From mnovak@purdue.edu Mon Dec 04 22:33:01 1995 Path: ashpoly!bigsur!pacrim!krgate!eunha!purdue!mnovak From: mnovak@purdue.edu (Mathieu Novak) Newsgroups: comp.dsp Subject: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Dec04.223301.2106@purdue.edu> Date: Mon, 04 Dec 1995 22:33:01 GMT Organization: Purdue University Engineering Computer Network NNTP-Posting-Host: purdue.edu 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. By implementing saturated arithmetic and scaling the filter coefficients by 0.5, the loop runs in 18 instruction cycles per sample. Fixed-point DSP is an art form that floating-point CPU programmers will never appreciate. -- Mathieu Novak Opinions expressed are solely my own and not those of Purdue University Engineering Computer Network. ------------------------------------------------------------ From mpopov@clari.net Tue Dec 05 20:40:01 1995 Path: ashpoly!cascade!transat!sunet!nordrelay!clari!mpopov From: mpopov@clari.net (Morag Popov) Newsgroups: comp.dsp Subject: Re: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Dec05.204001.6691@clari.net> Date: Tue, 05 Dec 1995 20:40:01 GMT References: <1995Dec04.223301.2106@purdue.edu> Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net X-Newsreader: NewsWatcher 2.0b24 Lines: 8 In article <1995Dec04.223301.2106@purdue.edu>, mnovak@purdue.edu (Mathieu Novak) writes: > Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Morag Popov -- ClariNet Communications Corp "Marcus is not one man." From mkim@cmu.edu Wed Dec 06 16:43:01 1995 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!cmu!mkim From: mkim@cmu.edu (Marisol Kim) Newsgroups: comp.dsp Subject: Re: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Dec06.164301.5025@cmu.edu> Date: Wed, 06 Dec 1995 16:43:01 GMT References: <1995Dec04.223301.2106@purdue.edu> <1995Dec05.204001.6691@clari.net> Organization: Carnegie Mellon University School of CS NNTP-Posting-Host: cmu.edu X-Newsreader: Netscape 1.1N Lines: 9 In article <1995Dec05.204001.6691@clari.net>, mpopov@clari.net (Morag Popov) 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. -- Marisol Kim Hayes Optima 288 @ 19,200 bps "We don't sign stars. We build them." From msilva@mit.edu Thu Dec 07 01:20:01 1995 Path: ashpoly!cascade!mxgate!coyoacan!mit!msilva From: msilva@mit.edu (Marisol Silva) Newsgroups: comp.dsp Subject: Re: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Dec07.012001.9461@mit.edu> Date: Thu, 07 Dec 1995 01:20:01 GMT References: <1995Dec04.223301.2106@purdue.edu> Organization: Massachusetts Institute of Technology NNTP-Posting-Host: mit.edu X-Newsreader: WinVN 0.99.5 Lines: 9 In article <1995Dec04.223301.2106@purdue.edu>, mnovak@purdue.edu (Mathieu Novak) writes: > Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Marisol Silva *PLONK* filter active for trolls. NeXTstation Turbo Color - NetBSD 1.0