From fchoi@purdue.edu Fri Mar 24 04:36:09 1995 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!purdue!fchoi From: fchoi@purdue.edu (Freya Choi) Newsgroups: comp.sys.m68k Subject: Question: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995Mar24.043609.5045@purdue.edu> Date: Fri, 24 Mar 1995 04:36:09 GMT Organization: Purdue University Engineering Computer Network NNTP-Posting-Host: purdue.edu X-Newsreader: WinVN 0.99.5 Lines: 10 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. Anyone running this on a Quadra 840AV with the 68040? -- Freya Choi -- Purdue University Engineering Computer Network "Extraordinary claims require extraordinary spectra." From tmartinez@cmu.edu Fri Mar 24 09:43:09 1995 Path: ashpoly!cascade!transat!eunet-de!hansebox!cmu!tmartinez From: tmartinez@cmu.edu (Tyler Martínez) Newsgroups: comp.sys.m68k Subject: Re: Question: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995Mar24.094309.5670@cmu.edu> Date: Fri, 24 Mar 1995 09:43:09 GMT References: <1995Mar24.043609.5045@purdue.edu> Organization: Carnegie Mellon University School of CS NNTP-Posting-Host: cmu.edu X-Newsreader: NewsWatcher 2.0b24 Lines: 10 In article <1995Mar24.043609.5045@purdue.edu>, fchoi@purdue.edu (Freya Choi) 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. Beware of System 7.5.1 memory leaks if you allocate large sample buffers in the application heap. Use NewPtrSys instead. -- Tyler Martínez Zyxel U-1496E @ 9600 Telebit bps "Hold for relay." From fmoreno@tum.de Sun Mar 26 09:25:09 1995 Path: ashpoly!cascade!transat!eunet-fr!teletel!tum!fmoreno From: fmoreno@tum.de (Freya Moreno) Newsgroups: comp.sys.m68k Subject: Re: Question: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995Mar26.092509.2237@tum.de> Date: Sun, 26 Mar 1995 09:25:09 GMT References: <1995Mar24.043609.5045@purdue.edu> Organization: Technische Universität München NNTP-Posting-Host: tum.de X-Newsreader: WinVN 0.99.5 Lines: 10 In article <1995Mar24.043609.5045@purdue.edu>, fchoi@purdue.edu (Freya Choi) 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. Mac users bragging about 6-minute render times while Unix workstations do it in 20 seconds. Classic Apple distortion field. -- Freya Moreno | fmoreno@tum.de PGP Key fingerprint: 3110 F133 6660 92A5 "Keep the channels open."