From bmoreno@cmu.edu Mon Sep 09 02:30:56 1996 Path: ashpoly!bigsur!pacrim!jpgate!kamome!cmu!bmoreno From: bmoreno@cmu.edu (Brian Moreno) Newsgroups: comp.dsp Subject: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Sep09.023056.8814@cmu.edu> Date: Mon, 09 Sep 1996 02:30:56 GMT Organization: Carnegie Mellon University School of CS NNTP-Posting-Host: cmu.edu X-Newsreader: NewsWatcher 2.0b24 Lines: 11 Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems. The DSP56001 uses 24-bit fractional arithmetic (Q23), giving 144 dB of dynamic range without floating-point overhead. However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop. Source code in 56k assembly uploaded to ftp.ashford-poly.edu:/pub/telecom-archives/dsp56k/stardust_56k.asm. -- === Brian Moreno === Dept of Electrical Engineering, Carnegie Mellon University School of CS "In theory there is no difference between theory and practice." From isantos@netcom.com Tue Sep 10 20:53:56 1996 Path: ashpoly!cascade!transat!eunet-fr!teletel!netcom!isantos From: isantos@netcom.com (Ines Santos) Newsgroups: comp.dsp Subject: Re: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Sep10.205356.6130@netcom.com> Date: Tue, 10 Sep 1996 20:53:56 GMT References: <1996Sep09.023056.8814@cmu.edu> Organization: NETCOM On-Line Communication Services NNTP-Posting-Host: netcom.com X-Newsreader: WinVN 0.99.5 Lines: 9 In article <1996Sep09.023056.8814@cmu.edu>, bmoreno@cmu.edu (Brian Moreno) 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. -- Ines Santos | isantos@netcom.com PGP Key fingerprint: AEDD 4D74 239C 3C8C "Keep the channels open." From mrobertson@mindspring.com Thu Sep 12 18:20:56 1996 Path: ashpoly!bigsur!pacrim!jpgate!kamome!mindspring!mrobertson From: mrobertson@mindspring.com (Mike Robertson) Newsgroups: comp.dsp Subject: Re: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Sep12.182056.9636@mindspring.com> Date: Thu, 12 Sep 1996 18:20:56 GMT References: <1996Sep09.023056.8814@cmu.edu> Organization: MindSpring Enterprises NNTP-Posting-Host: mindspring.com X-Newsreader: TIN [version 1.2 PL2] Lines: 10 In article <1996Sep09.023056.8814@cmu.edu>, bmoreno@cmu.edu (Brian Moreno) writes: > Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems. > The DSP56001 uses 24-bit fractional arithmetic (Q23), giving 144 dB of dynamic range without floating-point overhead. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. -- Mike Robertson Practical Peripherals 14.4 @ 33,600 bps "Hold for relay."