Article View: comp.sys.next.software
Article #1216Re: 3.0 Number Crunching Speed - a Reason to Relax
From: Steven T. Abell
Date: Thu, 01 Oct 1992 06:21
Date: Thu, 01 Oct 1992 06:21
19 lines
1321 bytes
1321 bytes
Matt Dillon writes: > Transcendentals were left out of the 040. Motorola provides the trap > code and states that running transcendentals on an 040 through the trap > code is *FASTER* then running transcendentals inline on an 030/882 > combination. > Non-transcendentals, the majority of what NeXTSTEP uses, run much, much > faster on an 040 then an 030/882 combination. There was a comparison in NEXTWORLD about a year ago that included a transcendentals test on a NeXT and a MacIIfx with an 882. The 030/882 was faster, but clock differences may make the comparison invalid. It's pretty hard to imagine doing transcendentals faster in assembler than in microcode, even bad microcode, especially when there's a trap involved. Steve abell@netcom.com
Technical Details
Message-ID:
<1992Oct1.062132.29436@netcom.com>Path:
arc1-ca.pugleaf.net!sparky!uunet!decwrl!csus.edu!netcom.com!abellReferences:
<1992Sep22.024416.19581@wam.umd.edu> <1992Sep24.070405.15244@udel.edu> <dillon.0ocr@overload.Berkeley.CA.US>Show Full Headers
Complete Article Headers
Newsgroups: comp.sys.next.software Path: arc1-ca.pugleaf.net!sparky!uunet!decwrl!csus.edu!netcom.com!abell From: abell@netcom.com (Steven T. Abell) Subject: Re: 3.0 Number Crunching Speed - a Reason to Relax Message-ID: <1992Oct1.062132.29436@netcom.com> Sender: abell@netcom.com (Steven T. Abell) Organization: Netcom - Online Communication Services (408 241-9760 guest) References: <1992Sep22.024416.19581@wam.umd.edu> <1992Sep24.070405.15244@udel.edu> <dillon.0ocr@overload.Berkeley.CA.US> Date: Thu, 1 Oct 1992 06:21:32 GMT Lines: 19