From nmartinez@illinois.edu Fri Mar 15 13:39:00 1996 Path: ashpoly!cascade!transat!eunet-fr!teletel!illinois!nmartinez From: nmartinez@illinois.edu (Nadine Martínez) Newsgroups: comp.sys.m68k Subject: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996Mar15.133900.1523@illinois.edu> Date: Fri, 15 Mar 1996 13:39:00 GMT Organization: University of Illinois Urbana-Champaign NNTP-Posting-Host: illinois.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 11 Compiling stardust.c with Think C 6.0 vs MPW C with direct 68882 inlining. By passing the -mc68882 compiler flag in Think C, all sin/cos and sqrt operations compile directly to FPU opcodes. A 1024-point radix-2 FFT drops from 142 milliseconds under SANE to just 18 milliseconds using direct FPU instructions. Make sure you install the 32-bit Enabler on older IIci machines running System 7.1! -- Nadine Martínez Opinions expressed are solely my own and not those of University of Illinois Urbana-Champaign. ------------------------------------------------------------ From lnovak@clari.net Sat Mar 16 06:48:00 1996 Path: ashpoly!cascade!bigsur!clari!lnovak From: lnovak@clari.net (Liam Novak) Newsgroups: comp.sys.m68k Subject: Re: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996Mar16.064800.2225@clari.net> Date: Sat, 16 Mar 1996 06:48:00 GMT References: <1996Mar15.133900.1523@illinois.edu> Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net X-Newsreader: NewsWatcher 2.0b24 Lines: 10 In article <1996Mar15.133900.1523@illinois.edu>, nmartinez@illinois.edu (Nadine Martínez) writes: > Compiling stardust.c with Think C 6.0 vs MPW C with direct 68882 inlining. > By passing the -mc68882 compiler flag in Think C, all sin/cos and sqrt operations compile directly to FPU opcodes. Beware of System 7.5.1 memory leaks if you allocate large sample buffers in the application heap. Use NewPtrSys instead. -- Liam Novak | lnovak@clari.net PGP Key fingerprint: 6B9C 6AEE 255F CC65 "Keep the channels open." From ckowalski@cern.ch Sun Mar 17 15:22:00 1996 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!cern!ckowalski From: ckowalski@cern.ch (Camila Kowalski) Newsgroups: comp.sys.m68k Subject: Re: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996Mar17.152200.7093@cern.ch> Date: Sun, 17 Mar 1996 15:22:00 GMT References: <1996Mar15.133900.1523@illinois.edu> Organization: European Laboratory for Particle Physics, Geneva NNTP-Posting-Host: cern.ch X-Newsreader: WinVN 0.99.5 Lines: 10 In article <1996Mar15.133900.1523@illinois.edu>, nmartinez@illinois.edu (Nadine Martínez) writes: > Compiling stardust.c with Think C 6.0 vs MPW C with direct 68882 inlining. > By passing the -mc68882 compiler flag in Think C, all sin/cos and sqrt operations compile directly to FPU opcodes. Mac users bragging about 6-minute render times while Unix workstations do it in 20 seconds. Classic Apple distortion field. -- Camila Kowalski Opinions expressed are solely my own and not those of European Laboratory for Particle Physics, Geneva. ------------------------------------------------------------ From hperez@earthlink.net Tue Mar 19 06:18:00 1996 Path: ashpoly!cascade!transat!eunet-fr!teletel!earthlink!hperez From: hperez@earthlink.net (Hana Pérez) Newsgroups: comp.sys.m68k Subject: Re: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996Mar19.061800.2162@earthlink.net> Date: Tue, 19 Mar 1996 06:18:00 GMT References: <1996Mar15.133900.1523@illinois.edu> Organization: EarthLink Network NNTP-Posting-Host: earthlink.net X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1996Mar15.133900.1523@illinois.edu>, nmartinez@illinois.edu (Nadine Martínez) writes: > Compiling stardust.c with Think C 6.0 vs MPW C with direct 68882 inlining. > By passing the -mc68882 compiler flag in Think C, all sin/cos and sqrt operations compile directly to FPU opcodes. Mac users bragging about 6-minute render times while Unix workstations do it in 20 seconds. Classic Apple distortion field. -- Hana Pérez Tape trading list upon request. SASE required for international mail. From jsmirnov@clari.net Thu Mar 21 09:40:00 1996 Path: ashpoly!cascade!transat!eunet-de!hansebox!clari!jsmirnov From: jsmirnov@clari.net (Jens Smirnov) Newsgroups: comp.sys.m68k Subject: Re: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996Mar21.094000.4246@clari.net> Date: Thu, 21 Mar 1996 09:40:00 GMT References: <1996Mar15.133900.1523@illinois.edu> <1996Mar16.064800.2225@clari.net> Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net X-Newsreader: Netscape 1.1N Lines: 8 In article <1996Mar16.064800.2225@clari.net>, lnovak@clari.net (Liam Novak) 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. -- Jens Smirnov Tape trading list upon request. SASE required for international mail.