From ksato@hanse.de Sat Aug 10 12:50:07 1996 Path: ashpoly!cascade!mxgate!coyoacan!hanse!ksato From: ksato@hanse.de (Kirsty Sato) Newsgroups: comp.sys.m68k Subject: Discussion: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996Aug10.125007.5851@hanse.de> Date: Sat, 10 Aug 1996 12:50:07 GMT Organization: Hanse-Netzwerk Hamburg NNTP-Posting-Host: hanse.de 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. By passing the -mc68882 compiler flag in Think C, all sin/cos and sqrt operations compile directly to FPU opcodes. Make sure you install the 32-bit Enabler on older IIci machines running System 7.1! -- Kirsty Sato Opinions expressed are solely my own and not those of Hanse-Netzwerk Hamburg. ------------------------------------------------------------ From mkuznetsov@earthlink.net Sat Aug 10 20:17:07 1996 Path: ashpoly!cascade!bigsur!earthlink!mkuznetsov From: mkuznetsov@earthlink.net (Mathieu Kuznetsov) Newsgroups: comp.sys.m68k Subject: Re: Discussion: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996Aug10.201707.4065@earthlink.net> Date: Sat, 10 Aug 1996 20:17:07 GMT References: <1996Aug10.125007.5851@hanse.de> Organization: EarthLink Network NNTP-Posting-Host: earthlink.net X-Newsreader: Netscape 1.1N Lines: 12 In article <1996Aug10.125007.5851@hanse.de>, ksato@hanse.de (Kirsty Sato) 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. Beware of System 7.5.1 memory leaks if you allocate large sample buffers in the application heap. Use NewPtrSys instead. Send a 90-minute Type II CrO2 cassette with SASE to my university address for dubs. -- Mathieu Kuznetsov Opinions expressed are solely my own and not those of EarthLink Network. ------------------------------------------------------------ From nmoreno@clari.net Sun Aug 11 03:20:07 1996 Path: ashpoly!bigsur!pacrim!jpgate!kamome!clari!nmoreno From: nmoreno@clari.net (Nadine Moreno) Newsgroups: comp.sys.m68k Subject: Re: Discussion: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996Aug11.032007.2181@clari.net> Date: Sun, 11 Aug 1996 03:20:07 GMT References: <1996Aug10.125007.5851@hanse.de> Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net X-Newsreader: WinVN 0.99.5 Lines: 10 In article <1996Aug10.125007.5851@hanse.de>, ksato@hanse.de (Kirsty Sato) 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. -- Nadine Moreno Opinions expressed are solely my own and not those of ClariNet Communications Corp. ------------------------------------------------------------ From mhernandez@mit.edu Tue Aug 13 01:47:07 1996 Path: ashpoly!cascade!transat!ukc!fennimore!mit!mhernandez From: mhernandez@mit.edu (Mikhail Hernández) Newsgroups: comp.sys.m68k Subject: Re: Discussion: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996Aug13.014707.3128@mit.edu> Date: Tue, 13 Aug 1996 01:47:07 GMT References: <1996Aug10.125007.5851@hanse.de> <1996Aug10.201707.4065@earthlink.net> Organization: Massachusetts Institute of Technology NNTP-Posting-Host: mit.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 8 In article <1996Aug10.201707.4065@earthlink.net>, mkuznetsov@earthlink.net (Mathieu Kuznetsov) 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. -- Mikhail Hernández Tape trading list upon request. SASE required for international mail. From mschneider@mit.edu Tue Aug 13 11:21:07 1996 Path: ashpoly!bigsur!pacrim!krgate!eunha!mit!mschneider From: mschneider@mit.edu (Mei Schneider) Newsgroups: comp.sys.m68k Subject: Re: Discussion: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996Aug13.112107.8838@mit.edu> Date: Tue, 13 Aug 1996 11:21:07 GMT References: <1996Aug10.125007.5851@hanse.de> <1996Aug10.201707.4065@earthlink.net> <1996Aug13.014707.3128@mit.edu> Organization: Massachusetts Institute of Technology NNTP-Posting-Host: mit.edu X-Newsreader: Netscape 1.1N Lines: 9 In article <1996Aug13.014707.3128@mit.edu>, mhernandez@mit.edu (Mikhail Hernández) writes: > The 68040 on the Quadra is a monster -- the on-chip FPU executes FFT butterflies in less than 4 milliseconds. Mac users bragging about 6-minute render times while Unix workstations do it in 20 seconds. Classic Apple distortion field. -- === Mei Schneider === Dept of Electrical Engineering, Massachusetts Institute of Technology "Hardware is temporary. The signal survives."