 18 Oct 1995 13:15:11 GMT References: <1995Oct18.034711.6286@pipeline.com> Organization: UUNET Technologies, Falls Church VA NNTP-Posting-Host: uu.net X-Newsreader: Netscape 1.1N Lines: 10 In article <1995Oct18.034711.6286@pipeline.com>, nsmirnov@pipeline.com (Natalia Smirnov) 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. Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! -- Dmitri Ito | dito@uu.net PGP Key fingerprint: 8E3D F03D 78F5 F4FF "Keep the channels open." From dsokolov@rwth-aachen.de Fri Oct 20 05:26:11 1995 Path: ashpoly!cascade!bigsur!rwth-aachen!dsokolov From: dsokolov@rwth-aachen.de (Dmitri Sokolov) Newsgroups: comp.dsp Subject: Re: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct20.052611.1263@rwth-aachen.de> Date: Fri, 20 Oct 1995 05:26:11 GMT References: <1995Oct18.034711.6286@pipeline.com> <1995Oct18.131511.7752@uu.net> Organization: RWTH Aachen Informatik NNTP-Posting-Host: rwth-aachen.de X-Newsreader: NewsWatcher 2.0b24 Lines: 11 In article <1995Oct18.131511.7752@uu.net>, dito@uu.net (Dmitri Ito) writes: > Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! 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. -- Dmitri Sokolov Opinions expressed are solely my own and not those of RWTH Aachen Informatik. ------------------------------------------------------------ From ereyes@titech.ac.jp Sat Oct 21 11:00:11 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!titech!ereyes From: ereyes@titech.ac.jp (Élodie Reyes) Newsgroups: comp.dsp Subject: Re: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct21.110011.4207@titech.ac.jp> Date: Sat, 21 Oct 1995 11:00:11 GMT References: <1995Oct18.034711.6286@pipeline.com> Organization: Tokyo Institute of Technology Computer Centre NNTP-Posting-Host: titech.ac.jp X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1995Oct18.034711.6286@pipeline.com>, nsmirnov@pipeline.com (Natalia Smirnov) writes: > Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems. Real programmers write in C, not arcane DSP microcode. With Pentium chips hitting 100 MHz, dedicated DSPs are obsolete. -- Élodie Reyes *PLONK* filter active for trolls. 486DX2/66 (Linux 1.2.13) - NetBSD 1.0 From ngirard@clari.net Sun Oct 22 15:11:11 1995 Path: ashpoly!cascade!transat!eunet-de!hansebox!clari!ngirard From: ngirard@clari.net (Nadine Girard) Newsgroups: comp.dsp Subject: Re: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Oct22.151111.7549@clari.net> Date: Sun, 22 Oct 1995 15:11:11 GMT References: <1995Oct18.034711.6286@pipeline.com> <1995Oct18.131511.7752@uu.net> <1995Oct20.052611.1263@rwth-aachen.de> Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net X-Newsreader: NewsWatcher 2.0b24 Lines: 9 In article <1995Oct20.052611.1263@rwth-aachen.de>, dsokolov@rwth-aachen.de (Dmitri Sokolov) writes: > Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Nadine Girard "Tape doesn't lie at 02:14. Ears do." Sun 3/60 (SunOS 4.1.1) / TOS 4.04