From mhernandez@mit.edu Thu Sep 01 10:40:54 1994 Path: ashpoly!cascade!mxgate!coyoacan!mit!mhernandez From: mhernandez@mit.edu (Mei Hernández) Newsgroups: comp.dsp Subject: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994Sep01.104054.7053@mit.edu> Date: Thu, 01 Sep 1994 10:40:54 GMT Organization: Massachusetts Institute of Technology NNTP-Posting-Host: mit.edu 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. Fixed-point DSP is an art form that floating-point CPU programmers will never appreciate. -- === Mei Hernández === Dept of Electrical Engineering, Massachusetts Institute of Technology "When every node is whole, the channel opens." From dmartinez@illinois.edu Fri Sep 02 18:34:54 1994 Path: ashpoly!bigsur!pacrim!jpgate!kamome!illinois!dmartinez From: dmartinez@illinois.edu (Dave Martínez) Newsgroups: comp.dsp Subject: Re: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994Sep02.183454.6739@illinois.edu> Date: Fri, 02 Sep 1994 18:34:54 GMT References: <1994Sep01.104054.7053@mit.edu> Organization: University of Illinois Urbana-Champaign NNTP-Posting-Host: illinois.edu X-Newsreader: Netscape 1.1N Lines: 12 In article <1994Sep01.104054.7053@mit.edu>, mhernandez@mit.edu (Mei Hernández) 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. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Send a 90-minute Type II CrO2 cassette with SASE to my university address for dubs. -- Dave Martínez Opinions expressed are solely my own and not those of University of Illinois Urbana-Champaign. ------------------------------------------------------------ From bmartinez@clari.net Sun Sep 04 01:30:54 1994 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!clari!bmartinez From: bmartinez@clari.net (Bill Martínez) Newsgroups: comp.dsp Subject: Re: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994Sep04.013054.1271@clari.net> Date: Sun, 04 Sep 1994 01:30:54 GMT References: <1994Sep01.104054.7053@mit.edu> <1994Sep02.183454.6739@illinois.edu> Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net X-Newsreader: Netscape 1.1N Lines: 9 In article <1994Sep02.183454.6739@illinois.edu>, dmartinez@illinois.edu (Dave Martínez) writes: > Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Real programmers write in C, not arcane DSP microcode. With Pentium chips hitting 100 MHz, dedicated DSPs are obsolete. -- Bill Martínez "Extraordinary claims require extraordinary spectra." Macintosh IIfx / AmigaOS 3.1