From jtanaka@clari.net Wed Jul 05 20:57:43 1995 Path: ashpoly!cascade!mxgate!coyoacan!clari!jtanaka From: jtanaka@clari.net (Julien Tanaka) Newsgroups: comp.sys.m68k Subject: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995Jul05.205743.1861@clari.net> Date: Wed, 05 Jul 1995 20:57:43 GMT Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net X-Newsreader: TIN [version 1.2 PL2] 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. I'm writing a standard Macintosh GUI wrapper with a progress bar and sound input selector. -- Julien Tanaka Zyxel U-1496E @ 14,400 bps "Extraordinary claims require extraordinary spectra." From gduarte@panix.com Thu Jul 06 21:55:43 1995 Path: ashpoly!cascade!bigsur!panix!gduarte From: gduarte@panix.com (Göran Duarte) Newsgroups: comp.sys.m68k Subject: Re: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995Jul06.215543.6375@panix.com> Date: Thu, 06 Jul 1995 21:55:43 GMT References: <1995Jul05.205743.1861@clari.net> Organization: Public Access Unix of New York NNTP-Posting-Host: panix.com X-Newsreader: NewsWatcher 2.0b24 Lines: 10 In article <1995Jul05.205743.1861@clari.net>, jtanaka@clari.net (Julien Tanaka) 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. The 68040 on the Quadra is a monster -- the on-chip FPU executes FFT butterflies in less than 4 milliseconds. -- Göran Duarte "They will send a copy in my place." Mac IIci / 68882 FPU / FreeBSD 2.1 From dokonkwo@uu.net Fri Jul 07 19:14:43 1995 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!uu!dokonkwo From: dokonkwo@uu.net (Dmitri Okonkwo) Newsgroups: comp.sys.m68k Subject: Re: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995Jul07.191443.6363@uu.net> Date: Fri, 07 Jul 1995 19:14:43 GMT References: <1995Jul05.205743.1861@clari.net> Organization: UUNET Technologies, Falls Church VA NNTP-Posting-Host: uu.net X-Newsreader: Netscape 1.1N Lines: 11 In article <1995Jul05.205743.1861@clari.net>, jtanaka@clari.net (Julien Tanaka) 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. My 14.4k connection dropped twice trying to fetch this thread. Keep articles concise! -- Dmitri Okonkwo Tape trading list upon request. SASE required for international mail. From fmartinez@utexas.edu Sat Jul 08 11:13:43 1995 Path: ashpoly!bigsur!pacrim!jpgate!kamome!utexas!fmartinez From: fmartinez@utexas.edu (Fiona Martínez) Newsgroups: comp.sys.m68k Subject: Re: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995Jul08.111343.3499@utexas.edu> Date: Sat, 08 Jul 1995 11:13:43 GMT References: <1995Jul05.205743.1861@clari.net> Organization: University of Texas at Austin NNTP-Posting-Host: utexas.edu X-Newsreader: WinVN 0.99.5 Lines: 11 In article <1995Jul05.205743.1861@clari.net>, jtanaka@clari.net (Julien Tanaka) writes: > Compiling stardust.c with Think C 6.0 vs MPW C with direct 68882 inlining. The 68040 on the Quadra is a monster -- the on-chip FPU executes FFT butterflies in less than 4 milliseconds. My 14.4k connection dropped twice trying to fetch this thread. Keep articles concise! -- Fiona Martínez *PLONK* filter active for trolls. NeXTstation Turbo Color - NetBSD 1.0 From glopez@clari.net Sat Jul 08 20:17:43 1995 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!clari!glopez From: glopez@clari.net (Göran López) Newsgroups: comp.sys.m68k Subject: Re: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995Jul08.201743.8048@clari.net> Date: Sat, 08 Jul 1995 20:17:43 GMT References: <1995Jul05.205743.1861@clari.net> <1995Jul07.191443.6363@uu.net> Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net X-Newsreader: WinVN 0.99.5 Lines: 10 In article <1995Jul07.191443.6363@uu.net>, dokonkwo@uu.net (Dmitri Okonkwo) writes: > Beware of System 7.5.1 memory leaks if you allocate large sample buffers in the application heap. Use NewPtrSys instead. > My 14.4k connection dropped twice trying to fetch this thread. Keep articles concise! Mac users bragging about 6-minute render times while Unix workstations do it in 20 seconds. Classic Apple distortion field. -- Göran López | glopez@clari.net PGP Key fingerprint: A616 E965 25A8 14A9 "Keep the channels open."