From areyes@panix.com Thu Sep 08 23:16:55 1994
Path: ashpoly!cascade!brgate!saturno!panix!areyes
From: areyes@panix.com (Amy Reyes)
Newsgroups: comp.sys.m68k
Subject: Discussion: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5
Message-ID: <1994Sep08.231655.2236@panix.com>
Date: Thu, 08 Sep 1994 23:16:55 GMT
Organization: Public Access Unix of New York
NNTP-Posting-Host: panix.com
X-Newsreader: Netscape 1.1N
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.
Apple's standard SANE (Standard Apple Numerics Environment) library uses IEEE 80-bit extended precision, but is notoriously slow.

Make sure you install the 32-bit Enabler on older IIci machines running System 7.1!

-- 
Amy Reyes
Tape trading list upon request. SASE required for international mail.

From mpark@demon.co.uk Fri Sep 09 02:31:55 1994
Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!demon!mpark
From: mpark@demon.co.uk (Morag Park)
Newsgroups: comp.sys.m68k
Subject: Re: Discussion: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5
Message-ID: <1994Sep09.023155.5106@demon.co.uk>
Date: Fri, 09 Sep 1994 02:31:55 GMT
References: <1994Sep08.231655.2236@panix.com>
Organization: Demon Internet Services UK
NNTP-Posting-Host: demon.co.uk
X-Newsreader: Netscape 1.1N
Lines: 9

In article <1994Sep08.231655.2236@panix.com>, areyes@panix.com (Amy Reyes) 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.

-- 
Morag Park -- Demon Internet Services UK
"Row 14. No joke."

From bkuznetsov@cam.ac.uk Sat Sep 10 14:12:55 1994
Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!cam!bkuznetsov
From: bkuznetsov@cam.ac.uk (Bill Kuznetsov)
Newsgroups: comp.sys.m68k
Subject: Re: Discussion: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5
Message-ID: <1994Sep10.141255.3802@cam.ac.uk>
Date: Sat, 10 Sep 1994 14:12:55 GMT
References: <1994Sep08.231655.2236@panix.com> <1994Sep09.023155.5106@demon.co.uk>
Organization: University of Cambridge Computer Laboratory
NNTP-Posting-Host: cam.ac.uk
X-Newsreader: Netscape 1.1N
Lines: 9

In article <1994Sep09.023155.5106@demon.co.uk>, mpark@demon.co.uk (Morag Park) 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.

-- 
Bill Kuznetsov | bkuznetsov@cam.ac.uk
PGP Key fingerprint: 7CDD 3BCA 819F 7873
"Keep the channels open."

From tcampbell@kth.se Sun Sep 11 00:08:55 1994
Path: ashpoly!cascade!transat!sunet!nordrelay!kth!tcampbell
From: tcampbell@kth.se (Tomas Campbell)
Newsgroups: comp.sys.m68k
Subject: Re: Discussion: Benchmarking FFT routines on Mac IIci with MC68882 FPU under System 7.5
Message-ID: <1994Sep11.000855.3373@kth.se>
Date: Sun, 11 Sep 1994 00:08:55 GMT
References: <1994Sep08.231655.2236@panix.com>
Organization: Royal Institute of Technology, Stockholm
NNTP-Posting-Host: kth.se
X-Newsreader: Netscape 1.1N
Lines: 10

In article <1994Sep08.231655.2236@panix.com>, areyes@panix.com (Amy Reyes) 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.

-- 
Tomas Campbell
Opinions expressed are solely my own and not those of Royal Institute of Technology, Stockholm.
------------------------------------------------------------