rodney@gitpyr.UUCP (RODNEY RICKS) (09/21/86)
In the September 18, 1986 issue of Electronic Design magazine, there is an
article on page 27 describing the new microprocessor from Motorola, the 68030.
The chip is said to run software 20% to 30% faster than the 68020. The chip
is supposed to be available in mid-1987, along with an improved math
coprocessor, the 68882, which will be up to 25% faster than the 68881.
Oh yeah, by the way, the chip will have a built-in, paged memory-management
unit (in case anybody cares :-)). The MMU is a subset of the 68551 PMMU.
The chips have several other interesting features, which I will leave to the
article, so that I won't be accused of plagaurism (sp?).
The bus data transfer rate of the 68030 is 40 Mbytes/sec.
Hopefully, now we won't have to put up with the Intel fans (short
for fanatics? :-)) telling us that Intel has the best microprocessor.
Now, when will I be able to connect one to my Amiga?
Rodney Ricks,
(Not officially respresenting...) The 64 Store. Atlanta, Georgia 30339
UUCP: ...!{akgua,allegra,amd,hplabs,ihnp4,seismo,ut-ngp}!gatech!gitpyr!rodney
or : !gatech!gt-oscar!rodney
Mail: 4265 Hidden Valley Dr. College Park, Ga. 30349tomj@oakhill.UUCP (Tom Johnson) (09/25/86)
In article <2270@gitpyr.UUCP> rodney@gitpyr.UUCP (RODNEY RICKS) writes: >In the September 18, 1986 issue of Electronic Design magazine, there is an >article on page 27 describing the new microprocessor from Motorola, the 68030. >The chip is said to run software 20% to 30% faster than the 68020. The chip >is supposed to be available in mid-1987, along with an improved math >coprocessor, the 68882, which will be up to 25% faster than the 68881. > >Rodney Ricks The Electronic Design article you reference contained a misprint. The 68030 is *not* 20-30% faster than the 68020, it *is* 50- 75% faster than the 68020 (at equivalent clock speeds). Also, the 68882 is *not* 25% faster than the 68881, it *is* 100-300% faster (2x-4x). tom johnson ut-sally/oakhill/tomj
cmcmanis@sun.uucp (Chuck McManis) (09/25/86)
> In the September 18, 1986 issue of Electronic Design magazine, there is an > article on page 27 describing the new microprocessor from Motorola, the 68030. ... edited out ... Ah yes another volley in the "Well this stuff is OK but look at what we will have next year!" contest. Anyone for starting a "Best Announced but not yet available product" award? -- --Chuck McManis uucp: {anywhere}!sun!cmcmanis BIX: cmcmanis ARPAnet: cmcmanis@sun.com These opinions are my own and no one elses, but you knew that didn't you.
swalton@well.UUCP (Stephen R. Walton) (09/26/86)
In article <2270@gitpyr.UUCP> rodney@gitpyr.UUCP (RODNEY RICKS) writes: >In the September 18, 1986 issue of Electronic Design magazine, there is an >article on page 27 describing the new microprocessor from Motorola, the 68030. >The chip is said to run software 20% to 30% faster than the 68020. The chip >is supposed to be available in mid-1987, along with an improved math >coprocessor, the 68882, which will be up to 25% faster than the 68881. > I hope the '82 is more like twice as fast as the 68881. >The bus data transfer rate of the 68030 is 40 Mbytes/sec. Think about this. That's a memory access time of 25 nanoseconds, more than 4 times as fast as the memory we use in our PC's, Macs, Amigas, and Apples. >Now, when will I be able to connect one to my Amiga? Probably soon. But it won't be noticably faster than a CSA 68020. An editorial on the subject of the 68040 (yes, 40) in a recent issue of a small PC magazine commented that these chips are so fast that a fair amount of high-speed cache memory is absolutely essential to take full advantage of them. And that means minicomputer pricing, not micros. Unless one of you hot-shot chip designers out there can put together a 1 MB DRAM with 25 ns access which will sell for less than $100 apiece :-> Stephen Walton, speaking for myself Dudley Moore: You must be a nut case! Peter Cook: They said that about Galileo! They said that about Einstein! Moore: Yeah, well they've said it about a lot of nut cases, too. -from the movie "Bedazzled"
dillon@CORY.BERKELEY.EDU (Matt Dillon) (09/26/86)
Anybody know if the Amiga's OS can handle a 68030? If it does indeed have an MMU on-chip, and were interfacable.. <*place your wishes here*>. -Matt
campbell@sauron.UUCP (Mark Campbell) (09/26/86)
In article <7637@sun.uucp> cmcmanis@sun.uucp (Chuck McManis) writes: >> In the September 18, 1986 issue of Electronic Design magazine, there is an >> article on page 27 describing the new microprocessor from Motorola, the 68030. > ... edited out ... > >Ah yes another volley in the "Well this stuff is OK but look at what we will >have next year!" contest. Anyone for starting a "Best Announced but not >yet available product" award? Okay...the Sun 3/200. (or did you not mean *generally* available?) -- Mark Campbell Phone: (803)-791-6697 E-Mail: !ncsu!ncrcae!sauron!campbell
rick@seismo.CSS.GOV (Rick Adams) (09/26/86)
In article <7637@sun.uucp>, cmcmanis@sun.uucp (Chuck McManis) writes: > Ah yes another volley in the "Well this stuff is OK but look at what we will > have next year!" contest. Anyone for starting a "Best Announced but not > yet available product" award? Well, we ordered the several Floating Point Accelerators from Sun back in August 1985 (as soon as they were announced). We still haven't seen them, nor heard a ship date. I've heard that they were shipping 5 or 6 times now. We have seen great benchmark results from Sun though.... I vote for the sun FPA, it looks great on paper. (The GKS product is also unobtainable, despite being in the catalog for about a year. However, it is no where near as impressive as the FPA.) I talked to someone who actually had his hands on one, so they may be really shipping them now. He may have had a beta test version though. I wouldn't call them available until you can actually get one without waiting a year. (Hmm. A year is about what you are complaining about with Motorola) ---rick P.S Aside from the FPA and GKS vaporware, we are happy Sun customers. (Although I wish they would quit gratuitiously changing the way programs worked. My latest example is that repquota no longer prints disk usage for people without quotas UNLESS you specify a flag. It would have been backwards compatible to have a flag to supress printing disk usage for people with no quotas. Oh well, it only took me an hour to figure out way the quota system "wasn't working" anymore. I hate surprises.)
dillon@CORY.BERKELEY.EDU (Matt Dillon) (09/27/86)
>From: swalton@well.UUCP (Stephen R. Walton) >>The bus data transfer rate of the 68030 is 40 Mbytes/sec. > >Think about this. That's a memory access time of 25 nanoseconds, more than >4 times as fast as the memory we use in our PC's, Macs, Amigas, and Apples. 100ns access time @ 40 Mbytes/sec (remember: 32 bit data bus). I think that the quickest transfer is 64 Mbytes/sec, which would come out to about 62ns access time. This means that your rams actually have to be a bit faster when you take into account bus buffering. It isn't unreasonable, and I think you'd be able to (finally) take advantage of dynamic-ram's block-read ability. I especially like the fact that they do MMU address translation in parallel with checking the cache. Anybody know what kind of page-table cache the MMU has? -Matt
hamilton@uiucuxc.CSO.UIUC.EDU (09/28/86)
>>The bus data transfer rate of the 68030 is 40 Mbytes/sec. > >Think about this. That's a memory access time of 25 nanoseconds, more than >4 times as fast as the memory we use in our PC's, Macs, Amigas, and Apples. thinking about this, i suspect that 40MB/sec is achieved with 32-bit transfers, one every 100 nanoseconds. wayne hamilton U of Il and US Army Corps of Engineers CERL UUCP: {ihnp4,pur-ee,convex}!uiucdcs!uiucuxc!hamilton ARPA: hamilton%uiucuxc@a.cs.uiuc.edu USMail: Box 476, Urbana, IL 61801 CSNET: hamilton%uiucuxc@uiuc.csnet Phone: (217)333-8703 CIS: [73047,544] PLink: w hamilton
hadeishi@husc4.harvard.edu (mitsuharu hadeishi) (09/28/86)
In article <2270@gitpyr.UUCP> Rodney Ricks writes: >The chip is said to run software 20% to 30% faster than the 68020. I believe I read a press release about the 68030 in which it was stated that the 68030 was capable of 8 MIPS, approximately 8 times that of the VAX 11/780, and 2 times that of the 80386. This is at a clock rate of 16-20 Mhz, I'm not sure which. The 68020 is capable of about 1.5 MIPS at a clock rate of 14 Mhz. This puts the 68030 at about 5 times faster than the 68020 at the same clock rate. Apparently the 68030 uses what is called "Harvard parallel architecture". Now that Motorola has released this chip, they have in *every category of chip* a far superior chip than does Intel. 68030 > 80386, 68020 > 80286, 68010 and 68000 >> 80186, 8086, 68008 > 8088. The 68000 series chips are in every case more orthogonally designed, faster, more compatible with each other, and easier to program. -Mitsu
michaelk@copper.UUCP (Michael Kersenbrock) (09/28/86)
>In article <2270@gitpyr.UUCP> rodney@gitpyr.UUCP (RODNEY RICKS) writes: >>In the September 18, 1986 issue of Electronic Design magazine, there is an >>article on page 27 describing the new microprocessor from Motorola, the 68030. >>The chip is said to run software 20% to 30% faster than the 68020. The chip >>is supposed to be available in mid-1987, along with an improved math >>coprocessor, the 68882, which will be up to 25% faster than the 68881. >> > >I hope the '82 is more like twice as fast as the 68881. > >>The bus data transfer rate of the 68030 is 40 Mbytes/sec. > >Think about this. That's a memory access time of 25 nanoseconds, more than >4 times as fast as the memory we use in our PC's, Macs, Amigas, and Apples. > Um....your 8-bit uP background is showing. If the 68030 is handling data at 40 MBytes/sec, that's 100 nS per bus cycle time. The '030, like the '020 is 32-bit machine -- it does 4-bytes at a time. >>Now, when will I be able to connect one to my Amiga? > >Probably soon. But it won't be noticably faster than a CSA 68020. >An editorial on the subject of the 68040 (yes, 40) in a recent issue of >a small PC magazine commented that these chips are so fast that a fair >amount of high-speed cache memory is absolutely essential to take full >advantage of them. And that means minicomputer pricing, not micros. >Unless one of you hot-shot chip designers out there can put together a >1 MB DRAM with 25 ns access which will sell for less than $100 apiece :-> Well, 1M Drams are normally 100nS, so even with the "normal" chips, it may get by with only a few wait-states. In any case, one of the reasons that the '030 is supposed to be faster is that it has a built-in data-cache in addition to the program-cache that the '020 has. > > Stephen Walton, speaking for myself > >Dudley Moore: You must be a nut case! >Peter Cook: They said that about Galileo! They said that about Einstein! >Moore: Yeah, well they've said it about a lot of nut cases, too. > -from the movie "Bedazzled" -- Mike Kersenbrock Tektronix Computer Aided Software Engineering Aloha, Oregon
radford@calgary.UUCP (Radford Neal) (09/29/86)
In article <1837@well.UUCP>, swalton@well.UUCP (Stephen R. Walton) writes: > >The bus data transfer rate of the 68030 is 40 Mbytes/sec. > > Think about this. That's a memory access time of 25 nanoseconds, more than > 4 times as fast as the memory we use in our PC's, Macs, Amigas, and Apples. If the 68030 is anything like the 68020, you probably only need 100ns cycle time memories, since the bus can be four bytes wide. The access time will have to be less, though, to allow for setting up the address. > >Now, when will I be able to connect one to my Amiga? > > Probably soon. But it won't be noticably faster than a CSA 68020... > An editorial on the subject of the 68040 (yes, 40) in a recent issue of > a small PC magazine commented that these chips are so fast that a fair > amount of high-speed cache memory is absolutely essential to take full > advantage of them. And that means minicomputer pricing, not micros. > Unless one of you hot-shot chip designers out there can put together a > 1 MB DRAM with 25 ns access which will sell for less than $100 apiece :-> I think it should be possible to build a small, fast, cheap machine using the 68030 at nearly full speed. The trick would be to not use cache memory, just make it all fast. You only need 16 8KX8 static RAM's to make up 128K of memory. So a *really* fast 128K Mac ought to be possible. The question is, does anyone want one? Or would they rather have 2M of slower memory for the same price? Radford Neal The University of Calgary
daveh@cbmvax.cbm.UUCP (Dave Haynie) (09/29/86)
> Xref: cbmvax net.micro.68k:327 net.micro.amiga:2836 net.micro.atari16:2165 net.micro.mac:2966 > > In article <262@husc6.HARVARD.EDU> hadeishi@husc4.UUCP (mitsuharu hadeishi) writes: > >>Apparently the 68030 >>uses what is called "Harvard parallel architecture". Now that Motorola >>has released this chip, they have in *every category of chip* a far >>superior chip than does Intel. 68030 > 80386, 68020 > 80286, 68010 and >>68000 >> 80186, 8086, 68008 > 8088. The 68000 series chips are in every case >>more orthogonally designed, faster, more compatible with each other, >>and easier to program. > > I agree that the Motorola chips are nicer in many ways than the Intel chips. > But to say Motorola has "released" the chip is stretching things a bit. Not > only has the 68030 not been formally introduced as a product yet, but the > artciles I've read (e.g., Electronics, September 18, 1986) say volume > production is more than a year away. The 80386 is here now, and actually > has a system available (Compaq) which uses it. In other words, it's not > really fair to compare that which Intel has now with what Motorola will have > in a year. A more fair comparison is between the 68020 and the 80386. > > Rich Simoni Yea, they aren't expecting first silicon on the 68030 until next spring or so, last I heard. And sure you CAN get a '386 system, like the Compaq, today, or a 68020 system (a "Turbo" Amiga is in the same price range). But instead of comparing virtual performance, the actual performance should really be compared. Today's MS-DOS based 80386 system, like the '286 systems that preceed it, are still emulating an 8088, still have software than only supports 640K or so of normally mapped memory, etc. You can add UN*X for a better throughput. The "Turbo" Amiga, and Amiga with a 14.4MHz 68020 board and some fast 32 bit RAM will work much better with the "base" system software. Every progam that runs on the Amiga has the potential to use the whole 4 Gigabyte address space along with the faster speed and wider bus. Also, some run-time loaded Amiga libraries can be replaced with libraries that take advantage of the 68020's additional op-codes, 68881 coprocessor, etc. Still, programs will have to be recompiled with a 68020 compiler to take full advantage of the processor, and of course, UN*X could be added. I think a close comparison of these two systems would indicate the current "off-the-shelf" winner in the Motorola-Intel battle, at least for the $3000 to $5000 price range. We'll have to wait for an 80386 based workstation, to compare against a Sun-3 or the like, for the title in that price range. -- ============================================================================ Dave Haynie {caip,ihnp4,allegra,seismo}!cbmvax!daveh These opinions are my own, though if you try them out, and decide that you really like them, a small donation would be appreciated.
kds@mipos3.UUCP (Ken Shoemaker ~) (09/29/86)
I can't speak for Intel, and none of this should be taken as the opinion
of Intel, but...
...Motorola's "announcement" came in the same week that they chided
Intel for the 386's non-availability; the week that Compaq introduced
their box. Compaq must have gotten at least a few from us, since they
seem to have filled the channels of distribution pretty well, or at least
out here Computerland seems to have it, and Fry's is advertising it
(for a discount!) in their newspaper ads.
To put things in a historical sense, Motorola held the same kind of blitz
a few months after Intel came out with the 286, describing the 68020,
the 68881(?) floating point unit, and the whatever-the-number-is paged
MMU, all well before their general availiblity, or even their first
silicon. I suppose that Intel should feel flattered in that it seems
to me that Mot, in each case, has conceeded defeat: the 68010 to the
286, the 68020 to the 386, and that in order to maintain any kind of
market share, they have to have a media blitz, rolling out their newest,
half baked, family jewels. Its all pretty comic!
And what about the new family jewels? Should I feel vindicated that Mot
has finally decided that on-chip memory management is a good idea? And
what about performance? Why would a new, ultra-high end user choose
the 68030 over, say, the Clipper, or the MIPS chips, each of which
(if you believe the numbers) should exceed the 68030's performance (with
the promise of delivering more sooner) and are available, in some semblance,
today?
--
The above views are personal.
I've seen the future, I can't afford it...
Ken Shoemaker, Microprocessor Design, Intel Corp., Santa Clara, California
uucp: ...{ hplabs|amdcad|qantel|pur-ee|scgvaxd|oliveb }!intelca!mipos3!kds
csnet/arpanet: kds@mipos3.intel.comsimoni@Shasta.STANFORD.EDU (Richard Simoni) (09/29/86)
In article <262@husc6.HARVARD.EDU> hadeishi@husc4.UUCP (mitsuharu hadeishi) writes: >Apparently the 68030 >uses what is called "Harvard parallel architecture". Now that Motorola >has released this chip, they have in *every category of chip* a far >superior chip than does Intel. 68030 > 80386, 68020 > 80286, 68010 and >68000 >> 80186, 8086, 68008 > 8088. The 68000 series chips are in every case >more orthogonally designed, faster, more compatible with each other, >and easier to program. I agree that the Motorola chips are nicer in many ways than the Intel chips. But to say Motorola has "released" the chip is stretching things a bit. Not only has the 68030 not been formally introduced as a product yet, but the artciles I've read (e.g., Electronics, September 18, 1986) say volume production is more than a year away. The 80386 is here now, and actually has a system available (Compaq) which uses it. In other words, it's not really fair to compare that which Intel has now with what Motorola will have in a year. A more fair comparison is between the 68020 and the 80386. Rich Simoni Center for Integrated Systems Stanford University simoni@sonoma.stanford.edu ...!decwrl!glacier!shasta!simoni
tomj@oakhill.UUCP (Tom Johnson) (09/30/86)
In article <1837@well.UUCP> swalton@well.UUCP (Stephen R. Walton) writes: > >>The bus data transfer rate of the 68030 is 40 Mbytes/sec. > >Think about this. That's a memory access time of 25 nanoseconds, more than >4 times as fast as the memory we use in our PC's, Macs, Amigas, and Apples. >... >Unless one of you hot-shot chip designers out there can put together a >1 MB DRAM with 25 ns access which will sell for less than $100 apiece :-> > > Stephen Walton, speaking for myself Try again, just because there is a 40MB/s bus transfer rate *does not* mean that there is a 25 ns bus cycle! If the processor transfers 4 bytes/xfer, then the bus transfer rate will be 40 MB/s on a **100** ns bus cycle time. If the transfer is 2 bytes/xfer then the bus cycle would be 50ns, etc. Additionally, your assumption does not take into account capabilities such as those on the 68030 to support new DRAM technology (page mode, static column, nibble mode) which allows the transfer of 4 long words (16 bytes) in as little as 5 clock cycles. tom johnson inhp4!ut-sally!oakhill!tomj
tim@ism780c.UUCP (Tim Smith) (09/30/86)
In article <807@cbmvax.cbmvax.cbm.UUCP> daveh@cbmvax.cbm.UUCP (Dave Haynie) writes: > [Talking about "Turbo" Amiga] > etc. Still, programs will have to be recompiled with a 68020 compiler to > take full advantage of the processor, and of course, UN*X could be added. Does the "Turbo" Amiga have an MMU? If not, then adding UN*X is going to be very hard. Adding unix to the Compaq 386 box will be very easy. -- What's the difference between a duck? Tim Smith USENET: sdcrdcf!ism780c!tim Compuserve: 72257,3706 Delphi or GEnie: mnementh
hadeishi@husc4.harvard.edu (mitsuharu hadeishi) (09/30/86)
In re: Motorola's continual announcements . . . is it fair to compare the 68030 to the 80386, etc.? Yes. Firstly, it is clear that Motorola's architecture is a lot cleaner than Intel's. Certainly their chips may be a year or so behind Intel (i.e. 68020 and 80286, 68010 and 80186, etc.); but their chips are far superior. Even if you insist on comparing the 68010 to the 80286 it is not clear at all that the 80286 wins hands down; the '010 has an architecture more well-suited to implementation of multi-level interrupt multitasking systems. The fact is that for implementation of real systems the 68010 in many cases is a far more appropriate choice than the 80286. (about Unix implementation, see BYTE article last month.) And of course the 68020 is far and away the superior chip compared to the 80286, and the '020 has been available in real computer systems for some time. From this point of view it seems reasonable to compare the '286 to the '020 rather than the '010, and here the comparison is clear. '020s are used (as you all know) in Sun workstations, TurboAmigas, etc. '286s are used in ATs. There is really no comparison between the two classes of machine. Another important point to make is that operating systems implemented on the 68000 series chips will run fairly well on the higher-grade chips; UNIX, AmigaDos/Intuition, Mac Operating System, ST GEM/TOS, et cetera are all capable of taking advantage of the higher-level chips' capabilities (especially UNIX and AmigaDos/Intuition), although some optimization wouldn't hurt. From the point of view of the Intel chips, the 8088-based MS-DOS is nothing more than a kludge on the 80286 and 80386; for that operating system to take advantage of those chips a complete re-write must be done, and is in the process of being completed as we speak. This new operating system however will most likely be incompatible to a great degree with the old, 8/16-bit MS-DOS and will contain major bugs for some time. AmigaDOS/Intuition has been running on the 68000 for about a year now, and Unix implementations for the 68000/68010 chips have been floating around for at least 3 or more years. To compare the 80386 with the 68020 is absurd; the '020 has been around for some time now and has a lot of debugged software/operating systems running on it; the '030 will inherit that software and blaze with it whereas what will the '386 do? Run character-stream UNIX and 640K MS-DOS? -Mitsu (hadeihi@husc4.UUCP) P.S. BTW, have you guys heard about the 78000 (!?) See Nanobytes in the latest BYTE (the //GS issue). This is a RISC uprocessor with a rated speed of 20 (VAX-equivalent?) MIPS . . . this kind of speed totally blows away the RT PC (note: the RT PC performed just a hair better than an PC AT in benchmarks . . . see PC World or PC magazine of a month ago. . . not very impressive . . .)
hamilton@uiucuxc.CSO.UIUC.EDU (09/30/86)
hadeishi@husc4.harvard.edu sez: >Now that Motorola >has released this chip, they have in *every category of chip* a far >superior chip than does Intel. 68030 > 80386, 68020 > 80286, 68010 and >68000 >> 80186, 8086, 68008 > 8088. The 68000 series chips are in every case >more orthogonally designed, faster, more compatible with each other, >and easier to program. i agree with your sentiment, but i think you have the chips mismatched. it should be more like: 68030 > ???, 68020 > 80386, 680[01]0 > 80286. wayne hamilton U of Il and US Army Corps of Engineers CERL UUCP: {ihnp4,pur-ee,convex}!uiucdcs!uiucuxc!hamilton ARPA: hamilton%uiucuxc@a.cs.uiuc.edu USMail: Box 476, Urbana, IL 61801 CSNET: hamilton%uiucuxc@uiuc.csnet Phone: (217)333-8703 CIS: [73047,544] PLink: w hamilton
braner@batcomputer.TN.CORNELL.EDU (braner) (10/01/86)
[] Gee, that was to much to ignore... Can anybody point me to the numerous powerful multi-user super-boxes that are using the '286 or '386 (AT doesn't count since it is actually used by single users, and in 8088 emulation mode, primarily)? As far as I can tell, designers that are not REQUIRED by their employers to use Intel chips have long used the 68K family for that purpose! If you want REALLY high-end computing power, you get a CRAY and such. If you want a real big bang per buck, you get a 68K personal machine. Unless, of course, you only feel comfortable in the crowd, the crowd that will use 8080 software forever (they think). - Moshe Braner BTW: the ST + 32081 math coprocessor beats (for number crunching) the portion of the power of the VAX I work on that actually labors for me. Price? still < $1000
wsr@lmi-angel.UUCP (Wolfgang Rupprecht) (10/01/86)
In article <> hadeishi@husc4.UUCP (mitsuharu hadeishi) writes: >[random discussion about 68030]... Now that Motorola >has released this chip [well, no, they will release it in a year from >now.. -wsr], they have in *every category of chip* a far >superior chip than does Intel. 68030 > 80386, 68020 > 80286, 68010 and >68000 >> 80186, 8086, 68008 > 8088. The 68000 series chips are in every case >more orthogonally designed, faster, more compatible with each other, >and easier to program. Up to now I have always hated intel chips. In general they are *so* contorted/poorly thought out, (some would even say buggy). Up to now that is. I'm ready to to believe in 80386's. Lets face it, the lack of *usable* memory management for the 68000 is laughable. The 68020 mmu is still vapor-hardware. When it finally arrives, one gets a chip that has to have *lots* of logic around it just to allow it to work in the same circuits that the 68020 used to work in. It will run *much* slower than the 68020 however. Remember, the 68020 is in the same generation as the 80386. (lets hear some applause for intel.) -- Wolfgang Rupprecht {harvard|decvax!cca|mit-eddie}!lmi-angel!wsr
farren@hoptoad.uucp (Mike Farren) (10/01/86)
In article <280@husc6.HARVARD.EDU> hadeishi@husc4.UUCP (mitsuharu hadeishi) writes: > >'020s are used (as you all know) in Sun workstations, TurboAmigas, etc. >'286s are used in ATs. There is really no comparison between the two >classes of machine. That's true - there's over 10 times as many '286 machines out there. There really ISN'T a comparison... > > Another important point to make is that operating systems implemented on >the 68000 series chips will run fairly well on the higher-grade chips; >UNIX, AmigaDos/Intuition, Mac Operating System, ST GEM/TOS, et cetera are >all capable of taking advantage of the higher-level chips' capabilities Depends entirely on the foresight and ability of the systems programmer (about whom, in most cases, I have grave doubts). Mac OS dies on anything higher than a 68000... > >'386 do? Run character-stream UNIX and 640K MS-DOS? You betcha. At the same time. Quickly. With graphics additions as hardware/ software becomes available. I have said before and I will say again now - religious chip wars are useless and S*T*U*P*I*D. All of the debate and struggle gains you not one thing. Personally, I'd rather be aware of and comfortable with as many of the different architectures/chips/systems available - it makes me that much more employable (and that much more affluent). When I look at what power I have sitting on my desk now, IBM *OR* Amiga or whatever, and compare it with the most powerful computers available when I started in this business, I find it really hard to bitch too much about a chip - even, god help me, the 6502. -- ---------------- "... if the church put in half the time on covetousness Mike Farren that it does on lust, this would be a better world ..." hoptoad!farren Garrison Keillor, "Lake Wobegon Days"
david@randvax.UUCP (David Shlapak) (10/01/86)
> I can't speak for Intel, and none of this should be taken as the opinion > of Intel, but... > followed by a couple pages of Motorola bashing on the part of an Intel > employee... Look, after all the Amiga-vs-Mac stuff we've had around here the last thing we need is another "My dog's better than your dog" "debate" on this net. Maybe we need a net.penis.length to contain these discussions. --- das "Let's try to keep our reginalistic emotions under our big, ugly hats, shall we?" --- Milo Bloom
stever@videovax.UUCP (Steven E. Rice, P.E.) (10/01/86)
In article <1837@well.UUCP>, Stephen R. Walton (swalton@well.UUCP) writes: >> The bus data transfer rate of the 68030 is 40 Mbytes/sec. > > Think about this. That's a memory access time of 25 nanoseconds, more than > 4 times as fast as the memory we use in our PC's, Macs, Amigas, and Apples. Whoa!!!!! The 68020 (and presumably the 68030) transfers long words (32 bits, 4 bytes) on its 32-bit data bus. What you are tripped up by is specmanship! 40 Mbytes/sec is 10 Mlong-words/sec. This comes to 100 nsec, which is fast, but doable. Steve Rice ---------------------------------------------------------------------------- {decvax | hplabs | ihnp4 | uw-beaver}!tektronix!videovax!stever
olson@endor.harvard.edu (Eric Olson) (10/01/86)
In article <783@oakhill.UUCP> tomj@oakhill.UUCP (Tom Johnson) writes: >In article <1837@well.UUCP> swalton@well.UUCP (Stephen R. Walton) writes: >> >>>The bus data transfer rate of the 68030 is 40 Mbytes/sec. >> >>Think about this. That's a memory access time of 25 nanoseconds, more than >>4 times as fast as the memory we use in our PC's, Macs, Amigas, and Apples. >>... >> Stephen Walton, speaking for myself > >Try again, just because there is a 40MB/s bus transfer rate *does not* mean >that there is a 25 ns bus cycle! If the processor transfers 4 bytes/xfer, >then the bus transfer rate will be 40 MB/s on a **100** ns bus cycle time. >... >tom johnson inhp4!ut-sally!oakhill!tomj The 68020 (and presumably, the 030) has input lines asking the external device (memory, peripheral, etc) how wide it is. That way the processor can ask for a long from a byte-wide device and do 4 reads after the device says "but I'm only a little bitty byte wide". But I would be frightened away from an architecture where a true 32 bit processor has byte-wide main memory. Still, you build the machine with the fastest memory you can afford, and if it ain't fast enough, use wait states. Faster processor speed is still significant, especially in a processor with an on-chip cache. Remember: memory halfs in cost and doubles in speed every n years (where n used to be 5, and seems to be approaching 1). I realize there is a theoretical limit, until we all wise up and start using transphasors (transistors that switch laser light). This brings up an interesting point. In the mid-70's, everyone started making processors with MORE instructions that do MORE processing each. This was, in part, due to memory bandwidth limitations. The new genre of RISC machines is going exactly the other way: processors with FEWER instructions, each of which does LESS processing, because memory bandwidth has improved. The RISC concept allows more specific manipulation of processing power (and therefore requires more intelligent compilers, to take advantage of the quantization of the processing unit). When's the 78000 due out? -Eric
kds@mipos3.UUCP (Ken Shoemaker ~) (10/02/86)
In article <280@husc6.HARVARD.EDU> hadeishi@husc4.UUCP (mitsuharu hadeishi) writes: > > In re: Motorola's continual announcements . . . is it fair >to compare the 68030 to the 80386, etc.? > > Yes. > > Firstly, it is clear that Motorola's architecture is a lot >cleaner than Intel's. Certainly their chips may be a year or so behind >Intel (i.e. 68020 and 80286, 68010 and 80186, etc.); but their chips are >far superior. Even if you insist on comparing the 68010 to the 80286 >it is not clear at all that the 80286 wins hands down; the '010 has an >architecture more well-suited to implementation of multi-level interrupt >multitasking systems. The fact is that for implementation of real >systems the 68010 in many cases is a far more appropriate choice than >the 80286. (about Unix implementation, see BYTE article last month.) >And of course the 68020 is far and away the superior chip compared >to the 80286, and the '020 has been available in real computer systems >for some time. From this point of view it seems reasonable to compare the >'286 to the '020 rather than the '010, and here the comparison is clear. >'020s are used (as you all know) in Sun workstations, TurboAmigas, etc. >'286s are used in ATs. There is really no comparison between the two >classes of machine. Of course, there is no comparison between the two classes of machines, because you in here introduce more than two classes of machines. I would not put Sun workstations and Turbo Amigas in the same class of machine simply because the Amiga does not support any kind of memory management. If you want to have a grown-up machine, one that supports multi-user/multi-tasking you really need to have this. I have used 68k boxes without that have tried to do it, and it works, but you are just asking for trouble. That the same 286-based box can support an unprotected os like ms-dos and a protected os like Unix means that it can function in either of these two classes. Not until you have the 68030, or you make an entire subsystem plug in with a 68020 and an mmu, can you claim to have upgraded your Amiga the way plugging a 286 upgrades a pc. > Another important point to make is that operating systems implemented on >the 68000 series chips will run fairly well on the higher-grade chips; >UNIX, AmigaDos/Intuition, Mac Operating System, ST GEM/TOS, et cetera are >all capable of taking advantage of the higher-level chips' capabilities >(especially UNIX and AmigaDos/Intuition), although some optimization >wouldn't hurt. From the point of view of the Intel chips, the 8088-based >MS-DOS is nothing more than a kludge on the 80286 and 80386; for that >operating system to take advantage of those chips a complete re-write >must be done, and is in the process of being completed as we speak. >This new operating system however will most likely be incompatible to >a great degree with the old, 8/16-bit MS-DOS and will contain major bugs >for some time. AmigaDOS/Intuition has been running on the 68000 for >about a year now, and Unix implementations for the 68000/68010 chips have >been floating around for at least 3 or more years. To compare the 80386 In the first place, msdos, while not taking advantage of the functionality of the 286 or the 386, is certainly not a kludge when run on them. In addition, when you go from one type of 68* based box to another, you also need to rewrite the operating system, because the memory management systems are incompatible. People have also generated 286-based cards that plug into plain old IBM pcs and have gained a significant performance upgrade, while running their old software. I'm sure the same thing can and will happen with the 386. And Unix implementations for 8086-based chips have been floating around for at least 5 years. MS-DOS already is a 16-bit os, and always has been. When going from a 68020 to a 68030, you will have to rewrite the os to take advantage of the memory management that the 68030 provides...otherwise you just have a little faster 68020, much like you can use a 286 as a faster 8086. How about if we compare apples to apples? Can you say "incoherency?" >P.S. BTW, have you guys heard about the 78000 (!?) See Nanobytes in >the latest BYTE (the //GS issue). This is a RISC uprocessor with a >rated speed of 20 (VAX-equivalent?) MIPS . . . this kind of speed >totally blows away the RT PC (note: the RT PC performed just a hair >better than an PC AT in benchmarks . . . see PC World or PC magazine >of a month ago. . . not very impressive . . .) Oh, this is great. Yet another paper product. This one, they haven't even had a press release on! I guess it must be more than a year away (:-)). -- The above views are personal. I've seen the future, I can't afford it... Ken Shoemaker, Microprocessor Design, Intel Corp., Santa Clara, California uucp: ...{ hplabs|amdcad|qantel|pur-ee|scgvaxd|oliveb }!intelca!mipos3!kds csnet/arpanet: kds@mipos3.intel.com
das@well.UUCP (David Shayer) (10/02/86)
U> >>In article <262@husc6.HARVARD.EDU> hadeishi@husc4.UUCP (mitsuharu hadeishi) writes: >>Now that Motorola >>has released this chip, they have in *every category of chip* a far >>superior chip than does Intel. >I agree that the Motorola chips are nicer in many ways than the Intel chips. >But to say Motorola has "released" the chip is stretching things a bit. Not >only has the 68030 not been formally introduced as a product yet, but the >artciles I've read (e.g., Electronics, September 18, 1986) say volume >production is more than a year away. The 80386 is here now, and actually >has a system available (Compaq) which uses it. In other words, it's not >really fair to compare that which Intel has now with what Motorola will have >in a year. A more fair comparison is between the 68020 and the 80386. To say that Compaq has "released" an 80386 based computer is stretching things a bit too, don't you think. They announced it, but said it would be available after January. This is similar to what Motorola announced. By the way, IBM is said to be readying an announcment in January for their 80386 based machine, which will be available in June. And don't forget that a multitasking operating system won't be available until around June either. Until then, these 80386 machines are just big fast PC's. Meanwhile, Apple is readying a 68020 based machine for release early next year. Isn't vaporware fun! ----------------------- David Shayer @ the Well "Most of the time, for most programmers, what a compiler produces is not object code, but error messages."
hr@uicsl.UUCP (10/02/86)
RE: > Think about this. That's a memory access time of 25 nanoseconds, more than > 4 times as fast as the memory we use in our PC's, Macs, Amigas, and Apples. >>If the 68030 is anything like the 68020, you probably only need 100ns >>cycle time memories, since the bus can be four bytes wide. The access time >>will have to be less, though, to allow for setting up the address. Would someone who has access to the product announcement check this?: I thought I read that the chip would be able to handle one of the "funny" modes available on some DRAMs, nibble, page, or some such. My understanding is that this would allow "short cycling" on consecutive memory accesses. How much of a speedup would this give? ---- harold ravlin {ihnp4,pur-ee}!uiucdcs!uicsl!hr
olson@endor.harvard.edu (Eric Olson) (10/02/86)
In article <1165@hoptoad.uucp> farren@hoptoad.UUCP (Mike Farren) writes: >Mac OS dies on anything higher than >a 68000... This isn't true. The Prodigy 4, a Mac running a 16 MHz 68020 w/68881 FPC works just fine. Some things want to run with the cache off (because they contain self-modifying code). I'm not sure whether the 128K Mac ROMs are a requirement for this, but, most definitely, the Mac WILL RUN ON A 68020. -Eric
kds@mipos3.UUCP (Ken Shoemaker ~) (10/03/86)
In article <1870@well.UUCP> das@well.UUCP (David Shayer) writes: > >To say that Compaq has "released" an 80386 based computer is stretching >things a bit too, don't you think. They announced it, but said it would >be available after January. This is similar to what Motorola announced. hardly. They are advertised, and are being sold, at least in this area by Businessland, Computerland, and Fry's. Talking to salesmen at the stores, they are going like hotcakes, but they are getting them in, and selling them to real-live people. -- The above views are personal. I've seen the future, I can't afford it... Ken Shoemaker, Microprocessor Design, Intel Corp., Santa Clara, California uucp: ...{ hplabs|amdcad|qantel|pur-ee|scgvaxd|oliveb }!intelca!mipos3!kds csnet/arpanet: kds@mipos3.intel.com
hadeishi@husc4.harvard.edu (mitsuharu hadeishi) (10/03/86)
In article <glub@Well.UUCP> David Shayer writes: >Until then, these 80386 machines are just big >fast PC's. Meanwhile, Apple is readying a 68020 based machine for >release early next year. Although the Prodigy 4 and Turbo Amiga boxes are available NOW (read: faster than a VAX.) And C-Amiga is readying the Ranger, a 68020-based machine for release early next year as well (to also use a 2-megabyte version of the custom chips, supporting much higher graphics resolutions, non-interlaced hi-res color, AND a 4K per blit limit (up from 1K).) >Isn't vaporware fun! Yup. -Mitsu
keithe@tekgvs.UUCP (10/03/86)
In article <1870@well.UUCP> das@well.UUCP (David Shayer) writes: >By the way, IBM is said to be readying an announcment in January for >their 80386 based machine, which will be available in June. And don't >forget that a multitasking operating system won't be available until >around June either. Meanwhile, Apple is readying a 68020 based machine >for release early next year. > >Isn't vaporware fun! > Making predictions is difficult, especially about the future. keith
tim@ism780c.UUCP (Tim Smith) (10/03/86)
[ followups to net.micro.68k please! ] In article <1870@well.UUCP> das@well.UUCP (David Shayer) writes: > >To say that Compaq has "released" an 80386 based computer is stretching >things a bit too, don't you think. They announced it, but said it would >be available after January. This is similar to what Motorola announced. The Compaq machine may not be available yet, but the 80386 is. For now, the 68030 is just something to be aware of when you design a 68020 system so that you will be able to upgrade when the '30 becomes available. -- member, all HASA divisions POELOD ECBOMB -------------- ^-- Secret Satanic Message Tim Smith USENET: sdcrdcf!ism780c!tim Compuserve: 72257,3706 Delphi or GEnie: mnementh
kim@amdahl.UUCP (Kim DeVaughn) (10/03/86)
In article <1870@well.UUCP>, das@well.UUCP (David Shayer) writes: > To say that Compaq has "released" an 80386 based computer is stretching > things a bit too, don't you think. They announced it, but said it would > be available after January. This is similar to what Motorola announced. Nope. Fry's Electronics (and supermarket) here in Sunnyvale had one in already and sold it (I *think* the salesman said $5995). They said they were expecting "several" more in "any day now". They may not be widely available yet, but they aren't "vaporware" either! /kim -- UUCP: {sun,decwrl,hplabs,pyramid,ihnp4}!amdahl!kim DDD: 408-746-8462 USPS: Amdahl Corp. M/S 249, 1250 E. Arques Av, Sunnyvale, CA 94086 CIS: 76535,25 [ Any thoughts or opinions which may or may not have been expressed ] [ herein are my own. They are not necessarily those of my employer. ]
daveh@cbmvax.cbm.UUCP (Dave Haynie) (10/04/86)
> Keywords: here we go again... > Of course, there is no comparison between the two classes of machines, > because you in here introduce more than two classes of machines. I would not > put Sun workstations and Turbo Amigas in the same class of machine simply > because the Amiga does not support any kind of memory management. If you > want to have a grown-up machine, one that supports multi-user/multi-tasking > you really need to have this. Its nice to have, but required only if your OS demands it. > Not until you have the 68030, or you make an entire subsystem > plug in with a 68020 and an mmu, can you claim to have upgraded your Amiga > the way plugging a 286 upgrades a pc. I don't claim that the 68020 upgrade for a 68000 machine is like a '286 upgrade for an 8088 machine. I claim its a generation removed. The 8088 is an 8/16 bit processor, the '286 a 16/16. The 68000 is a 16/32, the '020 a 32/32. A processor with a good MMU can run UNIX, quickly, and have all the advantages of a protected OS. A non-protected OS can suffer from misbehaved programs and slow context swaps (if its multitasking, which the MS-DOS operating system emphatically isn't. Oranges vs. oranges, remember). A Turbo upgrade for an Amiga can deliver much the same level of performance as a Sun and still use standard off-the-shelf Amiga software and operating system, and it can take advantage of the extra capabilities of the 68020. MS-DOS runs fine on a '286 PC, but it can't use the extended memory addressing capabilities -- unless you kludge it, MS-DOS has a ceiling of 640K. AmigaDOS has a ceiling of 4 Gigabytes. I wouldn't try to run UNIX on a basic Turbo Amiga; UNIX isn't designed to run well without an MMU; the context swapping between processes is going to make it a dog. But the Amiga Exec was designed to provide a fast multitasking environment on a machine without an MMU. It supports 68000, 68010, and 68020 processors in a processor-independent way; the Amiga Exec knows which processor is in place and provides functions to handle the differences, like the different exception stacks. All the AmigaDOS software will work as is with the 68020, or it can be upgraded to take advantage of special 68020 enhancements. These upgrades, however, can be incremental, as each of the many software subsystems in the Amiga Kernal and DOS can be replace independently of each other. How much of the '286 instruction set, etc. will even work under MS-DOS, assuming I were to, say, rewrite the floating point libraries for the '286 and '287 coprocessor? Of course an unprotected multitasking environment is dangerous for poorly behaved programs, but so is a single tasking environment on a machine without adequate hardware support (like a PC). I wouldn't expect to see a good UNIX running on the Amiga without an MMU; then again, an 80286 UNIX has to deal with its own set of inefficiencies, namely the 64K paging limitations that still exist even in protected mode. For a state of the art UNIX environment I need to an an MMU to my Amiga, you need to add a '386 to your PC. An integral/standard MMU is a good idea; Intel finally go one right :-). > In the first place, msdos, while not taking advantage of the functionality > of the 286 or the 386, is certainly not a kludge when run on them. True. Lots of folks think the 8086 modes are kludges; I believe they're not. And the problems with the '286 native mode and MS-DOS are due to both the design of the '286 and the design of MS-DOS. > In addition, when you go from one type of 68* based box to another, you > also need to rewrite the operating system, because the memory management > systems are incompatible. Not helped much by the two different late-coming 680xx MMUs. I know, any standard MMU is better than none at all. > People have also generated 286-based cards that plug into plain old IBM pcs > and have gained a significant performance upgrade, while running their old > software. I'm sure the same thing can and will happen with the 386. That's really only the advantage of going to a 16 bit (or 32 bit) bus versus the old 8 bot of the 8088, plus the faster speed of the thing. But like I said, I can get a much more significant increase in power with the 68020 card in the Turbo Amiga -- double the clock speed, double the data fetch, and extend that memory limit to the full 32 bit address bus. Now I know that this can't be done with every 68000 based micro; many of them do thing that require a 68000 only. That's an OS design issue. And why I'm glad I have an Amiga. > And Unix implementations for 8086-based chips have been floating around for > at least 5 years. MS-DOS already is a 16-bit os, and always has been. > When going from a 68020 to a 68030, you will have to rewrite the os to take > advantage of the memory management that the 68030 provides...otherwise you > just have a little faster 68020, much like you can use a 286 as a faster > 8086. How about if we compare apples to apples? Can you say "incoherency?" I'll leave the "incoherency" to you. You're flaming for running a 68000 with no MMU, then you suggest Unix for 8086-based chips? If I want segmentation as a form of memory mamagement, there's nothing stopping me from requiring my 68000 environment to support only register relative addressing. But anything that _requires_ 64K segments shouldn't be bothering with UNIX. And as I said, my move to an '020 or '030 already gives me much more than just a faster 68000. To use an MMU would require a few sections of the OS to be rewritten, but its no big deal. The hard part is going the other way, which is why memory management is required for UNIX speed, protection issues aside. >>P.S. BTW, have you guys heard about the 78000 (!?) See Nanobytes in >>the latest BYTE (the //GS issue). This is a RISC uprocessor with a >>rated speed of 20 (VAX-equivalent?) MIPS . . . this kind of speed >>totally blows away the RT PC (note: the RT PC performed just a hair >>better than an PC AT in benchmarks . . . see PC World or PC magazine >>of a month ago. . . not very impressive . . .) > > Oh, this is great. Yet another paper product. This one, they haven't > even had a press release on! I guess it must be more than a year > away (:-)). > -- > The above views are personal. > > I've seen the future, I can't afford it... > Ken Shoemaker, Microprocessor Design, Intel Corp., Santa Clara, California > uucp: ...{ hplabs|amdcad|qantel|pur-ee|scgvaxd|oliveb }!intelca!mipos3!kds > csnet/arpanet: kds@mipos3.intel.com -- ============================================================================ Dave Haynie {caip,ihnp4,allegra,seismo}!cbmvax!daveh These opinions are my own, though if you try them out, and decide that you really like them, a small donation would be appreciated.
rodney@gitpyr.UUCP (RODNEY RICKS) (10/04/86)
In article <205@mipos3.UUCP> kds@mipos3.UUCP (Ken Shoemaker ~) writes: > Not until you have the 68030, or you make an entire subsystem >plug in with a 68020 and an mmu, can you claim to have upgraded your Amiga >the way plugging a 286 upgrades a pc. Actually, one can easily support this claim. In both cases, you still have the same level of memory protection as before. Correct me if I'm wrong (and I know you will :-)), but MSDOS doesn't work in the 80286 protected mode, does it? Since this is this case, upgrading to a 286 does not give any more protection when using the standard OS. Sure, you have memory protection for UNIX and its clones, but if that's what you're talking about, then you're comparing apples and oranges (which is quickly becoming a very popular term in this discussion). >>P.S. BTW, have you guys heard about the 78000 (!?) See Nanobytes in > >Oh, this is great. Yet another paper product. This one, they haven't >even had a press release on! I guess it must be more than a year >away (:-)). According to the BYTE article, it is due in the first quarter of next year. Rodney Ricks, (Not officially respresenting...) The 64 Store. Atlanta, Georgia 30339 UUCP: ...!{akgua,allegra,amd,hplabs,ihnp4,seismo,ut-ngp}!gatech!gitpyr!rodney or : !gatech!gt-oscar!rodney Mail: 4265 Hidden Valley Dr. College Park, Ga. 30349
stuart@BMS-AT.UUCP (Stuart D. Gathman) (10/04/86)
In article <200@mipos3.UUCP>, kds@mipos3.UUCP (Ken Shoemaker ~) writes: > silicon. I suppose that Intel should feel flattered in that it seems > to me that Mot, in each case, has conceeded defeat: the 68010 to the . . . > market share, they have to have a media blitz, rolling out their newest, > half baked, family jewels. Its all pretty comic! I enjoy just watching the competition! The one-upmanship of Motorola and Intel makes for a furious rate of advance in technology. Let's just have a clean, fair fight and we all benefit. -- Stuart D. Gathman <..!seismo!{vrdxhq|dgis}!BMS-AT!stuart>
henry@utzoo.UUCP (Henry Spencer) (10/04/86)
> And what about the new family jewels? Should I feel vindicated that Mot > has finally decided that on-chip memory management is a good idea? ... Only when Intel decides that on-chip caches are a good idea. The 68030 has *both*. So does the Clipper, by the way. The MIPS chip set likewise does the MMU and half (the hard half) of the cache. Where's Intel? > ...what about performance? Why would a new, ultra-high end user choose > the 68030 over, say, the Clipper, or the MIPS chips, each of which > (if you believe the numbers) should exceed the 68030's performance (with > the promise of delivering more sooner) and are available, in some semblance, > today? Probably no particularly good reason except compatibility. I notice, however, that the line of argument has shifted slightly: we don't hear the word "Intel" any more. -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,decvax,pyramid}!utzoo!henry
henry@utzoo.UUCP (Henry Spencer) (10/04/86)
> ...Remember, the 68020 is in the same generation as the 80386...
Really? 386-based machines have just barely started shipping. And all
the software runs them as if they were 8088s, or 286s at best if you're
lucky. 68020 machines have been in the field, with software support, for
quite some time now. I'd say the 020 has a half-generation lead.
--
Henry Spencer @ U of Toronto Zoology
{allegra,ihnp4,decvax,pyramid}!utzoo!henrygnu@hoptoad.uucp (John Gilmore) (10/04/86)
[==> Followups have been redirected to net.micro.68k <== If you care
about this topic, please start reading about it there. The mac/amiga/atari
users have enough to talk about without Intel/Motorola flames. I'm just glad
nobody has tried including net.micro.pc, then the flames would really start!]
I believe it is correct to compare the 80386 with the 68020, on a feature
by feature basis. They are both the first 32-bit processor from each
company, have similar cycle and instruction times, addressing capabilities,
etc.
In the past Intel has been first to market each level of chip.
However, they spent too much time at the 16-bit level (on the 186 and
286, both of which took longer than expected), and Motorola worked hard
to get out of the gate first at the 32-bit level. (My favorite "why
IBM chose the 8088" rumor is that Motorola couldn't build them 1/4
million chips the first year and Intel could since they had been
ramping up longer. Be interesting to see if Motorola's taking the lead
at 32 bits makes any difference now.)
My impression is that the relative power of each chip relates strongly
to its design date. Thus, Motorola chips have tended to be more
powerful than the corresponding Intel chips, because they were designed
slightly later. The 80386 should be faster than the 68020, since it is
coming out about a year later. They made a big mistake in leaving out
the cache, though they did put in a neat feature for interleaved memory
access, so it probably runs in roughly the same speed range unless you
give it a fancy memory subsystem. (If a fancy memory is assumed,
though, then the faster possible clock rate on the 68020 comes into
play, maybe making it an even heat again.) The 68030 will be faster
than both, since it will be yet another year later. And so on...
I'm not going to argue the merits of Intel versus Motorola. It is clear
to me that both have learned from each other. Now it is up to us, the
users, to learn to write our stuff portably so we can move it to the machine
of whoever is learning the fastest at each point in time.
--
John Gilmore {sun,ptsfa,lll-crg,ihnp4}!hoptoad!gnu jgilmore@lll-crg.arpa
May the Source be with you!henry@utzoo.UUCP (Henry Spencer) (10/04/86)
> ... When going from a 68020 to a 68030, you will have to > rewrite the os to take advantage of the memory management that the 68030 > provides...otherwise you just have a little faster 68020, much like you > can use a 286 as a faster 8086. How about if we compare apples to apples? Okay, we'll compare apples to apples: how's the 32-bit-mode code generator for the 386 coming? If you don't have that, you just have a little faster 80286. Rewriting the code generator and recompiling everything strikes me as rather more work than hacking up the MMU section of the kernel. *Especially* since the 286->386 conversion is also going to mean hacking up the MMU stuff to cope with 32-bit addresses. -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,decvax,pyramid}!utzoo!henry
hadeishi@husc4.harvard.edu (mitsuharu hadeishi) (10/04/86)
In article <1177@hoptoad.uucp> John Gilmore writes: >[==> Followups have been redirected to net.micro.68k <== If you care >about this topic, please start reading about it there. The mac/amiga/atari >users have enough to talk about without Intel/Motorola flames. I'm just glad >nobody has tried including net.micro.pc, then the flames would really start!] I don't have time to read another newsgroup, so I'm keeping this thread here. Sorry! >My impression is that the relative power of each chip relates strongly >to its design date. Thus, Motorola chips have tended to be more >powerful than the corresponding Intel chips, because they were designed >slightly later. The 80386 should be faster than the 68020, since it is >coming out about a year later. Agreed. >I'm not going to argue the merits of Intel versus Motorola. It is clear >to me that both have learned from each other. Now it is up to us, the >users, to learn to write our stuff portably so we can move it to the machine >of whoever is learning the fastest at each point in time. It is really not very easy to write portably for the two micro- processors, since they are such massively different architectures. If you're writing in a high-level language, then things are difficult because of the massive differences between MS-DOS and 68000-based operating systems. If you're willing to restrict yourself to character- screen oriented applications, then you might have a slim chance working completely in a high level environment (read: SLOW on an 8086.) I think the basic issue comes down to elegance. The Motorola designers decided to throw away the past when they came up with the 68000, and I think they made the right decision. Intel instead decided to sidle themselves with an endless nightmare of backward-compatibility. The 80386 operates, for example, on 32-bit objects when in native mode, and to get it to address words you have to switch to '286 emulation mode. (Although both address bytes with no problems.) The '386 has to have two emulation modes ('286 and 8086) in order for software to run properly; this makes the whole software engineering project all the more complicated. I feel sorry for the Intel designers in particular who have to put in all this extraneous crap to maintain compatibility. The 68000 series, on the other hand, is upward compatible with no need for emulation modes. (With the exception of MOVE SR,(ea)) 68000 code will run unmodified on a 68010 and a 68020. The only differences are in the interrupt exception handlers, and even this is not a problem because the 68020 will still be able to handle 68000-style interrupts without having to change the program code (only the exception handlers.) AmigaDOS/Intuition does this with no problems. The issue here is elegance as well as performance. It is much easier to take a large base of 68000 software and run it to full advantage on a 68020 because 68000 code is essentially written as though it were running on a true 32-bit machine. When it is run on a true 32-bit machine, it *really* blazes. The fact that the '020 and '030 need no "emulation modes" makes the upgrade path infinitely smoother; an MS-DOS task running under emulation mode on a 386 still has to deal with a 640K RAM limit, even if you're using virtual memory. This isn't really a serious limitation in most cases, but it is certainly ugly and to my mind unnecessarily complex and non-uniform. -Mitsu
merchant@dartvax.UUCP (Peter Merchant) (10/04/86)
> ...the advantages of a protected OS. A non-protected OS can suffer from > misbehaved programs and slow context swaps (if its multitasking, which the > MS-DOS operating system emphatically isn't. Oranges vs. oranges, remember). > A Turbo upgrade for an Amiga can deliver much the same level of performance > as a Sun and still use standard off-the-shelf Amiga software and operating > system, and it can take advantage of the extra capabilities of the 68020. > MS-DOS runs fine on a '286 PC, but it can't use the extended memory > addressing capabilities -- unless you kludge it, MS-DOS has a ceiling of > 640K. AmigaDOS has a ceiling of 4 Gigabytes. > -- > ============================================================================ > Dave Haynie {caip,ihnp4,allegra,seismo}!cbmvax!daveh Quick trivia question: How about MS-Windows? That does multitasking...how well does it work an 80386? I've almost been tempted to go to Computer- land and give it a try... -- "Oh, I can't wait all my life Peter Merchant On a street of broken dreams."
mash@mips.UUCP (10/05/86)
In article <205@mipos3.UUCP> kds@mipos3.UUCP (Ken Shoemaker ~) writes: >In article <280@husc6.HARVARD.EDU> hadeishi@husc4.UUCP (mitsuharu hadeishi) writes: >>P.S. BTW, have you guys heard about the 78000 (!?) See Nanobytes in >>the latest BYTE (the //GS issue). This is a RISC uprocessor with a >>rated speed of 20 (VAX-equivalent?) MIPS . . . this kind of speed >>totally blows away the RT PC (note: the RT PC performed just a hair >>better than an PC AT in benchmarks . . . see PC World or PC magazine >>of a month ago. . . not very impressive . . .) > >Oh, this is great. Yet another paper product. This one, they haven't >even had a press release on! I guess it must be more than a year >away (:-)). Ken has a right to be skeptical on this one! It also illustrates the silly ways in which 3rd and 4th-hand information propagate around. I quote what the Byte issue says: "Sources with Motorola's microprocessor group (Austin, TX) confirm that two new central processing units will be released by that company in the near future. (Stuff on 68030)...Due in the first quarter of next year is the 20-MIPS 78000 CPU. Sources said the 78000 is a RISC processor that represents an evolutionary progression of the 68020." What we have is: a) Somebody from Moto talking off-the-record to somebody on the Byte staff. It is possible that not all Moto people would agree. b) Byte prints this, so now it looks official-looking. c) Somebody else mentions it in the net, and now, a chip that (I'd guess) hasn't yet been taped out "totally blows away the RT PC." Now it sounds like it exists right now... No wonder people get confused between claims and reality! One has to wonder what even the 2 sentences in the article actually MEAN? "Due in the first quarter.." Does that mean: a) Appears in a machine you can login on? b) Appears in a board? c) Production quantities? d) Sample quantities? e) First silicon? f) Announcement of when any of a-e will happen? "20-MIPS 78000 CPU" Does that mean? a) Runs at 20Mhz, and can do peak 1 instruction per cycle? b) Is 20X a VAX 11/780 doing real work? Note that the conversion from peak MIPS to VAX-equivalent MIPS (or any other kind of real-live MIPS) can take on radically different factors: for example, on 68020s, the division factor seems like 4 (i.e., 16.67Mhz designs gives 8 peak Mips (2 cycles/instr) ==> 2 VAX mips. Clippers have a factor of about 6.6 (33 peak Mips ==> 5 Mips (according to published claims)). Depending on what the 78000 really will be, it might be as low as a 5Mips part (one would expect it to be higher). "RISC processor that represents an evolutionary progression of the 68020." What does this mean? a) Does it run 68020 object code? b) Does it run a subset of the 68020 instructions? c) Is it different object code, but "philosophically like the 68020"? Sigh. As one can see, if one reads the literature uncritically, things that sound like information can be found to be rather content-free, i.e., "real" information is found to be "virtual." Note: none of this is meant as an attack on the 78000, but on the weird process by which information leaks around and is jumbled up. As far as I know, the 78000 hasn't been announced by anyone for the record, and is mostly discussed with Moto customers (existing or potential) who want to do high-end systems and show signs of picking other vendors' micro-processors. Perhaps someone from Motorola might care to comment on the original article: one can choose to believe a vendor or not, but nobody should ascribe high credibility to 3rd-hand off-the-record comments, or knock the vendor for such "information". -- -john mashey DISCLAIMER: <generic disclaimer, I speak for me only, etc> UUCP: {decvax,ucbvax,ihnp4}!decwrl!mips!mash, DDD: 408-720-1700, x253 USPS: MIPS Computer Systems, 930 E. Arques, Sunnyvale, CA 94086
hsu@eneevax.UUCP (Dave Hsu) (10/05/86)
In article <205@mipos3.UUCP> kds@mipos3.UUCP (Ken Shoemaker ~) writes: >In article <280@husc6.HARVARD.EDU> hadeishi@husc4.UUCP (mitsuharu hadeishi) writes: >> >> In re: Motorola's continual announcements . . . is it fair >>to compare the 68030 to the 80386, etc.? >> >> Yes. > >Of course, there is no comparison between the two classes of machines, because >you in here introduce more than two classes of machines. I would not >put Sun workstations and Turbo Amigas in the same class of machine simply >because the Amiga does not support any kind of memory management. If you >want to have a grown-up machine, one that supports multi-user/multi-tasking >you really need to have this. I have used 68k boxes without that have >tried to do it, and it works, but you are just asking for trouble. That >the same 286-based box can support an unprotected os like ms-dos and >a protected os like Unix means that it can function in either of these >two classes. Not until you have the 68030, or you make an entire subsystem >plug in with a 68020 and an mmu, can you claim to have upgraded your Amiga >the way plugging a 286 upgrades a pc. Your assertion about classifying 68k based machines is more or less correct; I wouldn't want to lump the Amiga and the Sun together. Yet. But sooner or later Commodore will get their act together, and you can bet on the existence of an MMU in Atari's TT box next year, and that whatever design one uses, it will be infinitely easier to live with than the Intel segmentation design. A simple matter of extracting limited function from a limited design. >> Another important point to make is that operating systems implemented on >>the 68000 series chips will run fairly well on the higher-grade chips; >>UNIX, AmigaDos/Intuition, Mac Operating System, ST GEM/TOS, et cetera are >>all capable of taking advantage of the higher-level chips' capabilities > >In the first place, msdos, while not taking advantage of the functionality >of the 286 or the 386, is certainly not a kludge when run on them. In >addition, when you go from one type of 68* based box to another, you also >need to rewrite the operating system, because the memory management systems >are incompatible. You just walked down the wrong alley for an Intel person. I dare say that there are far more diversified and completely alien implementations of the MS-DOS BIOS and kernal than there ever have been 68k implementations of Unix. In fact, the amount of dirty work needed to port Unix memory management from one 68k system to another pretty much is based on only two variables: which version of Unix, and whether or not you're demand paged. The minimum functionality required to run Unix well is easily accomodated by just about every 68k MMU design I've ever seen. What's a '386 user to do, use the warped `MMU' inherent in its predecessors? No...you need a rewrite. Motorola passed that trial years ago when the '010 was introduced. > People have also generated 286-based cards that plug >into plain old IBM pcs and have gained a significant performance upgrade, while >running their old software. I'm sure the same thing can and will happen >with the 386. What a coincidence! Just about everybody else in the world has too! And if you're not convinced, I have some Apple-II+ accelerator boards I'd like you to meet. > And Unix implementations for 8086-based chips have been >floating around for at least 5 years. Amazing. Commercial Unix implementations for 68000-based machines have been floating around for well over 5 years; check the ads in your old Byte issues. >MS-DOS already is a 16-bit os, and always has been. What are you trying to say? "And it always will be?" We're not discussing 6502's here. The 68k family is already a 32-bit processor, and always has been. Bus sizing is just another one of its strengths, if you will. > When going from a 68020 to a 68030, you will have to >rewrite the os to take advantage of the memory management that the 68030 >provides...otherwise you just have a little faster 68020, much like you >can use a 286 as a faster 8086. How about if we compare apples to apples? >Can you say "incoherency?" Can you say "Yes, you are becoming incoherent"? When going from a 68020 to a 68030, you will have to throw away that PMMU chip you had sitting next to it. Code changes? Motorola claims you won't need any. I'll be more than happy to throw away another large power-sucking chip; I need the real-estate anyhow. When going from a '286 to a '386, you will have to rewrite the OS to take advantage of a real MMU. You guys DO have a real MMU, don't you? Otherwise, you have a little faster '286. As for improvements in Motorola's performance, a large number of programs benefitted wonderfully from the DBxx change in the '010, and that's the name of the game. Otherwise, why would NEC try to market the V20/V30 family? >-- >The above views are personal. > >I've seen the future, I can't afford it... > >Ken Shoemaker, Microprocessor Design, Intel Corp., Santa Clara, California Yes, the good people of Oakhill are jumping the gun a bit, but it isn't going to bother me a bit. As far a Joe Blow is concerned, his computer suddenly became faster and less power grubbing with a minimum of hardware redesign on my behalf. As far as I'm concerned, I just won an MMU and a data cache and saved myself a lot of money. Keep your iAPX toys, guys; I might look you over if you start running circles around Vax 8800's. If. The opinions expressed within are entirely mine, and may not reflect the views of the University of Maryland. But a hell of a lot of my friends share them. -dave -- David Hsu (301) 454-1433 || -8798 || -8715 "I know no-thing!" -eneevax Communications & Signal Processing Laboratory / EE Systems Staff Systems Research Center, Bldg 093 / Engineering Computer Facility The University of Maryland -~- College Park, MD 20742 ARPA: hsu@eneevax.umd.edu UUCP: [seismo,allegra,rlgvax]!umcp-cs!eneevax!hsu "Hoy?"
magore@watdcsu.UUCP (M.A.Gore - ICR) (10/05/86)
In article <7183@utzoo.UUCP> henry@utzoo.UUCP (Henry Spencer) writes: >> And what about the new family jewels? Should I feel vindicated that Mot >> has finally decided that on-chip memory management is a good idea? ... > >Only when Intel decides that on-chip caches are a good idea. The 68030 >has *both*. So does the Clipper, by the way. The MIPS chip set likewise >does the MMU and half (the hard half) of the cache. Where's Intel? > > >> ...what about performance? Why would a new, ultra-high end user choose >> the 68030 over, say, the Clipper, or the MIPS chips, each of which >> (if you believe the numbers) should exceed the 68030's performance (with >> the promise of delivering more sooner) and are available, in some semblance, >> today? > >Probably no particularly good reason except compatibility. I notice, >however, that the line of argument has shifted slightly: we don't hear >the word "Intel" any more. >-- > Henry Spencer @ U of Toronto Zoology > {allegra,ihnp4,decvax,pyramid}!utzoo!henry Hi, I just would like to bring up a new topic to think about. I find that working at the chip level can be a great joy but *most* consumers do not. They mostly run programs and only care if they run -fast enough-. As far as writing programs go the compiler is where the real sweet occurs yet compiler writers are *few* compared to end users. Then we have to weigh just how many *good* applications can be addressed with only a compiler rather then an assembler. This problem is more complex -to be sure- but I would say that the compiler that is most mature is the winner (at the time). That is because a buggy compiler can cause much pain to developers... Market competition has to occur to 'debug' the compilers etc.... What I am getting at is by no means the fault of hardware design -just the first in high volume to market wins-... I have a 68K and an 8086! but I can use both... Ok everybody - lets here it for 'to each his own' (any cheers?). I think most people don't get down to the hardware level like the rest of us. Well for now I'll think about getting so sleep and dream of some *real* cpu's not like the toys we have today but say in 10 years. (You know 10 years from now when we have the computing power of a CRAY in a kid's toy.....:-) # Mike Gore # Institute for Computer Research. # These ideas/concepts do not imply views held by the University of Waterloo.
ewhac@well.UUCP (Leo 'Bols Ewhac' Schwab) (10/05/86)
[ The following is a uuencoded program that will kill the line eater: ]
> > > O F F I C I A L A N N O U N C E M E N T < < <
At the risk of contributing further to the problem, I hereby make
the following announcement:
This line of discussion has ceased to be interesting. As a result,
the signal-to-noise ratio has gone way down.
Talking about vaoprware, after I think about it a while, only
depresses me. I'm interested in what I can do now, not next year. Waiting
is not my thing, doing neat stuff is.
We already went through The Boing Wars once already. Face it:
designers will create wonderful ideas and stupid mistakes. Some have more
of one than the other (I will privately mail my opinions to anyone who's
interested; I don't think there will be a big response, do you?). This is
the point: If machine 'x' really is sh*t, then why did person 'y' buy it?
Furthermore, why did person 'y' keep machine 'x' after "seeing for himself"
that it was sh*t? One machine will not please everyone, and that's why
there are all different kinds. This is one of the advantages of living in
America.
Please redirect followups to these threads of thought to the
appropriate news group; I'm getting tired of sorting through 50 articles a
day, only half of which are remotely interesting. To get an idea of what
it's like, visit net.jokes sometime ("There are 320 new messages in
net.jokes, only 7 of which are funny. Read now [ynq]?").
Don't followup to this article, as talk.bizarre might get annoyed
(snicker, snicker...).
_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_
________ ___
\ /___--__ Leo L. Schwab
___ ___ /\ ---##\ ihnp4!ptsfa!well!ewhac
/ X \_____ | __ _---)) ..or..
/ /_\-- -----+==____\ // \ _ well ---\
___ ( o---+------------------O/ \/ \ dual ----> !unicom!ewhac
\ / ___ \_ (`o ) hplabs -/ ("AE-wack")
____ \___/ \_/
Recumbent Bikes: "Work FOR? I don't work FOR
The _O_n_l_y Way To Fly! anybody! I'm just having fun."ingoldsby@calgary.UUCP (10/07/86)
The continuing debate on the relative merits of the MC68030 / 80386 seems to be a bit pointless. The computer industry is famous for the speed at which equipment is designed, introduced, used and falls obsolete. Of course Motorola's product is superior to the 80386 because there is an almost 1.5 year difference in technology. Doubtless the 80486 (which is probably under development right now) will be superior to the MC68030. What most people seem to forget is that processors are chosen for a particular product for a variety of reasons, only one of which is their programming characteristics. Interfacing requirements, # of support chips, price, etc. are all important characteristics for a manufacturer. I'm sure that for given applications either Motorola or Intel win. Instead of these pointless `A is > B because B is built by 3 toed sloths' arguments, why not try to restrict postings to actual specific comparisons between the two or information which may not yet be generally known. Terry Ingoldsby ...ihnp4!alberta!calgary!ingoldsby
baron@transys.UUCP (10/07/86)
In article <1870@well.UUCP> das@well.UUCP (David Shayer) writes: >>>In article <262@husc6.HARVARD.EDU> hadeishi@husc4.UUCP (mitsuharu hadeishi) >writes: >>>Now that Motorola >>>has released this chip, they have in *every category of chip* a far >>>superior chip than does Intel. > [some lines deleted] >To say that Compaq has "released" an 80386 based computer is stretching >things a bit too, don't you think. They announced it, but said it would >be available after January. This is similar to what Motorola announced. Hmmm. According to my brother ( a tech at microsoft ), Microsoft just installed over 40 of the Compaq 386 machines. His comments: 1. Makes an AT look sick. 2. Extremely fast MS-DOS. >By the way, IBM is said to be readying an announcment in January for >their 80386 based machine, which will be available in June. And don't >forget that a multitasking operating system won't be available until >around June either. Until then, these 80386 machines are just big >fast PC's. Meanwhile, Apple is readying a 68020 based machine for >release early next year. I have heard (from SCO) that SCO will be releasing Xenix 386 soon. They may already have it running internally. When last I spoke to them they mentioned that Xenix 286 research was being held back in favor of the 386 based machine. An example, they informed me they would never release a large model kernel for the 286. >Isn't vaporware fun! Yes, but in this case a multi-tasking OS will probably come available sooner than most people think. Porting anything written for the 286 is a piece of cake. >----------------------- >David Shayer @ the Well
jmvm@oakhill.UUCP (John VanMunster) (10/09/86)
In article <427@vaxb.calgary.UUCP> ingoldsby@calgary.UUCP writes: >The continuing debate on the relative merits of the MC68030 / 80386...... >....try to restrict postings to actual specific comparisons between the >two or information which may not yet be generally known. > > Terry Ingoldsby Agreed. But there is no need to cross-post to all the various pc groups!! Enough is enough. Lets have more of the useful discussions, relative to the specific groups, that had given the net its reputation for an excellent source of fast and pertinent information and news. -- John VanMunster (512) 440-2015 Motorola Inc., OakHill, Tx. 78735-8598 path:...{harvard,ihnp4,seismo,gatech,nbires}..!ut-sally!oakhill!jmvm (uucp) _____________________________________________________________________________ Disclaimer: The views contained herein are my own, not those of my employer or the administration of any computer sytem which I may be currently using. _____________________________________________________________________________ "Computers aren't inhuman - Only humans can be inhuman"..Revenge of the Nerds.