From owner-freebsd-hackers Tue May 19 23:30:43 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id XAA25655 for freebsd-hackers-outgoing; Tue, 19 May 1998 23:30:43 -0700 (PDT) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from quark.ChrisBowman.com (crbowman.erols.com [209.122.47.155]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id XAA25505 for ; Tue, 19 May 1998 23:30:10 -0700 (PDT) (envelope-from crb@ChrisBowman.com) Received: from localhost (crb@localhost) by quark.ChrisBowman.com (8.8.8/8.8.8) with SMTP id CAA01936; Wed, 20 May 1998 02:30:14 -0400 (EDT) (envelope-from crb@ChrisBowman.com) X-Authentication-Warning: quark.ChrisBowman.com: crb owned process doing -bs Date: Wed, 20 May 1998 02:30:11 -0400 (EDT) From: "Christopher R. Bowman" To: Greg Lehey cc: Wilko Bulte , freebsd-hackers@FreeBSD.ORG Subject: Re: Original PC (was: talk (fwd)) In-Reply-To: <19980520144300.M20476@freebie.lemis.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Wed, 20 May 1998, Greg Lehey wrote: >On Tue, 19 May 1998 at 23:57:14 +0200, Wilko Bulte wrote: >> As Oliver Fromme wrote... >>> In list.freebsd-hackers Mike wrote: >>>> [...] >>>> I've always heard (I have no motorola experience, yet) that motorola asm >>>> blows x86 away when it comes to efficiency. A friend I have develops for >>>> Be and he's always ranting about it. :) >>> >>> He's right. The x86 has 4 general-purpose registers, each of >>> them 16 bits (they were extended to 32 bits in the 80386) and >>> 4 address registers of the same size. And there are certain >>> restrictions on their usage, e.g. you can only use the CX >>> register as counter in the "loop" instruction etc. >>> >>> On the other hand, the Motorola 68k has 8 general-purpose >>> registers of 32 bits and 8 address registers (also 32 bits). >>> There is no restriction on their use, except that the 8th >>> address regsiter is the default stack pointer. >>> >>> I programmed on both architectures in assembler, and I have to >>> say that the 68k is definitely easier to program, and the >>> higher number of registers allows for efficient programming. >>> >>> Maybe it was the biggest mistake ever made in computer history >>> when IBM selected the 8088 for their first PC back in 1979. >>> (Or was it 1978? Don't know, I probably couldn't even spell >>> the word "Computer" correctly back then.) If they used the >>> 68000 -- which was already available at that time -- we would >>> have less problems today, I guess. >> >> An attractive (to me ;-) explanation is that IBM did not want to use the >> 68K because it was a threat (performance wise) to their high profit machine >> range. >> >> Urban legend or not, it sure sounds OK ;-) > >I just realized what this thread was about after deleting a lot of >messages unread, so if I repeat something that has already gone past, >please forgive me. > >Did it really take IBM so long to develop the PC? My recollection was >that it was a sort of half-hearted effort after the devastating >success of the 5100. In any case, the obvious reason for the choice >of processor was the software available--CP/M 86 and 86-DOS for the >8088, nothing for the 68K. On top of that, the 8088 was cheaper >because it had 8 bit memory (remember that most chips in those days >were single bit). I don't think they had the slightest concern about >attacking their mainframe machines, which were as fast as they needed >to be (quite literally). Vague recollections that Intel let them liscense the processor, or become a second source whereas Motorola wouldn't. --------- Christopher R. Bowman crb@ChrisBowman.com My home page To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message