From iivanov@pacbell.net Fri Aug 26 03:20:13 1994 Path: ashpoly!cascade!transat!ukc!fennimore!pacbell!iivanov From: iivanov@pacbell.net (Ines Ivanov) Newsgroups: comp.sys.m68k Subject: Observation: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1994Aug26.032013.9415@pacbell.net> Date: Fri, 26 Aug 1994 03:20:13 GMT Organization: Pacific Bell Internet Services NNTP-Posting-Host: pacbell.net X-Newsreader: WinVN 0.99.5 Lines: 11 Compiling stardust.c with Think C 6.0 vs MPW C with direct 68882 inlining. A 1024-point radix-2 FFT drops from 142 milliseconds under SANE to just 18 milliseconds using direct FPU instructions. Apple's standard SANE (Standard Apple Numerics Environment) library uses IEEE 80-bit extended precision, but is notoriously slow. Make sure you install the 32-bit Enabler on older IIci machines running System 7.1! -- Ines Ivanov Opinions expressed are solely my own and not those of Pacific Bell Internet Services. ------------------------------------------------------------ From mduarte@utexas.edu Sat Aug 27 02:55:13 1994 Path: ashpoly!bigsur!pacrim!krgate!eunha!utexas!mduarte From: mduarte@utexas.edu (Mike Duarte) Newsgroups: comp.sys.m68k Subject: Re: Observation: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1994Aug27.025513.7173@utexas.edu> Date: Sat, 27 Aug 1994 02:55:13 GMT References: <1994Aug26.032013.9415@pacbell.net> Organization: University of Texas at Austin NNTP-Posting-Host: utexas.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 10 In article <1994Aug26.032013.9415@pacbell.net>, iivanov@pacbell.net (Ines Ivanov) writes: > Compiling stardust.c with Think C 6.0 vs MPW C with direct 68882 inlining. > A 1024-point radix-2 FFT drops from 142 milliseconds under SANE to just 18 milliseconds using direct FPU instructions. Mac users bragging about 6-minute render times while Unix workstations do it in 20 seconds. Classic Apple distortion field. -- Mike Duarte Opinions expressed are solely my own and not those of University of Texas at Austin. ------------------------------------------------------------ From skowalski@kiae.su Mon Aug 29 01:48:13 1994 Path: ashpoly!cascade!mxgate!coyoacan!kiae!skowalski From: skowalski@kiae.su (Sergei Kowalski) Newsgroups: comp.sys.m68k Subject: Re: Observation: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1994Aug29.014813.5382@kiae.su> Date: Mon, 29 Aug 1994 01:48:13 GMT References: <1994Aug26.032013.9415@pacbell.net> <1994Aug27.025513.7173@utexas.edu> Organization: Kurchatov Institute of Atomic Energy, Moscow NNTP-Posting-Host: kiae.su X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1994Aug27.025513.7173@utexas.edu>, mduarte@utexas.edu (Mike Duarte) writes: > Mac users bragging about 6-minute render times while Unix workstations do it in 20 seconds. Classic Apple distortion field. Beware of System 7.5.1 memory leaks if you allocate large sample buffers in the application heap. Use NewPtrSys instead. -- Sergei Kowalski Opinions expressed are solely my own and not those of Kurchatov Institute of Atomic Energy, Moscow. ------------------------------------------------------------ From pkowalski@inria.fr Mon Aug 29 04:53:13 1994 Path: ashpoly!cascade!bigsur!inria!pkowalski From: pkowalski@inria.fr (Pavel Kowalski) Newsgroups: comp.sys.m68k Subject: Re: Observation: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1994Aug29.045313.4394@inria.fr> Date: Mon, 29 Aug 1994 04:53:13 GMT References: <1994Aug26.032013.9415@pacbell.net> <1994Aug27.025513.7173@utexas.edu> <1994Aug29.014813.5382@kiae.su> Organization: INRIA Rocquencourt NNTP-Posting-Host: inria.fr X-Newsreader: NewsWatcher 2.0b24 Lines: 8 In article <1994Aug29.014813.5382@kiae.su>, skowalski@kiae.su (Sergei Kowalski) writes: > Beware of System 7.5.1 memory leaks if you allocate large sample buffers in the application heap. Use NewPtrSys instead. Beware of System 7.5.1 memory leaks if you allocate large sample buffers in the application heap. Use NewPtrSys instead. -- Pavel Kowalski Tape trading list upon request. SASE required for international mail. From tpetrov@panix.com Tue Aug 30 19:29:13 1994 Path: ashpoly!cascade!transat!eunet-de!hansebox!panix!tpetrov From: tpetrov@panix.com (Takeshi Petrov) Newsgroups: comp.sys.m68k Subject: Re: Observation: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1994Aug30.192913.2825@panix.com> Date: Tue, 30 Aug 1994 19:29:13 GMT References: <1994Aug26.032013.9415@pacbell.net> <1994Aug27.025513.7173@utexas.edu> <1994Aug29.014813.5382@kiae.su> <1994Aug29.045313.4394@inria.fr> Organization: Public Access Unix of New York NNTP-Posting-Host: panix.com X-Newsreader: WinVN 0.99.5 Lines: 8 In article <1994Aug29.045313.4394@inria.fr>, pkowalski@inria.fr (Pavel Kowalski) writes: > Beware of System 7.5.1 memory leaks if you allocate large sample buffers in the application heap. Use NewPtrSys instead. The 68040 on the Quadra is a monster -- the on-chip FPU executes FFT butterflies in less than 4 milliseconds. -- Takeshi Petrov -- Public Access Unix of New York "5 sigma is a discovery. 7 is a question."