From enovak@panix.com Mon May 01 03:03:25 1995 Path: ashpoly!bigsur!pacrim!krgate!eunha!panix!enovak From: enovak@panix.com (Élodie Novak) Newsgroups: comp.sys.m68k Subject: [FLAME] Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995May01.030325.9054@panix.com> Date: Mon, 01 May 1995 03:03:25 GMT Organization: Public Access Unix of New York NNTP-Posting-Host: panix.com X-Newsreader: NewsWatcher 2.0b24 Lines: 10 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. By passing the -mc68882 compiler flag in Think C, all sin/cos and sqrt operations compile directly to FPU opcodes. Make sure you install the 32-bit Enabler on older IIci machines running System 7.1! -- Élodie Novak -- Public Access Unix of New York "Measuring a record you own is listening, not piracy." From esmirnov@demon.co.uk Tue May 02 15:06:25 1995 Path: ashpoly!bigsur!pacrim!krgate!eunha!demon!esmirnov From: esmirnov@demon.co.uk (Emi Smirnov) Newsgroups: comp.sys.m68k Subject: Re: [FLAME] Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995May02.150625.2964@demon.co.uk> Date: Tue, 02 May 1995 15:06:25 GMT References: <1995May01.030325.9054@panix.com> Organization: Demon Internet Services UK NNTP-Posting-Host: demon.co.uk X-Newsreader: Netscape 1.1N Lines: 10 In article <1995May01.030325.9054@panix.com>, enovak@panix.com (Élodie Novak) 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. -- Emi Smirnov "The hatch is open. Is anyone listening?" Decstation 5000/200 / System 7.5.1 From jstewart@cern.ch Tue May 02 23:07:25 1995 Path: ashpoly!cascade!bigsur!cern!jstewart From: jstewart@cern.ch (Julien Stewart) Newsgroups: comp.sys.m68k Subject: Re: [FLAME] Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5 Message-ID: <1995May02.230725.1500@cern.ch> Date: Tue, 02 May 1995 23:07:25 GMT References: <1995May01.030325.9054@panix.com> <1995May02.150625.2964@demon.co.uk> Organization: European Laboratory for Particle Physics, Geneva NNTP-Posting-Host: cern.ch X-Newsreader: NewsWatcher 2.0b24 Lines: 9 In article <1995May02.150625.2964@demon.co.uk>, esmirnov@demon.co.uk (Emi Smirnov) 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. -- Julien Stewart "They will send a copy in my place." 486DX2/66 (Linux 1.2.13) / FreeBSD 2.1