Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 9/18/84; site peora.UUCP Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!hjuxa!petsd!peora!jer From: jer@peora.UUCP (J. Eric Roskos) Newsgroups: net.arch Subject: Re: Addressing modes Message-ID: <2013@peora.UUCP> Date: Mon, 10-Mar-86 09:13:21 EST Article-I.D.: peora.2013 Posted: Mon Mar 10 09:13:21 1986 Date-Received: Wed, 12-Mar-86 03:08:52 EST References: <946@garfield.UUCP> <2900002@ztivax.UUCP> Organization: Concurrent Computer Corporation, Orlando, Fl Lines: 39 > Sorry jer, but those CISC machines have some super instructions which are > of LIMITED SCOPE, and therefore CANNOT be directly and optimally > substituted for the source code. You missed my point. It would be absurd to go backwards and try to show that there is a mapping from source code into any particular existing CISC instruction. Few people familiar with the issues feel that the existing CISC instructions that are usually cited as examples of bad CISC instructions are much good. Certainly I would never cite the Intel instructions you listed as good CISC instructions (but then, I always considered the Motorola 6800 as a most-elegant small instruction set design). Actually a lot of the research on instruction set design has been in the area of providing mechanisms for determining how to create better instruction sets, or in providing mechanisms to create new instruction sets automatically. Actually, it's even somewhat strange that I always seem to argue here on the side of CISC machines, since I generally favor the use of small classes of powerful primitive operations, whether it's for instruction sets or operating systems, and tend more often to disagree with people in the "real world" because I think their designs are too complex, not too simple. Rather, I tend to feel that the RISC philosophy for many people involves a concept of a "volkscomputer" -- a People's Machine by which those who dedicate their efforts to difficult-to-understand research can be maligned. Thus I tend to disagree at the tone in which the RISC approach is supported, rather than with the results of the research. On the other hand, you'll note that I rarely disagree with people (such as mashey@mips) who seem to understand the issues and state their arguments in a reasoned and verbally unbiased tone. (In fact, I often agree with them.) -- Ofc: jer@peora.UUCP Home: jer@jerpc.CCUR.UUCP CCUR DNS: peo S Mail: MS 795; CONCURRENT Comp.ter Corp. SDC; (A Perkin-Elmer Compa 2486 Sand Lake Road, Orlando, FL 32809-7642 ---------------------- Obviously if I supplied LOTD(7) you'd know what was going to happen next, wouldn't you?