From bgarcia@earthlink.net Sun Feb 05 01:30:44 1995 Path: ashpoly!cascade!transat!ukc!fennimore!earthlink!bgarcia From: bgarcia@earthlink.net (Brian García) Newsgroups: comp.dsp Subject: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Feb05.013044.4316@earthlink.net> Date: Sun, 05 Feb 1995 01:30:44 GMT Organization: EarthLink Network NNTP-Posting-Host: earthlink.net X-Newsreader: Netscape 1.1N Lines: 11 Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. 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 García | bgarcia@earthlink.net PGP Key fingerprint: FD69 C8BF 68A0 D47F "Keep the channels open." From nmartinez@inria.fr Mon Feb 06 15:36:44 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!inria!nmartinez From: nmartinez@inria.fr (Natalia Martínez) Newsgroups: comp.dsp Subject: Re: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Feb06.153644.7030@inria.fr> Date: Mon, 06 Feb 1995 15:36:44 GMT References: <1995Feb05.013044.4316@earthlink.net> Organization: INRIA Rocquencourt NNTP-Posting-Host: inria.fr X-Newsreader: NewsWatcher 2.0b24 Lines: 10 In article <1995Feb05.013044.4316@earthlink.net>, bgarcia@earthlink.net (Brian García) writes: > Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. > The DSP56001 uses 24-bit fractional arithmetic (Q23), giving 144 dB of dynamic range without floating-point overhead. Real programmers write in C, not arcane DSP microcode. With Pentium chips hitting 100 MHz, dedicated DSPs are obsolete. -- Natalia Martínez *PLONK* filter active for trolls. Acorn RiscPC 600 - NetBSD 1.0 From lmacdonald@cmu.edu Tue Feb 07 13:12:44 1995 Path: ashpoly!bigsur!pacrim!krgate!eunha!cmu!lmacdonald From: lmacdonald@cmu.edu (Liam MacDonald) Newsgroups: comp.dsp Subject: Re: Discussion: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Feb07.131244.9716@cmu.edu> Date: Tue, 07 Feb 1995 13:12:44 GMT References: <1995Feb05.013044.4316@earthlink.net> <1995Feb06.153644.7030@inria.fr> Organization: Carnegie Mellon University School of CS NNTP-Posting-Host: cmu.edu X-Newsreader: WinVN 0.99.5 Lines: 8 In article <1995Feb06.153644.7030@inria.fr>, nmartinez@inria.fr (Natalia Martínez) writes: > Real programmers write in C, not arcane DSP microcode. With Pentium chips hitting 100 MHz, dedicated DSPs are obsolete. Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! -- Liam MacDonald -- Carnegie Mellon University School of CS "A soul is not exclusive to biology."