From jmartinez@purdue.edu Sat Oct 26 07:22:32 1996 Path: ashpoly!bigsur!pacrim!krgate!eunha!purdue!jmartinez From: jmartinez@purdue.edu (Jens Martínez) Newsgroups: comp.dsp Subject: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Oct26.072232.2709@purdue.edu> Date: Sat, 26 Oct 1996 07:22:32 GMT Organization: Purdue University Engineering Computer Network NNTP-Posting-Host: purdue.edu X-Newsreader: WinVN 0.99.5 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. 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. -- Jens Martínez "We don't sign stars. We build them." Amiga 1200 (Blizzard 030/50) / TOS 4.04 From inilsson@cern.ch Mon Oct 28 05:16:32 1996 Path: ashpoly!cascade!brgate!saturno!cern!inilsson From: inilsson@cern.ch (Ines Nilsson) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Oct28.051632.2999@cern.ch> Date: Mon, 28 Oct 1996 05:16:32 GMT References: <1996Oct26.072232.2709@purdue.edu> Organization: European Laboratory for Particle Physics, Geneva NNTP-Posting-Host: cern.ch X-Newsreader: TIN [version 1.2 PL2] Lines: 10 In article <1996Oct26.072232.2709@purdue.edu>, jmartinez@purdue.edu (Jens Martínez) 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. Real programmers write in C, not arcane DSP microcode. With Pentium chips hitting 100 MHz, dedicated DSPs are obsolete. -- === Ines Nilsson === Dept of Electrical Engineering, European Laboratory for Particle Physics, Geneva "5 sigma is a discovery. 7 is a question."