Date: Wed, 20 May 1998 23:47:17 -0400 (EDT) From: Chuck Robey <chuckr@glue.umd.edu> To: Soren Kristensen <soekris@alameda.net> Cc: hackers@FreeBSD.ORG Subject: Re: Original PC and talk Message-ID: <Pine.BSF.3.96.980520233946.312a-100000@localhost> In-Reply-To: <3563A210.31CF@alameda.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 20 May 1998, Soren Kristensen wrote: > Lets get it clear why IBM choose the 8088: > > The PC design team wanted more than a 8 bit CPU in their new PC, but > they didn't want to go to full 16 bit, as most peripheral chips > (especially cheap ones) at that time was 8 bit and DRAM was only > avaliable in x1 types. (IBM actually planned to release a machine with > only 16 kbyte....) > And at the time the decision was made, sometime around 1980, the only > avaliable chip that fit their need was Intel's 8088. Motorola only had > the 68000 avaliable, 68008, the 8 bit bus version, came later. > > But IBM was also using the 8086 in their Displaywriter Word Processor, > which may have influenced the choice. (Funny machine with big 8 inch > floppies, I once played with cp/m-86 on it :-) > > So that's why we are stuck with the x86 family, like it or not. But I > don't think it is as bad as a lot a people make it. It don't have a nice > architecture (anybody remember the great 32000 series from national > semiconductor ?), but the x86 is quite powerfull in assembler (my > favorite progamming language....), Actually, as just about any modern text shows, the X86 architecture is the slowest one out there. Take a look at one of the recent Hennesey & Patterson texts, you'll see. even with its limited number of > registers. And who writes code in assembler anymore ? Which is precisely why you don't want a processor that makes life easy on assembly language writers, the compilers are the only ones spewing assembly anymore, you want a processor to execute instructions FAST. Intel's X86 architecture is miserable for real optimization, although one must admit that Intel has done just about everything they could do to make a horrible instruction set go quickly. All the tricks they use, the RISC processors use, but the RISC processors are designed to make best use of those tricks, and the X86 architecture isn't. All those tricks (the same ones) are why processors like the DEC Alpha are so hot. Things like register renaming don't give you much improvement if you're talking about such a tiny humber of registers to begin with (referring to the X86 here). > > And with todays chips sizes the underlying processor architecture dosn't > matter so much anyway, it's more a matter of cache sizes and memory > bandwidth. > There was actually a short time where the new Pentium Pro-200 was the > fastest processor in the world, measured in specint95 and specfp95.... > > > > Best Regards, > > > Soren Kristensen > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-hackers" in the body of the message > > ----------------------------+----------------------------------------------- Chuck Robey | Interests include any kind of voice or data chuckr@glue.umd.edu | communications topic, C programming, and Unix. 213 Lakeside Drive Apt T-1 | Greenbelt, MD 20770 | I run Journey2 and picnic (FreeBSD-current) (301) 220-2114 | and jaunt (NetBSD). ----------------------------+----------------------------------------------- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.980520233946.312a-100000>