roberts@NBS-VMS.arpa (ROBERTS, JOHN) (09/24/86)
> From: Joe Kwan <jhk%csun.uucp@BRL.ARPA> > ...I think it's save to assume that the 80386 > is still too new (IBM doesn't have a 386 PC out yet) and too costly > (at least for a Compaq Deskpro 386 {for the price that they want, > you could almost go out and buy a Sun}) at this time to consider. Remember that a 32-bit 80386 running at 16MHz is in the performance class of a 32-bit 68020 running at ~16MHz. (Precise comparisons depend on what is being done and who is conducting the tests, and are highly political.) The 80386 appears to have a number of distinct advantages over the 80286 (linear address space, on-chip MMU, etc.) (This based on product descriptions of the three processors, and not related to peripheral chips, choice of operating system, ....) <Standard disclaimers.> John Roberts roberts@nbs-vms.ARPA ------
mc68020@gilbbs.UUCP (Thomas J Keller) (09/26/86)
In article <4104@brl-smoke.ARPA>, roberts@NBS-VMS.arpa (ROBERTS, JOHN) writes: > > From: Joe Kwan <jhk%csun.uucp@BRL.ARPA> > > ...I think it's save to assume that the 80386 > > is still too new (IBM doesn't have a 386 PC out yet) and too costly > > (at least for a Compaq Deskpro 386 {for the price that they want, > > you could almost go out and buy a Sun}) at this time to consider. > > Remember that a 32-bit 80386 running at 16MHz is in the performance > class of a 32-bit 68020 running at ~16MHz. (Precise comparisons > depend on what is being done and who is conducting the tests, and > are highly political.) The 80386 appears to have a number of distinct > advantages over the 80286 (linear address space, on-chip MMU, etc.) > (This based on product descriptions of the three processors, and not > related to peripheral chips, choice of operating system, ....) Yeah, maybe. But it's still got the same braindamaged instruction set and register complement as the 80286, 80186, 8086, 8080, 8008. In otherwords it is a 32 bit high speed microcontroller chip. Intel *STILL* hasn't learned how to make *COMPUTERS*. (I doubt they ever will) -- Disclaimer: Disclaimer? DISCLAIMER!? I don't need no stinking DISCLAIMER!!! tom keller "She's alive, ALIVE!" {ihnp4, dual}!ptsfa!gilbbs!mc68020 (* we may not be big, but we're small! *)
magore@watdcsu.UUCP (M.A.Gore - ICR) (09/28/86)
In article <1038@gilbbs.UUCP> mc68020@gilbbs.UUCP (Thomas J Keller) writes: >In article <4104@brl-smoke.ARPA>, roberts@NBS-VMS.arpa (ROBERTS, JOHN) writes: >> > From: Joe Kwan <jhk%csun.uucp@BRL.ARPA> >> > ...I think it's save to assume that the 80386 >> > is still too new (IBM doesn't have a 386 PC out yet) and too costly >> > (at least for a Compaq Deskpro 386 {for the price that they want, >> > you could almost go out and buy a Sun}) at this time to consider. >> >> Remember that a 32-bit 80386 running at 16MHz is in the performance >> class of a 32-bit 68020 running at ~16MHz. (Precise comparisons >> depend on what is being done and who is conducting the tests, and >> are highly political.) The 80386 appears to have a number of distinct >> advantages over the 80286 (linear address space, on-chip MMU, etc.) >> (This based on product descriptions of the three processors, and not >> related to peripheral chips, choice of operating system, ....) > > Yeah, maybe. But it's still got the same braindamaged instruction set and > register complement as the 80286, 80186, 8086, 8080, 8008. In otherwords > it is a 32 bit high speed microcontroller chip. Intel *STILL* hasn't > learned how to make *COMPUTERS*. (I doubt they ever will) ^^^^^^^^^^^^^^^^^^ Yes but it's worse the the 68000 took so long to come out. While the teams designing the 68k were backstabbing each other Intel had chips. I like the 68k (have one) IBM *would* have used them had they been out in time. Note that the 68k ref. manual states that the 68k was designed with the 6809 in mind --- so as with Intel for GOOD market reasons. Market reasons are more important then the BEST hardware that can be had (sooooooo sad but true). We live with what we have at the time and that which earns money.... # Mike Gore # Institute for Computer Research. # These ideas/concepts do not imply views held by the University of Waterloo.
caf@omen.UUCP (Chuck Forsberg WA7KGX) (09/30/86)
In article <2586@watdcsu.UUCP> magore@watdcsu.UUCP (M.A.Gore - ICR) writes:
: Yes but it's worse the the 68000 took so long to come out. While
:the teams designing the 68k were backstabbing each other Intel had
:chips. I like the 68k (have one) IBM *would* have used them had they
:been out in time. Note that the 68k ref. manual states that the 68k
:was designed with the 6809 in mind --- so as with Intel for GOOD
:market reasons. Market reasons are more important then the BEST
:hardware that can be had (sooooooo sad but true).
If the stories of the IBM PC's short design cycle are anywhere near true,
the 68k was definitely available when the PC design was started.
I suspect IBM chose the 8088 to save some money on chepaer chips and an 8 bit
bus, and because Bill Gates had told them a ten times jump in memory size would
be sufficient to differentiate the IBM PC from the 8-bit machines of the time.
I'm sure IBM also wanted to make it easy to port 8 bit CP/M applications to
the PC. The main legacy that remains from this is MSDOS's CP/M-80 inherited
brain damage.
Chuck Forsberg WA7KGX Author of Pro-YAM communications Tools for PCDOS and Unix
...!tektronix!reed!omen!caf Omen Technology Inc "The High Reliability Software"
Voice: 503-621-3406 17505-V Northwest Sauvie Island Road Portland OR 97231
TeleGodzilla BBS: 621-3746 300/1200 CIS:70007,2304 Genie:CAF Source:TCE022
omen Any ACU 1200 1-503-621-3746 se:--se: link ord: Giznoid in:--in: uucp
omen!/usr/spool/uucppublic/FILES lists all uucp-able files, updated hourlytimothym@tekigm2.UUCP (Timothy D Margeson) (09/30/86)
Insert a dsicussion about the new Compaq 386 Deskpro priced similar to a SUN workstation....... In article <1038@gilbbs.UUCP> mc68020@gilbbs.UUCP (Thomas J Keller) writes: > Yeah, maybe. But it's still got the same braindamaged instruction set and > register complement as the 80286, 80186, 8086, 8080, 8008. In otherwords > it is a 32 bit high speed microcontroller chip. Intel *STILL* hasn't > learned how to make *COMPUTERS*. (I doubt they ever will) > >tom keller "She's alive, ALIVE!" >{ihnp4, dual}!ptsfa!gilbbs!mc68020 > >(* we may not be big, but we're small! *) Intel may not make good computers, true, but I'll bet Compaq and IBM sell alot more 386 based machines than Sun and the others using 68k combined! The issue here is economics (and a thing called compatibility)! p.s. written on an IBM PC-AT, running Crosstalk XVI, to a Gould 9000, running rn overlaid running vi, talking over a multi-national network of ATT and ITT phones lines..... Who gives a rip what the processor is, as long as the machine can be used as a tool to get work (or play) done?
sambo@ukma.uky.csnet (Father of micro-ln) (09/30/86)
In article <1038@gilbbs.UUCP> mc68020@gilbbs.UUCP (Thomas J Keller) writes: >In article <4104@brl-smoke.ARPA>, roberts@NBS-VMS.arpa (ROBERTS, JOHN) writes: >> Remember that a 32-bit 80386 running at 16MHz is in the performance >> class of a 32-bit 68020 running at ~16MHz.... > > Yeah, maybe. But it's still got the same braindamaged instruction set and > register complement as the 80286, 80186, 8086, 8080, 8008. In otherwords > it is a 32 bit high speed microcontroller chip. Intel *STILL* hasn't > learned how to make *COMPUTERS*. (I doubt they ever will) I beg to differ. The 80386 has the same braindamaged instruction set only if you refuse to use the new instructions and addressing modes it has. About the only truly special purpose registers, if my memory serves me cor- rectly, are the stack pointer, the instruction pointer, and a few special status registers. It still is not the cleanest architecture, but it is quite a bit improved from previous family members. For me, the advantages are that its speed is fairly decent, and it can run MS-DOS (which is itself brain- damaged) programs in a multi-user, multi-tasking environment. Unfortunately, it is impossible to ignore MS-DOS software. -- Samuel A. Figueroa, Dept. of CS, Univ. of KY, Lexington, KY 40506-0027 UUCP: cbosgd!ukma!sambo CSNET: sambo@uky.csnet "Micro-ln is great, if only people would start using it."
kimery@wdl1.UUCP (Sam Kimery) (10/01/86)
/ wdl1:net.micro / caf@omen.UUCP (Chuck Forsberg WA7KGX) / 2:24 pm Sep 29, 1986 /
In article <2586@watdcsu.UUCP> magore@watdcsu.UUCP (M.A.Gore - ICR) writes:
: Yes but it's worse the the 68000 took so long to come out. While
:the teams designing the 68k were backstabbing each other Intel had
:chips. I like the 68k (have one) IBM *would* have used them had they
:been out in time.
::If the stories of the IBM PC's short design cycle are anywhere near true,
::the 68k was definitely available when the PC design was started.
::
::I suspect IBM chose the 8088 to save some money on chepaer chips and an 8 bit
::bus, and because Bill Gates had told them a ten times jump in memory size would
::be sufficient to differentiate the IBM PC from the 8-bit machines of the time.
The story I heard was that Intel approached IBM with a computer design
and managed to sell IBM the desigh for the PC. The design time for the
PC was MUCH smaller than ANY other project at IBM. This makes the story
even easier to believe.
Sam Kimery
kimery@ford-wdl1
sun!wdl1!kimery
"This is all hearsay and in no way resembles an offical statement"
Merry Feast of the Pig.
tim@ism780c.UUCP (Tim Smith) (10/01/86)
In article <1038@gilbbs.UUCP> mc68020@gilbbs.UUCP (Thomas J Keller) writes: >In article <4104@brl-smoke.ARPA>, roberts@NBS-VMS.arpa (ROBERTS, JOHN) writes: >> >> are highly political.) The 80386 appears to have a number of distinct >> advantages over the 80286 (linear address space, on-chip MMU, etc.) The 80286 has an on-chip MMU also. > > Yeah, maybe. But it's still got the same braindamaged > instruction set and register complement as the 80286, 80186, > 8086, 8080, 8008. Note that they have gotten rid of all the restrictions on addressing modes. This goes a long way toward fixing the architecture. -- What's the difference between a duck? Tim Smith USENET: sdcrdcf!ism780c!tim Compuserve: 72257,3706 Delphi or GEnie: mnementh
timothym@tekigm2.UUCP (10/03/86)
Someone (read some biased UNIX lover) wrote the following...... > Yeah, maybe. But it's still got the same braindamaged instruction set and > register complement as the 80286, 80186, 8086, 8080, 8008. In otherwords > it is a 32 bit high speed microcontroller chip. Intel *STILL* hasn't > learned how to make *COMPUTERS*. (I doubt they ever will) He must not remember the early developement of the 68000 processors..... They WERE designed with downward compatibility to the 6809, and 6800 uP's. So why is Intel braindamaged, and Motorola not ? Some gift of God ? Of course, we should not fail to mention that VAX's are all crap too, and all are archaic technology, so we should shun them like the plague !!!! Long live GOULD, Rah ! Tim Margeson (206)253-5240 PO Box 3500 d/s C1-937 @@ 'Who said that?' Vancouver, WA. 98668 {allegra..inhp4..decvax..ucbvax}!tektronix!tekigm2!timothym
rra202@uiucuxa.CSO.UIUC.EDU (10/03/86)
The fact that IBM wanted to use the 68k for it's microprocessor is a clue as to which one is better. Can you think of any major computer out since the 68000 that (except for compatibility reasons) doesn't use a 68000. Alot of the software for a 80386 is going to be taken, without rewrite, from a 8088. That software is going to use the 'brain damaged register set' and assume a segmented archeture. The 68000 is the wave of the future. We're not big, but we will be.
jimc@iscuva.UUCP (10/06/86)
In article <1028@tekigm2.UUCP> timothym@tekigm2.UUCP (Timothy D Margeson) writes: >Someone (read some biased UNIX lover) wrote the following...... > >> Yeah, maybe. But it's still got the same braindamaged instruction set and >> register complement as the 80286, 80186, 8086, 8080, 8008. In otherwords >> it is a 32 bit high speed microcontroller chip. Intel *STILL* hasn't >> learned how to make *COMPUTERS*. (I doubt they ever will) > >He must not remember the early developement of the 68000 processors..... > >They WERE designed with downward compatibility to the 6809, and 6800 uP's. So >why is Intel braindamaged, and Motorola not ? Some gift of God ? Of course, Huh? I don't see how the 68000 family is compatible with the 6800 series, they have different architectures (accumulator-based vs general register), different register sets, different object code... As I recall, Mot specifically stated early on that they were scrapping compatiblity in favor of a better architecture. The only compatibilities I see are the '68' in the part number, and the fact that they share the same type of status register, and the *hardware* compatiblity the 68000 has with the old peripheral chips. Get it right. -- +----------------+ ! II CCCCCC ! Jim Cathey ! II SSSSCC ! ISC Systems Corp. ! II CC ! Spokane, WA ! IISSSS CC ! UUCP: ihnp4!tektronix!reed!iscuva!jimc ! II CCCCCC ! (509)927-5757 +----------------+ "With excitement like this, who is needing enemas?"
timothym@tekigm2.UUCP (10/09/86)
In article <264@iscuva.UUCP> jimc@iscuva.UUCP (Jim Cathey) writes: > >Huh? I don't see how the 68000 family is compatible with the 6800 >series, they have different architectures ... different object code... As I >recall, Mot specifically stated ... The only compatibilities I see are >the '68' in the part number, and the fact that they share the same *hardware* > >Get it right. I did get it right. Intel 8086 series processors are in no way software compatible with 8080 processors (try running ANY 8080 code on an 8086 based machine hardware). They are HARDWARE compatible though, as are 6800 and 68000 processors. Bite your own lip please (:-). Tim Margeson (206)253-5240 PO Box 3500 d/s C1-937 @@ 'Who said that?' Vancouver, WA. 98668 {allegra..inhp4..decvax..ucbvax}!tektronix!tekigm2!timothym
jimc@iscuva.UUCP (10/13/86)
In article <1068@tekigm2.UUCP> timothym@tekigm2.UUCP (Timothy D Margeson) writes: >>series, they have different architectures ... different object code... As I >I did get it right. Intel 8086 series processors are in no way software >compatible with 8080 processors (try running ANY 8080 code on an 8086 based >machine hardware). Sorry if my point wasn't clear, but it was primarily that the things have the same architecture (especially if you ingore segmentation [we should all be so blessed]), and by implication that an 8080 program (or 8008 program) can be mechanically translated to the newer machine, WHERE IT WILL LOOK NEARLY THE SAME. i.e. the only arithmetic will be done in A, and that much register shuffling must be done, and that there is little that can be done about this. Now, I think Intel did a wonderful job on the 8080 (and I prefer it to the 6800), and their 8051 is nearly a work of art. BUT when they refused to expand the addressing range of the 8086, and instead provided segmentation so that existing programs' structures (as in 8080 CP/M) could be maintained I think they made their big mistake. Programmers have been lamenting this ever since. I can live with accumulator based machines, and I can live with base registers (what 8086 segmentation regs are mostly used as), but I cannot tolerate not being able to directly address the entire range of the machine. Supposedly the 80386 finally provides this. I just wonder what other little hurts are there so that it can run older code. As for running 8080 code on an 8086, I thought every user of BASICA did that! :-) -- +----------------+ ! II CCCCCC ! Jim Cathey ! II SSSSCC ! ISC Systems Corp. ! II CC ! Spokane, WA ! IISSSS CC ! UUCP: ihnp4!tektronix!reed!iscuva!jimc ! II CCCCCC ! (509)927-5757 +----------------+ "With excitement like this, who is needing enemas?"
james@reality1.UUCP (james) (10/13/86)
In article <2586@watdcsu.UUCP>, magore@watdcsu.UUCP (M.A.Gore - ICR) writes: > Note that the 68k ref. manual states that the 68k > was designed with the 6809 in mind --- so as with Intel for GOOD > market reasons. The connection between the 6809 and 68000 was minor. No effort was made to keep the few bad features of the 6809 in the 68000, whereas the 8088 kept most of the problems of the 8080. The 6809 has an extremely powerful instruction set, one that Intel has only now matched with the 80386. I suppose that the arguement can be made that the 68000 is only a 6809 with larger registers, more of them, and a few additional addressing modes and variants of all of the above. In this sense Intel made more progress with 6809->68000 than Intel did with 8080->8088, but the bottom line is that with the 68000 you don't have all those little quirks that the 8088 has that tells you that it is a direct descendant of a more primitive processor. -- James R. Van Artsdalen ...!ut-ngp!utastro!osi3b2!james "Live Free or Die"
eeproks@gitpyr.gatech.EDU (K. J. Seefried iii) (10/14/86)
<>
I guess I'll add my two (admittedly inexpert) cents to the war between the
680x0 and the iAPX x86 processors. First of all, after limited experience
with both processor groups on various machines, I am a 68k partisan. I feel
that it is a better processor for most (perhaps not all) general computing
purposes. However, when we discuss thing like which would be better to have
in your machine, you can not immediately say that it must be the best processor.
For example, say you are a small buisiness owner. You must decide on a computer
to assist you in running your buisiness. The 68000 class machines are faster
(in general), more powerful and come highly recomended by "power computer users".
What machine does your buisness buy? An IBM (or clone) machine. Why? Because
it is a mature, stable machine with an ungodly amount of software availible
today, not 'in the near future'. There are simply no types of buisiness software
that you cant find in one form or another on a PC type machine.
Look back to the 8-bit days. The 6809 was generally considered the most capable
8-bit processor around. What kind of machine did people buy? One of the 808x/z80
machines. Again this was because of the vast amounts of software availible for those
machines. And don't let us forget the Apple ][. Pathetic little processor, huge
amount of software, lots of happy owners.
Again let me say that the 680x0 processors are the best, but when you discuss the
relative merits of one chip over the other, think of who will use it. After all,
The most powerful machines around are only as good as the software that drives them.
For most general computer users, the best chip for them is the 80x86 if for no other
reason than they can run more useful software right now than with any other machine.
And as much as I hate to say it the letters IBM on a machine can mean a whole lot.
------------------------------------------------------------------------
K. J. Seefried iii
P.O. Box 30104, Georgia Insitute of Technology, Atlanta Georgia, 30332
...!{akgua,allegra,amd,hplabs,ihnp4,seismo,ut-ngp}!gatech!gitpyr!eeproksclif@intelca.UUCP (Clif Purkiser) (10/14/86)
> > The fact that IBM wanted to use the 68k for it's microprocessor > is a clue as to which one is better. Can you think of any major > computer out since the 68000 that (except for compatibility reasons) > doesn't use a 68000. Alot of the software for a 80386 is going to be > taken, without rewrite, from a 8088. That software is going to use the > 'brain damaged register set' and assume a segmented archeture. > > The 68000 is the wave of the future. > > > We're not big, but we will be. What the definition of a major computer? In the 68000 world I can only think of a few: Mac, Amiga, Atari that have sold 50K+ units. I would of course agree that Sun, and Apollo and several other workstation companies make "major computers". I think your questions is "what computers use the 8086 architecture other than MS-DOS machines?" The answer is quite a few. Convergent Technologies: Several hundred thousands NGEN workstations use the 80186, and the 80286. (Incidentally they run CT's OS CTOS not MS-DOS this makes CTOS more popular than Unix ) Altos: sells thousands of 80x86 based computers running Xenix. Daisy: uses the 80286 for its engineering workstations These three companies to the best of my knowledge all used the 80x86 architecture eventhough the 68000 was available. In addition their are scores if not hundreds of companies that use 80x86 that don't make PC clones. Prime, ICL, Nixdorf, Compugraphic, and SCI are a few that come immediately to mind. However, I don't know if they choose the 80x86 prior to the availability of the 68000. My apologies to all of the companies I didn't list. -- Clif Purkiser, Intel, Santa Clara, Ca. {pur-ee,hplabs,amd,scgvaxd,dual,idi,omsvax}!intelca!clif I guess I better put back the old disclaimer line. These views are my own property. However anyone who wants them can have them.