Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 9/18/84; site harvard.ARPA Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!think!harvard!sasaki From: sasaki@harvard.ARPA (Marty Sasaki) Newsgroups: net.micro.68k,net.arch Subject: Re: RISC Message-ID: <224@harvard.ARPA> Date: Fri, 28-Jun-85 22:39:29 EDT Article-I.D.: harvard.224 Posted: Fri Jun 28 22:39:29 1985 Date-Received: Sun, 30-Jun-85 00:48:25 EDT References: <639@vax2.fluke.UUCP> <2743@nsc.UUCP> <576@terak.UUCP> <5690@utzoo.UUCP> <1109@peora.UUCP> <5716@utzoo.UUCP> <78@rtp47.UUCP> Reply-To: sasaki@harvard.UUCP (Marty sasaki) Distribution: net Organization: Harvard Science Center Lines: 18 Xref: watmath net.micro.68k:976 net.arch:1502 Summary: This may be an increadibly stupid question, but I remember reading an article (in CACM, I think) on RISC machines where most of the instructions occupied a single byte of memory. Through cleverness, something like 95% of all instructions executed occupied a single byte (including operands). This meant that programs would be smaller and would run faster by the simple fact that they required fewer memory references to get work done. All of the recent discussion on RISC machines hasn't mentioned this at all. Have things changed sufficiently in the recent past that this small program size doesn't matter? Is this a really dumb question? -- ---------------- Marty Sasaki net: sasaki@harvard.{arpa,uucp} Havard University Science Center phone: 617-495-1270 One Oxford Street Cambridge, MA 02138