From bbrennan@pipeline.com Sat Sep 21 19:33:27 1996 Path: ashpoly!cascade!mxgate!coyoacan!pipeline!bbrennan From: bbrennan@pipeline.com (Bastien Brennan) Newsgroups: comp.sys.m68k Subject: [SPECULATION] Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996Sep21.193327.4834@pipeline.com> Date: Sat, 21 Sep 1996 19:33:27 GMT Organization: The Pipeline Internet Network NNTP-Posting-Host: pipeline.com X-Newsreader: WinVN 0.99.5 Lines: 11 Compiling stardust.c with Think C 6.0 vs MPW C with direct 68882 inlining. Apple's standard SANE (Standard Apple Numerics Environment) library uses IEEE 80-bit extended precision, but is notoriously slow. A 1024-point radix-2 FFT drops from 142 milliseconds under SANE to just 18 milliseconds using direct FPU instructions. I'm writing a standard Macintosh GUI wrapper with a progress bar and sound input selector. -- Bastien Brennan | bbrennan@pipeline.com PGP Key fingerprint: 54C7 53AD 32BE 78B3 "Keep the channels open." From jtanaka@demon.co.uk Sun Sep 22 06:04:27 1996 Path: ashpoly!cascade!transat!eunet-de!hansebox!demon!jtanaka From: jtanaka@demon.co.uk (Joachim Tanaka) Newsgroups: comp.sys.m68k Subject: Re: [SPECULATION] Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996Sep22.060427.3005@demon.co.uk> Date: Sun, 22 Sep 1996 06:04:27 GMT References: <1996Sep21.193327.4834@pipeline.com> Organization: Demon Internet Services UK NNTP-Posting-Host: demon.co.uk X-Newsreader: WinVN 0.99.5 Lines: 10 In article <1996Sep21.193327.4834@pipeline.com>, bbrennan@pipeline.com (Bastien Brennan) writes: > Compiling stardust.c with Think C 6.0 vs MPW C with direct 68882 inlining. > Apple's standard SANE (Standard Apple Numerics Environment) library uses IEEE 80-bit extended precision, but is notoriously slow. The 68040 on the Quadra is a monster -- the on-chip FPU executes FFT butterflies in less than 4 milliseconds. -- Joachim Tanaka SupraFAXModem 28.8 @ 33,600 bps "When every node is whole, the channel opens." From stanaka@earthlink.net Mon Sep 23 06:19:27 1996 Path: ashpoly!cascade!mxgate!coyoacan!earthlink!stanaka From: stanaka@earthlink.net (Steve Tanaka) Newsgroups: comp.sys.m68k Subject: Re: [SPECULATION] Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996Sep23.061927.9004@earthlink.net> Date: Mon, 23 Sep 1996 06:19:27 GMT References: <1996Sep21.193327.4834@pipeline.com> <1996Sep22.060427.3005@demon.co.uk> Organization: EarthLink Network NNTP-Posting-Host: earthlink.net X-Newsreader: TIN [version 1.2 PL2] Lines: 11 In article <1996Sep22.060427.3005@demon.co.uk>, jtanaka@demon.co.uk (Joachim Tanaka) writes: > The 68040 on the Quadra is a monster -- the on-chip FPU executes FFT butterflies in less than 4 milliseconds. Beware of System 7.5.1 memory leaks if you allocate large sample buffers in the application heap. Use NewPtrSys instead. If you are posting from AOL, please read RFC 1036 before posting again. -- Steve Tanaka "When every node is whole, the channel opens." Amiga 1200 (Blizzard 030/50) / AmigaOS 3.1