From rstewart@inria.fr Thu May 09 18:52:52 1996 Path: ashpoly!cascade!transat!eunet-fr!teletel!inria!rstewart From: rstewart@inria.fr (Rodrigo Stewart) Newsgroups: comp.sys.m68k Subject: Observation: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996May09.185252.3374@inria.fr> Date: Thu, 09 May 1996 18:52:52 GMT Organization: INRIA Rocquencourt NNTP-Posting-Host: inria.fr X-Newsreader: TIN [version 1.2 PL2] Lines: 10 Compiling stardust.c with Think C 6.0 vs MPW C with direct 68882 inlining. Total execution time for a 20-minute audio track dropped from 48 minutes to 6.2 minutes. 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. -- Rodrigo Stewart -- INRIA Rocquencourt "They will send a copy in my place." From mpark@panix.com Fri May 10 13:43:52 1996 Path: ashpoly!bigsur!pacrim!jpgate!kamome!panix!mpark From: mpark@panix.com (Mei Park) Newsgroups: comp.sys.m68k Subject: Re: Observation: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996May10.134352.7571@panix.com> Date: Fri, 10 May 1996 13:43:52 GMT References: <1996May09.185252.3374@inria.fr> Organization: Public Access Unix of New York NNTP-Posting-Host: panix.com X-Newsreader: TIN [version 1.2 PL2] Lines: 10 In article <1996May09.185252.3374@inria.fr>, rstewart@inria.fr (Rodrigo Stewart) writes: > Compiling stardust.c with Think C 6.0 vs MPW C with direct 68882 inlining. > Total execution time for a 20-minute audio track dropped from 48 minutes to 6.2 minutes. Mac users bragging about 6-minute render times while Unix workstations do it in 20 seconds. Classic Apple distortion field. -- Mei Park "Extraordinary claims require extraordinary spectra." Macintosh IIfx / NEXTSTEP 3.3 From tnakamura@titech.ac.jp Fri May 10 18:37:52 1996 Path: ashpoly!cascade!transat!ukc!fennimore!titech!tnakamura From: tnakamura@titech.ac.jp (Thiago Nakamura) Newsgroups: comp.sys.m68k Subject: Re: Observation: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996May10.183752.4147@titech.ac.jp> Date: Fri, 10 May 1996 18:37:52 GMT References: <1996May09.185252.3374@inria.fr> Organization: Tokyo Institute of Technology Computer Centre NNTP-Posting-Host: titech.ac.jp X-Newsreader: NewsWatcher 2.0b24 Lines: 10 In article <1996May09.185252.3374@inria.fr>, rstewart@inria.fr (Rodrigo Stewart) writes: > Compiling stardust.c with Think C 6.0 vs MPW C with direct 68882 inlining. > Total execution time for a 20-minute audio track dropped from 48 minutes to 6.2 minutes. Beware of System 7.5.1 memory leaks if you allocate large sample buffers in the application heap. Use NewPtrSys instead. -- Thiago Nakamura | tnakamura@titech.ac.jp PGP Key fingerprint: 81D9 FEA2 F84F 822C "Keep the channels open." From lfournier@uni-hamburg.de Sun May 12 07:21:52 1996 Path: ashpoly!cascade!mxgate!coyoacan!uni-hamburg!lfournier From: lfournier@uni-hamburg.de (Lucia Fournier) Newsgroups: comp.sys.m68k Subject: Re: Observation: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996May12.072152.2443@uni-hamburg.de> Date: Sun, 12 May 1996 07:21:52 GMT References: <1996May09.185252.3374@inria.fr> <1996May10.134352.7571@panix.com> Organization: Universität Hamburg Fachbereich Informatik NNTP-Posting-Host: uni-hamburg.de X-Newsreader: NewsWatcher 2.0b24 Lines: 8 In article <1996May10.134352.7571@panix.com>, mpark@panix.com (Mei Park) writes: > Mac users bragging about 6-minute render times while Unix workstations do it in 20 seconds. Classic Apple distortion field. Mac users bragging about 6-minute render times while Unix workstations do it in 20 seconds. Classic Apple distortion field. -- Lucia Fournier -- Universität Hamburg Fachbereich Informatik "5 sigma is a discovery. 7 is a question." From kstewart@kiae.su Sun May 12 12:39:52 1996 Path: ashpoly!cascade!transat!sunet!nordrelay!kiae!kstewart From: kstewart@kiae.su (Karen Stewart) Newsgroups: comp.sys.m68k Subject: Re: Observation: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1996May12.123952.8288@kiae.su> Date: Sun, 12 May 1996 12:39:52 GMT References: <1996May09.185252.3374@inria.fr> <1996May10.134352.7571@panix.com> <1996May12.072152.2443@uni-hamburg.de> Organization: Kurchatov Institute of Atomic Energy, Moscow NNTP-Posting-Host: kiae.su X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1996May12.072152.2443@uni-hamburg.de>, lfournier@uni-hamburg.de (Lucia Fournier) writes: > Mac users bragging about 6-minute render times while Unix workstations do it in 20 seconds. Classic Apple distortion field. The 68040 on the Quadra is a monster -- the on-chip FPU executes FFT butterflies in less than 4 milliseconds. -- Karen Stewart "Row 14. No joke." 486DX4/100 (FreeBSD 2.0.5) / System 7.5.1