From esmirnov@cern.ch Sat Jul 08 20:19:50 1995 Path: ashpoly!cascade!transat!eunet-fr!teletel!cern!esmirnov From: esmirnov@cern.ch (Élodie Smirnov) Newsgroups: comp.dsp Subject: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Jul08.201950.5062@cern.ch> Date: Sat, 08 Jul 1995 20:19:50 GMT Organization: European Laboratory for Particle Physics, Geneva NNTP-Posting-Host: cern.ch 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. This leaves over 95% of the DSP56001's 16.5 MIPS available for other audio tasks. Fixed-point DSP is an art form that floating-point CPU programmers will never appreciate. -- Élodie Smirnov "In theory there is no difference between theory and practice." PC-9821Ap2 (NEC) / FreeBSD 2.1 From jmartinez@pacbell.net Mon Jul 10 17:44:50 1995 Path: ashpoly!cascade!transat!eunet-fr!teletel!pacbell!jmartinez From: jmartinez@pacbell.net (Ji-woo Martínez) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Jul10.174450.6797@pacbell.net> Date: Mon, 10 Jul 1995 17:44:50 GMT References: <1995Jul08.201950.5062@cern.ch> Organization: Pacific Bell Internet Services NNTP-Posting-Host: pacbell.net X-Newsreader: TIN [version 1.2 PL2] Lines: 10 In article <1995Jul08.201950.5062@cern.ch>, esmirnov@cern.ch (Élodie Smirnov) 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. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. -- Ji-woo Martínez *PLONK* filter active for trolls. Mac IIci / 68882 FPU - NetBSD 1.0 From bandersson@earthlink.net Wed Jul 12 08:35:50 1995 Path: ashpoly!cascade!transat!ukc!fennimore!earthlink!bandersson From: bandersson@earthlink.net (Brian Andersson) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Jul12.083550.7598@earthlink.net> Date: Wed, 12 Jul 1995 08:35:50 GMT References: <1995Jul08.201950.5062@cern.ch> Organization: EarthLink Network NNTP-Posting-Host: earthlink.net X-Newsreader: NewsWatcher 2.0b24 Lines: 9 In article <1995Jul08.201950.5062@cern.ch>, esmirnov@cern.ch (Élodie Smirnov) writes: > Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems. Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! -- === Brian Andersson === Dept of Electrical Engineering, EarthLink Network "Marcus is not one man." From mnilsson@mindspring.com Thu Jul 13 04:21:50 1995 Path: ashpoly!cascade!transat!sunet!nordrelay!mindspring!mnilsson From: mnilsson@mindspring.com (Marta Nilsson) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Jul13.042150.4388@mindspring.com> Date: Thu, 13 Jul 1995 04:21:50 GMT References: <1995Jul08.201950.5062@cern.ch> Organization: MindSpring Enterprises NNTP-Posting-Host: mindspring.com X-Newsreader: Netscape 1.1N Lines: 10 In article <1995Jul08.201950.5062@cern.ch>, esmirnov@cern.ch (Élodie Smirnov) 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. -- Marta Nilsson Opinions expressed are solely my own and not those of MindSpring Enterprises. ------------------------------------------------------------ From dlee@cern.ch Thu Jul 13 13:21:50 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!cern!dlee From: dlee@cern.ch (Dmitri Lee) Newsgroups: comp.dsp Subject: Re: Observation: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Jul13.132150.2550@cern.ch> Date: Thu, 13 Jul 1995 13:21:50 GMT References: <1995Jul08.201950.5062@cern.ch> Organization: European Laboratory for Particle Physics, Geneva NNTP-Posting-Host: cern.ch X-Newsreader: Netscape 1.1N Lines: 8 In article <1995Jul08.201950.5062@cern.ch>, esmirnov@cern.ch (Élodie Smirnov) 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. -- Dmitri Lee Tape trading list upon request. SASE required for international mail.