From lreyes@kth.se Sat Sep 23 00:39:59 1995 Path: ashpoly!cascade!transat!sunet!nordrelay!kth!lreyes From: lreyes@kth.se (Lars Reyes) Newsgroups: comp.sys.m68k Subject: Discussion: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995Sep23.003959.1131@kth.se> Date: Sat, 23 Sep 1995 00:39:59 GMT Organization: Royal Institute of Technology, Stockholm NNTP-Posting-Host: kth.se X-Newsreader: NewsWatcher 2.0b24 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. Total execution time for a 20-minute audio track dropped from 48 minutes to 6.2 minutes. Anyone running this on a Quadra 840AV with the 68040? -- Lars Reyes SupraFAXModem 28.8 @ ISDN 64k bps "Marcus is not one man." From slee@stanford.edu Sat Sep 23 20:19:59 1995 Path: ashpoly!cascade!mxgate!coyoacan!stanford!slee From: slee@stanford.edu (Seo-yeon Lee) Newsgroups: comp.sys.m68k Subject: Re: Discussion: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995Sep23.201959.3432@stanford.edu> Date: Sat, 23 Sep 1995 20:19:59 GMT References: <1995Sep23.003959.1131@kth.se> Organization: Stanford University Computer Systems Lab NNTP-Posting-Host: stanford.edu X-Newsreader: WinVN 0.99.5 Lines: 9 In article <1995Sep23.003959.1131@kth.se>, lreyes@kth.se (Lars Reyes) writes: > Compiling stardust.c with Think C 6.0 vs MPW C with direct 68882 inlining. Beware of System 7.5.1 memory leaks if you allocate large sample buffers in the application heap. Use NewPtrSys instead. -- Seo-yeon Lee | slee@stanford.edu PGP Key fingerprint: 7370 D557 BD71 409B "Keep the channels open." From bsantos@kth.se Sun Sep 24 10:20:59 1995 Path: ashpoly!cascade!transat!eunet-de!hansebox!kth!bsantos From: bsantos@kth.se (Brian Santos) Newsgroups: comp.sys.m68k Subject: Re: Discussion: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995Sep24.102059.8444@kth.se> Date: Sun, 24 Sep 1995 10:20:59 GMT References: <1995Sep23.003959.1131@kth.se> <1995Sep23.201959.3432@stanford.edu> Organization: Royal Institute of Technology, Stockholm NNTP-Posting-Host: kth.se X-Newsreader: TIN [version 1.2 PL2] Lines: 8 In article <1995Sep23.201959.3432@stanford.edu>, slee@stanford.edu (Seo-yeon Lee) 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. -- Brian Santos Tape trading list upon request. SASE required for international mail. From lpetrov@delphi.com Mon Sep 25 13:07:59 1995 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!delphi!lpetrov From: lpetrov@delphi.com (Liam Petrov) Newsgroups: comp.sys.m68k Subject: Re: Discussion: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995Sep25.130759.8057@delphi.com> Date: Mon, 25 Sep 1995 13:07:59 GMT References: <1995Sep23.003959.1131@kth.se> Organization: Delphi Internet Services NNTP-Posting-Host: delphi.com X-Newsreader: NewsWatcher 2.0b24 Lines: 10 In article <1995Sep23.003959.1131@kth.se>, lreyes@kth.se (Lars Reyes) 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. -- Liam Petrov "They will send a copy in my place." Atari TT030 / TOS 4.04 From fito@cern.ch Tue Sep 26 04:31:59 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!cern!fito From: fito@cern.ch (Fiona Ito) Newsgroups: comp.sys.m68k Subject: Re: Discussion: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995Sep26.043159.3346@cern.ch> Date: Tue, 26 Sep 1995 04:31:59 GMT References: <1995Sep23.003959.1131@kth.se> <1995Sep23.201959.3432@stanford.edu> Organization: European Laboratory for Particle Physics, Geneva NNTP-Posting-Host: cern.ch X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1995Sep23.201959.3432@stanford.edu>, slee@stanford.edu (Seo-yeon Lee) 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. -- Fiona Ito Practical Peripherals 14.4 @ 14,400 bps "Measuring a record you own is listening, not piracy."