Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!uunet!jarthur!elroy.jpl.nasa.gov!decwrl!sgi!rpw3@rigden.wpd.sgi.com From: rpw3@rigden.wpd.sgi.com (Rob Warnock) Newsgroups: comp.arch Subject: Re: Official comments on recent ARM postings Message-ID: <52301@sgi.sgi.com> Date: 1 Mar 90 06:20:23 GMT References: <8948@wpi.wpi.edu> <11245@encore.Encore.COM> Sender: rpw3@rigden.wpd.sgi.com Reply-To: rpw3@rigden.UUCP (Robert P. Warnock) Organization: Silicon Graphics, Inc., Mountain View, CA Lines: 30 In article <11245@encore.Encore.COM> jkenton@pinocchio.encore.com (Jeff Kenton) writes: +--------------- | From article <8948@wpi.wpi.edu>, by jhallen@wpi.wpi.edu (Joseph H Allen): | > (otherwise, I like ARM very much. A RISC processor you can actually program | > assembly language in! How fast is the latest ARM these days?) | Motorola's 88000 is pretty nice for writing assembly language, and MIPS looks | decent too, although I haven't tried it. No opinion on SPARC, but Intel's 860 | looks a little baroque. | Any other opinions? We haven't had a good religious war in a week or two. +--------------- I really enjoyed programming Am29000 assembler, once I had gotten used to a real 3-address machine. (Of course I am biased. I was a consultant to AMD at the time...) It took me a while to quit thinking of it as a stuttering 2-address machine -- at first I found myself writing everything as " , , ". But then the "shape" of the instruction set started sinking in, and I soon learned to use the 3-addr-ness to save instructions (cycles). (Though, just because of what is typically being computed, one still ends up writing many "2-address" instructions.) Overall, it feels really "clean". -Rob ----- Rob Warnock, MS-9U/510 rpw3@sgi.com rpw3@pei.com Silicon Graphics, Inc. (415)335-1673 Protocol Engines, Inc. 2011 N. Shoreline Blvd. Mountain View, CA 94039-7311