underdog@portia.Stanford.EDU (Dwight Joe) (02/06/90)
My suspicions are confirmed. The NEXT computer is in trouble. NEXT can only be saved if Steve Jobs replaces the 680X0 with RISC processor like the Sparc chip. In all compute intensive applications, the Sparcstation I beats the NEXT timewise. Worse, NEXT costs MORE than a Sparstation I. Too, the extra gadetry (like the DSP chip) on the NEXT is unlikely to be used by engineers doing compute-intensive applications. The DSP might help out in making a realistic video game; otherwise, its deadweight. What difference does it make if you can play Beethoven's fifth on the NEXT? I know. Steve's going to upgrade the NEXT to a 68040. Even then, the Sparc chip set is faster.
ccplumb@lion.waterloo.edu (Colin Plumb) (02/06/90)
In article <8859@portia.Stanford.EDU> underdog@portia.Stanford.EDU (Dwight Joe) writes: >NEXT can only be saved if Steve Jobs replaces the 680X0 >with RISC processor like the Sparc chip. In all compute >intensive applications, the Sparcstation I beats the NEXT >timewise. Worse, NEXT costs MORE than a Sparstation I. You can do a lot with architectures that are lacking in price/performance, but it's possible Steve Jobs has the same idea... have you noticed how many processor architects he's hired in the last year? -- -Colin
chari@nueces.cactus.org (Chris Whatley) (02/06/90)
Followups to comp.sys.next underdog@portia.Stanford.EDU (Dwight Joe) writes: >My suspicions are confirmed. The NEXT computer is in trouble. >NEXT can only be saved if Steve Jobs replaces the 680X0 >with RISC processor like the Sparc chip. In all compute >intensive applications, the Sparcstation I beats the NEXT >timewise. Worse, NEXT costs MORE than a Sparstation I. What store are you buying your computers from. You are incorrect. A NeXT with a 40MB winchester, a 256MB optical, 8mb of ram a 17-inch monitor with a 25Mhz '030 and a DSP is ~$7000.00 The last time I checked, (maybe it was about 2 mos. ago when we bought some sparcstations) the sparcs with a 19-inch mono display, two 80MB SCSI drives were over 10k. Sure you can crank through data at twice the speed at that price but where are you going to put it all? >Too, the extra gadetry (like the DSP chip) on the NEXT is >unlikely to be used by engineers doing compute-intensive >applications. The DSP might help out in making >a realistic video game; otherwise, its deadweight. >What difference does it make if you can play Beethoven's >fifth on the NEXT? Tell that to people who need cheap and fast data acqusition or the musicians who use it for sampling and sound processing. The NeXT comes with standard music and array processing libraries. The DSP is not just a fancy sound chip. >I know. Steve's going to upgrade the NEXT to a 68040. >Even then, the Sparc chip set is faster. What sparc machine do you think is going to beat an '040 at anything if Motorola's specs are correct? We have one of the fastest sparcstations that Sun makes in the department and it is only a 16 VAX mips machine. One of the tests a professor here ran was an application for his akcl LISP which takes 1800 seconds on his 6 MIPS '030 HP machine and takes 600 seconds on the Sun4. If the '040 @ 25Mhz will do 20 MIPS then we can expect it to run in a little over 400 seconds. Anyway, we can all do math. -- Christopher M. Whatley Research Systems Administrator - University of Texas Mathematics Work: chari@math.utexas.edu (preferably not NeXT Mail) (512/471-7711) Home: chari@nueces.cactus.org (NeXT Mail) (512/499-0475)
bs@linus.UUCP (Robert D. Silverman) (02/06/90)
In article <8859@portia.Stanford.EDU> underdog@portia.Stanford.EDU (Dwight Joe) writes:
:My suspicions are confirmed. The NEXT computer is in trouble.
:
:NEXT can only be saved if Steve Jobs replaces the 680X0
:with RISC processor like the Sparc chip. In all compute
^^^
:intensive applications, the Sparcstation I beats the NEXT
:timewise. Worse, NEXT costs MORE than a Sparstation I.
Try to be a little less dogmatic please. Your use of the word
'all' leaves a little to be desired. I have applications that
involve a lot of multi-precise integer arithmetic and rely
heavily on integer multiply and divide with remainder. These
applications run faster on the NEXT (and on the SUN-3/80 on my desk)
than they do on the Sparcstation I.
--
Bob Silverman
#include <std.disclaimer>
Internet: bs@linus.mitre.org; UUCP: {decvax,philabs}!linus!bs
Mitre Corporation, Bedford, MA 01730alan@oz.nm.paradyne.com (Alan Lovejoy) (02/06/90)
In article <8859@portia.Stanford.EDU< underdog@portia.Stanford.EDU (Dwight Joe) writes:
<My suspicions are confirmed. The NEXT computer is in trouble.
Agreed.
<NEXT can only be saved if Steve Jobs replaces the 680X0
<with RISC processor like the Sparc chip. In all compute
<intensive applications, the Sparcstation I beats the NEXT
<timewise. Worse, NEXT costs MORE than a Sparstation I.
Well, he has lots of choices here. I don't see why SPARC is to be
preferred, since it's just about the slowest of the RISC architectures.
Perhaps his cozy relationship with IBM will give him access to the
America (ROMPII) processor? Of course, the completely unsubstantiated
rumors I have heard is that he's intending to go with the 88k.
<Too, the extra gadetry (like the DSP chip) on the NEXT is
<unlikely to be used by engineers doing compute-intensive
<applications. The DSP might help out in making
<a realistic video game; otherwise, its deadweight.
<What difference does it make if you can play Beethoven's
<fifth on the NEXT?
Perhaps his target market is not engineers? Whatever. It's clear
that his prices, machine capabilities and marketing strategy are not
in harmony.
<I know. Steve's going to upgrade the NEXT to a 68040.
<Even then, the Sparc chip set is faster.
How do you know that? Have you benchmarked a 68040? Motorola claims that
the 68040 is faster, especially for floating point. While Motorola is
certainly biased, they do have one advantage: they can run benchmarks
on real 68040's. Since independent benchmarks are not yet available, perhaps
it would be best to desist from making unsupported claims?
____"Congress shall have the power to prohibit speech offensive to Congress"____
Alan Lovejoy; alan@pdn; 813-530-2211; AT&T Paradyne: 8550 Ulmerton, Largo, FL.
Disclaimer: I do not speak for AT&T Paradyne. They do not speak for me.
Mottos: << Many are cold, but few are frozen. >> << Frigido, ergo sum. >>swarren@convex.com (Steve Warren) (02/07/90)
In article <8859@portia.Stanford.EDU> underdog@portia.Stanford.EDU (Dwight Joe) writes: [...] >I know. Steve's going to upgrade the NEXT to a 68040. >Even then, the Sparc chip set is faster. I have seen assertions in several places that the '040 is faster than SPARC at equal clock rates. Does anyone have 'real' knowledge about this? -- --Steve ------------------------------------------------------------------------- {uunet,sun}!convex!swarren; swarren@convex.COM
ken@mm.uucp (Ken Seefried iii) (02/07/90)
In article <8859@portia.Stanford.EDU> underdog@portia.Stanford.EDU (Dwight Joe) writes: >My suspicions are confirmed. The NEXT computer is in trouble. > Yawn...here we go again... > >NEXT can only be saved if Steve Jobs replaces the 680X0 >with RISC processor like the Sparc chip. > Now, I'm no rabbid NeXT fan, but this is, of course, complete rubbish... While NeXT will certainly do a RISC box (an 88000, more likely than a SPARC), there is no indication that they are going to roll over and die if it doesn't happen (that is, there are lots of reasons why NeXT might fail...RISC isn't nessesarily one of them). Sure...the SPARC is faster. And to someone who doesn't look to hard, it's cheaper. But then you go and price adding Sybase, Allegro CL, Mathematica, Objective-C (tho soon free from GNU), and all the other software, and that wizzy SPARCstation 1 is pretty damn expensive... Then do this...add a 256MB Optical drive to your SS1. And a DSP SBus board (oops...there are none...sorry). And a 400-dpi laser printer. Suddenly, your talking about a $25k++ machine (mono). > >Too, the extra gadetry (like the DSP chip) on the NEXT is >unlikely to be used by engineers doing compute-intensive >applications. > And maybe I'm crazy, but I bet I can compute a 100x100 FFT faster on the 56001 than you can on your SAPRC FPU. That sounds like something "used by engineers doing compute-intensive applications". > >The DSP might help out in making >a realistic video game; otherwise, its deadweight. > Tremendously shortsighted. This is the kind of logic that says "why build personal computers...if you want to do *real* computer work, you have your *wonderful* CDC 6600! After all, number crunching is all thats important." Engineering applications are not the only application for workstations. Multi-media applications for one, derive great benifit from the DSP. Anything more complicated than simple voice synthesis really needs DSP-like hardware. Desktop publishing applications for the NeXT are left as an excercise to the reader... > >What difference does it make if you can play Beethoven's >fifth on the NEXT? > Well, none, I suppose, if thats all you can think of to do with that hardware... > >I know. Steve's going to upgrade the NEXT to a 68040. > And the '040 is no slouch...someone else can quote the numbers. > >Even then, the Sparc chip set is faster. > And the MIPS chip is going to be faster than that (I think)...and the Cray 3 faster still...big deal. Since when does being the fastest instantly make you the best computer (the Mac being case-in-point...vastly superior to other, faster machines in certain areas). Sure, I'd like the NeXT to be faster...a lot faster. But you know what I'd like more? I'd like the SS1 to come with a good database, Mathematica and 256MB of storage at no additional cost. The NeXT will get faster...the other is a pipe dream. -- Ken Seefried iii ...!<anywhere>!uunet!gatech!mm!ken MetaMedia, Inc. ken%mm.uucp@gatech.edu Atlanta, Georgia, USA obquote: "I feel...like a god..."
hascall@cs.iastate.edu (John Hascall) (02/07/90)
In article <8859@portia.Stanford.EDU> underdog@portia.Stanford.EDU (Dwight Joe) writes: }NEXT can only be saved if Steve Jobs replaces the 680X0 }with RISC processor like the Sparc chip. In all compute }intensive applications, the Sparcstation I beats the NEXT }timewise. Worse, NEXT costs MORE than a Sparstation I. Actually, there are probably already more RISC/UNIX/whatever workstation vendors than the market can support, so I don't see the profit in leaving their niche to join the crowd. }Too, the extra gadetry (like the DSP chip) on the NEXT is }unlikely to be used by engineers doing compute-intensive }applications... Time for a reality reminder. Not everyone in the world is an engineer, not everyone needs or wants an engineering workstation. In fact, while workstations may be the glamour boys, there really aren't that many of them compared to the many "less exciting" uses of [micro]processors. John Hascall
underdog@portia.Stanford.EDU (Dwight Joe) (02/07/90)
In article <7341@pdn.paradyne.com> alan@oz.paradyne.com (Alan Lovejoy) writes: | |<NEXT can only be saved if Steve Jobs replaces the 680X0 |<with RISC processor like the Sparc chip. In all compute |<intensive applications, the Sparcstation I beats the NEXT |<timewise. Worse, NEXT costs MORE than a Sparstation I. | |Well, he has lots of choices here. Agreed. But Sparc now has a huge software base that could easily be modified to fit the I/O of the NEXT, which BADLY needs software, if NEXT were to incorporate the SPARC chipset. |<Too, the extra gadetry (like the DSP chip) on the NEXT is |<unlikely to be used by engineers doing compute-intensive |<applications. The DSP might help out in making |<a realistic video game; otherwise, its deadweight. |<What difference does it make if you can play Beethoven's |<fifth on the NEXT? | |Perhaps his target market is not engineers? Whatever. It's clear |that his prices, machine capabilities and marketing strategy are not |in harmony. | Ain't that the truth. What I don't understand about NEXT is what kind of major market is it supposed to be aimed at. Some dude mentioned that the world has musicians as well as engineers. Fine. But musicians seem like too small of a market for the kind of sales projection that Jobs made. Then, Jobs points out that the NEXT is great for word-processing. So what? IBM/MAC can do a more than adequate job. Why spend close to 10K for a NEXT? Jobs initially aimed his machine at higher education. Now, he's shooting for business. I'll tell you why. In the higher education market, his machine loses twice: high price and low performance. People who do reports, write short stories, etc. aren't going to shell out that kind of money for just word processing. That covers liberal arts. Look at the other principal field: the technoids. Why would an engineering dept. pay close to 10K for a NEXT when the dept. can get a Sparcstation I for close to that price and for an order of magnitude performance in number-crunching. So, Jobs was squeezed out of the higher education market. The one thing that Jobs didn't count on back in mid 1980 was the rise of the RISC machines. What hurt him was the very long product development time of NEXT. He expected that back in mid-1980 (when he conceived of NEXT) there would be nothing like RISC. The graphics (assuming no RISC machines) and the powerful 680x0 (assuming no RISC machines) would have been great for running some snazzy graphics-filled simulations of say, molecular reactions. The DSP chip (assuming no RISC machines) would have been great for signal processing experiments in EE. Now, against RISC, NEXT with its 680x0 looks flimsy. Most of the RISC computers have great graphics and better number crunching. As for DSP, I see no reason why a DSP peripheral can't be attached to one of these RISC computers. The conclusion is that RISC has effectively locked NEXT out of a major market. The other major market is controlled by IBM/MAC. So, back to my original question, what kind of major market is NEXT supposed to be aimed at? What makes NEXT better than the dominant machines in those markets? Note that when MAC appeared, it had significant software features that were not present in other machines. (features = easy-to-use user interface, incredibly easy to use word processing, icon-driven menus etc) But NEXT doesn't have any significant software feature that are not present in other machines. |<I know. Steve's going to upgrade the NEXT to a 68040. |<Even then, the Sparc chip set is faster. | |How do you know that? Have you benchmarked a 68040? source: Businessweek
alderson@Belarius (Rich Alderson) (02/07/90)
In article <8905@portia.Stanford.EDU>, underdog@portia (Dwight Joe) writes: >So, Jobs was squeezed out of the higher education market. The one thing that >Jobs didn't count on back in mid 1980 was the rise of the RISC machines. What >hurt him was the very long product development time of NEXT. He expected that >back in mid-1980 (when he conceived of NEXT) there would be nothing like RISC. Can you support the claim that Jobs was thinking about the NeXT box in 1980? He didn't even take a look at the Macintosh until 1982 or thereabouts--when they wouldn't let him have the Lisa to screw with. As far as I know, the NeXT box was designed AFTER Jobs left Apple, in response to his published statement of intent for what his new company would work on. Certainly, being a marketing weenie and not a techie, he didn't have any hand in designing the hardware itself. Rich Alderson alderson@jessica.stanford.edu
underdog@portia.Stanford.EDU (Dwight Joe) (02/07/90)
In article <8913@portia.Stanford.EDU> alderson@jessica.stanford.edu (Rich Alderson) writes: |In article <8905@portia.Stanford.EDU>, underdog@portia (Dwight Joe) writes: ||So, Jobs was squeezed out of the higher education market. The one thing that ||Jobs didn't count on back in mid 1980 was the rise of the RISC machines. ||What hurt him was the very long product development time of NEXT. He ||expected that back in mid-1980 (when he conceived of NEXT) there would ||be nothing like RISC. |Can you support the claim that Jobs was thinking about the NeXT box in 1980? |He didn't even take a look at the Macintosh until 1982 or thereabouts--when |they wouldn't let him have the Lisa to screw with. | |As far as I know, the NeXT box was designed AFTER Jobs left Apple, in response |to his published statement of intent for what his new company would work on. |Certainly, being a marketing weenie and not a techie, he didn't have any hand |in designing the hardware itself. | You might be right on the time period. I thought that he had left in the mid 80s. My knowledge of the "long product development time" of the NEXT comes from sundry mag. articles and, in particular, a PBS show. I am sure of the "long product development time" that plagued the NEXT. At the time of conception, which must have been before the PBS show, Job could not possibly have thought that RISC would be a threat to the NEXT because RISC just wasn't commercially popular at the time of the PBS broadcast.
frk@mtxinu.COM (Frank Korzeniewski) (02/07/90)
In article <8905@portia.Stanford.EDU> underdog@portia.Stanford.EDU (Dwight Joe) writes: > >|<I know. Steve's going to upgrade the NEXT to a 68040. >|<Even then, the Sparc chip set is faster. >| >|How do you know that? Have you benchmarked a 68040? > >source: Businessweek WOW! The definitive source for technical information! I read in EE Times that the 68040 is faster than the Sparc. But after all, what do they know. They are only some technoid rag. Frank Korzeniewski (frk@mtxinu.com)
iyengar@grad2.cis.upenn.edu (Anand Iyengar) (02/07/90)
In article <8859@portia.Stanford.EDU> underdog@portia.Stanford.EDU (Dwight Joe) writes: >NEXT can only be saved if Steve Jobs replaces the 680X0 >with RISC processor like the Sparc chip. In all compute >intensive applications, the Sparcstation I beats the NEXT >timewise. Worse, NEXT costs MORE than a Sparstation I. Hardware alone doesn't make the system, nor is it the sole basis for people's buying decisions. For most of the computer intensive code that I run, a DEC 3100 (MIPS) is ~2 times as fast as a sparcstation (and I think that they're fairly close in price). That doesn't change the fact that the school will probably buy sparcstations before 3100's. Little things like the OS seem to matter (:-). >Too, the extra gadetry (like the DSP chip) on the NEXT is >unlikely to be used by engineers doing compute-intensive >applications. The DSP might help out in making >a realistic video game; otherwise, its deadweight. One man's gadget is another man's workhorse. There exist Real (tm) people who feel that they need such a beast. >What difference does it make if you can play Beethoven's fifth on the NEXT? Their system is good for more than just playing music. Like it or not, the software, and its integration with the nifty hardware, does count. There are a number of things that I just can't do on a sparcstation, which are easy on the nExt. And, to be fair, one can play Beethoven on the sparcstation, too; it just doesn't sound as good. >I know. Steve's going to upgrade the NEXT to a 68040. >Even then, the Sparc chip set is faster. That depends on how much faster "faster" is. I'd argue that the nExt needed to be faster from the start, but not for just number-crunching. Even starting up an (graphical) application on the nExt takes too long (insert magic numbers for acceptable response in an interactive environment here). I'm not sure that the 68040 will (or won't -- wait until it gets here) make this better. Also, nExt is undoubtedly designing a higher performance system, so I'd expect the next-to-next nExt (after the '040 version) to be faster. Anand. -- "The closer you get to your destination, the more you're slip-sliding away..." {inter | bit}net: iyengar@eniac.seas.upenn.edu uucp: !$ | uunet --- Lbh guvax znlor vg'yy ybbx orggre ebg-guvegrrarg? --- Disclaimer: I get in enough trouble speaking for me...
daveh@xtenk.sgi.com (David A Higgen) (02/07/90)
On a totally flippant note... after the XT in the IBM_PC world came, of course, the AT. Now, if you apply the same substitution to NEXT, you get... NEAT!! Can this be an accident?? daveh
antony@lbl-csam.arpa (Antony A. Courtney) (02/07/90)
In article <49956@sgi.sgi.com> daveh@xtenk.sgi.com (David A Higgen) writes: >On a totally flippant note... after the XT in the IBM_PC world came, >of course, the AT. Now, if you apply the same substitution to NEXT, you >get... NEAT!! Can this be an accident?? > > > daveh More relevant to the discussion of the NeXT is probably the comparison of the Lisa to the Mac. The Lisa was slow, overpriced, and uncompetetive. That wasn't of much importance. The machine was important because it was a machine which people at apple could do R&D for. The Macintosh embodied the design concepts of the Lisa, but it was very clear that the fundamental mistakes the engineers made were not repeated in the Mac. If you look at the NeXT as a Lisa of sorts, then it is a very good machine. The NeXT is also good because it embodies certain ideas which are very important. Sure, the NeXT has a DSP whether you want it or not. That means EVERYONE has stereo sound. Sure, everyone may not want it. But history has shown that the best way to bring technology to the people is by REQUIRING it. May be fascist, but it works. Just look at what a mess things were in for a while when PCs didn't come standard with a mouse. Designers couldn't ASSUME the user had a mouse, and that made the overhead of application writing very extreme. And look at workstations: Every Sun has a 19" screen. Imagine if there were other screens available? Do you think everyone would have bought a 19" screen? No way. And what would that have done to the development of window systems? I suspect it would have hampered it severely... In general, the Lisa idea is a very important one. I think it is a very sound practice to design something new and exciting and then do it again from the ground up, once the engineers involved have learned a few things. I think the bext possible thing to do in the window system arena is to rm -rf /usr/local/src/X and start over. X is slow, clunky and it is a mess. The protocol is overly complex and there are several fundamental design errors. I think a lot of people recognize this. But because everyone wants to 'standardize' X, there isn't any way to get away from it. This same principle is also what made UNIX so spiffy. Researchers wrote Multics. It sucked. But people learned an awful lot about what should and shouldn't be in an OS and how to implement OSs. Then people scrapped it and wrote UNIX based on things which had been learned from previous OSs. Imagine what the world would be like if UNIX and any other technological developments in the OS arena had to conform to a SMID--'Standard Multics Interface Definition'. :-) I'll admit it isn't clear whethewr or not NeXT looks upon it's current box as a Lisa. If so, I look very forward to the next NeXT. :-) antony -- ******************************************************************************* Antony A. Courtney antony@lbl.gov Advanced Development Group ucbvax!lbl-csam.arpa!antony Lawrence Berkeley Laboratory AACourtney@lbl.gov
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (02/07/90)
In article <4791@helios.ee.lbl.gov> antony@lbl-csam.arpa (Antony A. Courtney) writes: | May be fascist, but it works. Just look at what a mess things were in for a | while when PCs didn't come standard with a mouse. Designers couldn't ASSUME | the user had a mouse, and that made the overhead of application writing very | extreme. Whatever gave you the idea that PC come standard with a mouse? We have about 1200 and I would guess <300 have a mouse. It's an extra cost option on most systems. | And look at workstations: Every Sun has a 19" screen. Imagine if | there were other screens available? Do you think everyone would have bought | a 19" screen? No way. And what would that have done to the development of | window systems? I suspect it would have hampered it severely... This will be a shock, but all Suns do not come standard with a 19 inch screen, or we have been getting optional small monitors ;-) Almost all of our Sun/4 and Sparcstations have the 15 inch color screen. | This same principle | is also what made UNIX so spiffy. Researchers wrote Multics. It sucked. But | people learned an awful lot about what should and shouldn't be in an OS and | how to implement OSs. Then people scrapped it and wrote UNIX based on things | which had been learned from previous OSs. I suspect that you have never used Multics and don't recall that UNIX was written because there was not enough access to Multics. UNIX is just beginning to implement some of the ideas which have been working in Multics for two decades, such as mapping files to memory. The only reason Multics is not where UNIX is today is that it was developed by one company which didn't know how to sell computers and then rights went to another. If Multics had been ported to minis and micros as soon as the hardware would support it, a lot of people running it on large machines would use it on everything. There was some negotiation to buy the Multics rights from Honeywell and port it to the 386 (you really need those four levels of privilege), but I was told that Honeywell was afraid that it would cut into the GCOS market. That's too bad. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "Stupidity, like virtue, is its own reward" -me
gerry@zds-ux.UUCP (Gerry Gleason) (02/07/90)
In article <8905@portia.Stanford.EDU> underdog@portia.Stanford.EDU (Dwight Joe) writes: >In article <7341@pdn.paradyne.com> alan@oz.paradyne.com (Alan Lovejoy) writes: >|<NEXT can only be saved if Steve Jobs replaces the 680X0 >|<with RISC processor like the Sparc chip. In all compute >|<intensive applications, the Sparcstation I beats the NEXT >|<timewise. Worse, NEXT costs MORE than a Sparstation I. >|Well, he has lots of choices here. >Agreed. But Sparc now has a huge software base that could >easily be modified to fit the I/O of the NEXT, which BADLY >needs software, if NEXT were to incorporate the SPARC chipset. Huh? SPARC has a larger software base than 680x0? Where have you been? If your talking about stuff for SunOS, sure, but NEXT was never intended to leverage from that software base, or they could be doing it now. IMHO, software is NEXT's main problem. Job's decided to invend yet another wizzy system interface, and at a time when the whole industry wants system software standards. When the MAC first came out there wasn't anything comparable, now there is, so it is unlikely that the NEXT environment will have sufficient software written for it. >So, Jobs was squeezed out of the higher education market. The >one thing that Jobs didn't count on back in mid 1980 was the >rise of the RISC machines. What hurt him was the very long >product development time of NEXT. He expected that back in mid-1980 >(when he conceived of NEXT) there would be nothing like RISC. And this really long product development is probably due to his choice to reinvent the wheel in terms of software, but since this is comp.arch, maybe we could discuss whether RISC was predictable in mid-1980 (do you mean mid 1980's, 1980 seems much to early for the inception of the NEXT concept). By the mid 1980's it was very clear than RISC would be an important technology. In addition to being a big selling point, a RISC processor would have lessened the impact of the machines departure from software standards. Another case in point, I had the opportunity to work on a project using AT&T's CRISP processor, and was very surprised to find out that a proposal to build this chip had been around since before they built the first of the 32100 family. By the time they put up the resources to build the CRISP, the 32100 was well established and Sun was nearly ready to market their SPARC strategy, so the project fizzled. Had the 32100 been built from the CRISP proposal, the rest would be history, it would have been the first commercial RISC based processor, and it would have become the porting base for UNIX in the early 80's. Of course who can really predict what the market would have done, but such an early RISC processor would have put the pressure on Intel and Motorola much earlier, perhaps the 80386 would not have been built (or not been all that successful since it would have been competing from day one with Intel RISC). >|<I know. Steve's going to upgrade the NEXT to a 68040. >|<Even then, the Sparc chip set is faster. >|How do you know that? Have you benchmarked a 68040? Has anyone seen a 68040? I thought not. You are comparing a chip that won't ship until this summer with one that is in a machine that has been in production for some time. This occurs over and over in the RISC/CISC debate, but that doesn't seem to keep people from making these silly comparison's. BTW, what are some current prices on RISC chips? I have read that 80486's are ~950$ in thousand quantity, and someone posted 68040's are expected to be ~750$. I suppose you should include the MMU and FPU in the RISC prices since they are on the chip for the comparable CISC's, but since a large percentage of users don't need and FPU including this unit probably distorts the comparison. From day one I expected RISC processors to get to commodity prices very quickly (i.e. prices based almost completely on the cost to make to chip). Has that happened yet? Gerry Gleason
alan@oz.nm.paradyne.com (Alan Lovejoy) (02/07/90)
In article <8913@portia.Stanford.EDU< alderson@jessica.stanford.edu (Rich Alderson) writes: <In article <8905@portia.Stanford.EDU>, underdog@portia (Dwight Joe) writes: <>So, Jobs was squeezed out of the higher education market. The one thing that <>Jobs didn't count on back in mid 1980 was the rise of the RISC machines. What <>hurt him was the very long product development time of NEXT. He expected that <>back in mid-1980 (when he conceived of NEXT) there would be nothing like RISC. <Can you support the claim that Jobs was thinking about the NeXT box in 1980? <He didn't even take a look at the Macintosh until 1982 or thereabouts--when <they wouldn't let him have the Lisa to screw with. I presume that Dwight is the victim of a typo. By "mid 1980," I think he meant "mid 1980s." Makes a big difference. ____"Congress shall have the power to prohibit speech offensive to Congress"____ Alan Lovejoy; alan@pdn; 813-530-2211; AT&T Paradyne: 8550 Ulmerton, Largo, FL. Disclaimer: I do not speak for AT&T Paradyne. They do not speak for me. Mottos: << Many are cold, but few are frozen. >> << Frigido, ergo sum. >>
paul@taniwha.UUCP (Paul Campbell) (02/08/90)
In article <20056@netnews.upenn.edu> iyengar@grad2.cis.upenn.edu.UUCP (Anand Iyengar) writes: >In article <8859@portia.Stanford.EDU> underdog@portia.Stanford.EDU (Dwight Joe) writes: >>I know. Steve's going to upgrade the NEXT to a 68040. >>Even then, the Sparc chip set is faster. > That depends on how much faster "faster" is. I'd argue that the nExt >needed to be faster from the start, but not for just number-crunching. Even >starting up an (graphical) application on the nExt takes too long (insert >magic numbers for acceptable response in an interactive environment here). >I'm not sure that the 68040 will (or won't -- wait until it gets here) make Look closely at the architecture (try the Byte article), the NeXT board multiplexes data/addresses from the '030 to make it look like the '040 (or NuBus) - this means that their silicon is probably already set up for the '040, my bet is that Steve is just waiting for the parts in quantities/prices that he can afford to ship them. BTW I think he lucked out on the '040 bus interface, my guess is that the reason that it looks so much like NuBus is because Apple buys the most 68ks. Paul -- Paul Campbell UUCP: ..!mtxinu!taniwha!paul AppleLink: CAMPBELL.P You know there's something wrong when 100,000 people marching in Moscow makes page 1 and 400,000 in Washington doesn't .....
terry@uts.amdahl.com (Lewis T. Flynn) (02/08/90)
In article <4791@helios.ee.lbl.gov> antony@lbl-csam.arpa (Antony A. Courtney) writes: [some interesting points about computer systems evolution] >is also what made UNIX so spiffy. Researchers wrote Multics. It sucked. But >people learned an awful lot about what should and shouldn't be in an OS and >how to implement OSs. Then people scrapped it and wrote UNIX based on things I couldn't let this pass. Multics may not have been a commercial success, but it hardly sucked. Its main problem may have been that Honeywell (who inherited it when they bought the GE computer division) was never more than lukewarm about it and tried to kill it umpteen times over the years. They seem to have finally succeeded. but as late as '87 or so there were still over a hundred systems running it. Only something with a great deal to offer could generate such loyalty to orphan software (and hardware, if I remember correctly). I would speculate that if matters had been only slightly different in one of several possible places, Multics would be the system we were all using and writing software for. Terry disclaimer: As far as I know, Amdahl has no opinion on Multics.
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (02/08/90)
In article <160@zds-ux.UUCP> gerry@zds-ux.UUCP (Gerry Gleason) writes: | BTW, what are some current prices on RISC chips? I have read that | 80486's are ~950$ in thousand quantity, and someone posted 68040's | are expected to be ~750$. I suppose you should include the MMU and | FPU in the RISC prices since they are on the chip for the comparable | CISC's, but since a large percentage of users don't need and FPU | including this unit probably distorts the comparison. From day one | I expected RISC processors to get to commodity prices very quickly | (i.e. prices based almost completely on the cost to make to chip). | Has that happened yet? I think you are making (or almost making) an important point here. One of the reasons for the interest in the I860 CPU is that a lot of the glue chips are on the CPU substrate. This allows a RISC UNIX box to be built for less than a system with separate cache, MMU, and FPU. It makes the board layout easier, and time to market shorter. Please don't read this as "Intel is great," just that there are going to be some new workstations out (they may be shipping already) with the 860 just for this reason. My personal feeling is that I no longer care what the CPU looks like or who makes it, I look for available software and performance, either for one task or general purpose. Having explained why vendors were interested in the 860, and what it offers relative to SPARC, would someone tell me why DG thinks the 88k is better than SPARC? Serious question, what has the 88k got that SPARC doesn't? -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "Stupidity, like virtue, is its own reward" -me
underdog@portia.Stanford.EDU (Dwight Joe) (02/08/90)
In article <160@zds-ux.UUCP> gerry@zds-ux.UUCP (Gerry Gleason) writes: |In article <8905@portia.Stanford.EDU> underdog@portia.Stanford.EDU (Dwight Joe) writes: ||In article <7341@pdn.paradyne.com> alan@oz.paradyne.com (Alan Lovejoy) writes: |||Well, he has lots of choices here. || ||Agreed. But Sparc now has a huge software base that could ||easily be modified to fit the I/O of the NEXT, which BADLY ||needs software, if NEXT were to incorporate the SPARC chipset. | |uh? SPARC has a larger software base than 680x0? Where have you |been? I meant that SPARC, compared to the other RISCs, is the best choice in terms of having the largest software base. |IMHO, software is NEXT's main problem. Agreed. That's the problem of any new machine. If JOBS does indeed upgrade the NEXT to a RISC machine, then SPARC is a good choice software-wise.
bbc@sicilia.rice.edu (Benjamin Chase) (02/08/90)
alan@oz.nm.paradyne.com (Alan Lovejoy) writes: ><NEXT can only be saved if Steve Jobs replaces the 680X0 ><with RISC processor like the Sparc chip. >Perhaps his cozy relationship with IBM will give him access to the >America (ROMPII) processor? Does his cozy relationship go any further than the deal involving the NeXT windowing system (what's it called, NeXT StEP)? To me, that just looked like IBM being conservative, betting money on all the horses, so that it couldn't lose (not that it would really "win" either, but losing all the races you've bet on is no fun). From a business standpoint, using the America processor seems to me to be a very unwise move, and I doubt it will happen. "Let's see, I'll just crawl underneath IBM's huge thumb." First thing is the issue of compilers for the chip. Who's got them? Well, IBM does, but those are probably proprietary. Maybe NeXT will retarget the GNU compilers to generate code for this processor, giving the fruits of their labors to the FSF? Or pay a software house to retarget a C++ compiler? These are all ticklish things, and a shoddy job of retargetting a compiler could make more difference in the ultimate performance of the box than the choice of processor. He would do well to get a good processor _and_ a good compiler for it. [Pushing this thread back towards comp.arch fodder] I thought the NeXT architecture, and its choice of O/S, was designed to eventually adapt to multiple processors (no, I don't mean the DSP, you boneheads, I already know about that :-). Any opinions on this? Other PC and workstation class machines are slowly appearing with multiple processors. Any totally unsubstantiated rumors about the NeXT that someone would care to spew on the net? :-) -- Ben Chase <bbc@rice.edu>, Rice University, Houston, Texas
alderson@Belarius (Rich Alderson) (02/08/90)
In article <8914@portia.Stanford.EDU>, underdog@portia (Dwight Joe) writes: >In article <8913@portia.Stanford.EDU> alderson@jessica.stanford.edu > (Rich Alderson) writes: >|In article <8905@portia.Stanford.EDU>, underdog@portia (Dwight Joe) writes: >||The one thing that Jobs didn't count on back in mid 1980 was the rise of the >||RISC machines. >|Can you support the claim that Jobs was thinking about the NeXT box in 1980? >You might be right on the time period. I thought that he had left in the mid >80s. Misunderstanding: "Mid 1980" is not equivalent to "the mid 80s." The falling out between Jobs and Sculley came about 1985 or so. The NeXT box was finally announced late in 1987, more than a year later than Jobs' (typically) overly optimistic expectations. I read your original article as claiming that Jobs was involved with the NeXT a year before the IBM PC was announced. Anyway, I like the idea of the NeXT box: All that great software bundled in with great hardware. My problems with it are due to the lossy nature of the optical disk and the hard disk problems that have cropped up. Rich Alderson alderson@jessica.stanford.edu
bbc@sicilia.rice.edu (Benjamin Chase) (02/08/90)
In article <BBC.90Feb7154213@sicilia.rice.edu> I write: > Does [Steve Jobs'] cozy relationship [with IBM] go any further than > the deal involving the NeXT windowing system (what's it called, > NeXT StEP)? To me, that [deal] just looked like IBM being > conservative, betting money on all the horses, so that it couldn't > lose (not that it would really "win" either, but losing all the races > you've bet on is no fun). Weeeeellllll, I just read on comp.sys.next that two days ago IBM announced they will support NeXTStEP on their AIX machines. Talk about timing. Apparently IBM has decided that nExTsTeP, along with OSF/Motif, will win, place, or show, and that the X Window System will not pay, at least on IBM hardware. [Followup-to: comp.sys.next] -- Ben Chase <bbc@rice.edu>, Rice University, Houston, Texas
mash@mips.COM (John Mashey) (02/08/90)
In article <160@zds-ux.UUCP> gerry@zds-ux.UUCP (Gerry Gleason) writes: >In article <8905@portia.Stanford.EDU> underdog@portia.Stanford.EDU (Dwight Joe) writes: >>In article <7341@pdn.paradyne.com> alan@oz.paradyne.com (Alan Lovejoy) writes: >>|<NEXT can only be saved if Steve Jobs replaces the 680X0 >>|<with RISC processor like the Sparc chip. In all compute >>So, Jobs was squeezed out of the higher education market. The >>one thing that Jobs didn't count on back in mid 1980 was the >>rise of the RISC machines. What hurt him was the very long >>product development time of NEXT. He expected that back in mid-1980 >>(when he conceived of NEXT) there would be nothing like RISC. >this is comp.arch, maybe we could discuss whether RISC was predictable >in mid-1980 (do you mean mid 1980's, 1980 seems much to early for >the inception of the NEXT concept). By the mid 1980's it was very >clear than RISC would be an important technology. In addition to >being a big selling point, a RISC processor would have lessened the >impact of the machines departure from software standards. Why don't we just kill off all this silly speculation: all of this is 100% wrong: NeXT has certainly been aware of RISC very early. At the time they had to make their choice of processor [this was 1986/87], a 68030 was a very reasonable choice, as there was NO RISC available with: high performance low cost large volume supply with sure supplies After all, at that point: - Clipper performance wasn't that strong, and I'm not sure when the Fairchild uncertainty was going on, but it might have been around then. - MIPS was a 120-person company relying on foundries (not semi partners), and NeXT would have been incredibly gutsy at that point to have used MIPS. - SPARC wasn't announced yet, and the low level of integration of the gate array designs surely would have exceeded NeXT cost goals. - 88K was far away - i860 was even further off. Anyway, one might criticize them for not guessing how long the software would take, and then trying to guess which RISCs would do well and picking one of them, but I think it would have taken amazing precognition in early 1987 to predict what's happened since... and amazing courage to have bet the company on things not yet proven. Had they started a year later, maybe things would be different, but don't ding them by claiming something was obvious, when it wasn't at all. > >Another case in point, I had the opportunity to work on a project >using AT&T's CRISP processor, and was very surprised to find out that >a proposal to build this chip had been around since before they Yes, it is sad that CRISP didn't get out, as it at least had some elegant and clever ideas. >BTW, what are some current prices on RISC chips? I have read that >80486's are ~950$ in thousand quantity, and someone posted 68040's >are expected to be ~750$. I suppose you should include the MMU and >FPU in the RISC prices since they are on the chip for the comparable >CISC's, but since a large percentage of users don't need and FPU >including this unit probably distorts the comparison. From day one >I expected RISC processors to get to commodity prices very quickly >(i.e. prices based almost completely on the cost to make to chip). >Has that happened yet? For sure, in 10,000s, I've heard of 12.5MHZ R2000/R2010 chipsets at $100 for the pair, which means you build a pretty reasonable cache & memory interface [i.e., the whole core] for $200-$250. This was a while ago, so the higher clock rates are probably creeping down from the $10/mip (CPU) that was being quoted a year back. (Note that such a configuration should be usually faster on integer, and even faster on FP, than a 25MHz 486 with external cache.) -- -john mashey DISCLAIMER: <generic disclaimer, I speak for me only, etc> UUCP: {ames,decwrl,prls,pyramid}!mips!mash OR mash@mips.com DDD: 408-991-0253 or 408-720-1700, x253 USPS: MIPS Computer Systems, 930 E. Arques, Sunnyvale, CA 94086
khb@chiba.kbierman@sun.com (Keith Bierman - SPD Advanced Languages) (02/08/90)
In article <BBC.90Feb7154213@sicilia.rice.edu> bbc@sicilia.rice.edu (Benjamin Chase) writes:
[Pushing this thread back towards comp.arch fodder]
Nice move :>
I thought the NeXT architecture, and its choice of O/S, was designed
to eventually adapt to multiple processors (no, I don't mean the DSP,
you boneheads, I already know about that :-). Any opinions on this?
Other PC and workstation class machines are slowly appearing with
multiple processors. Any totally unsubstantiated rumors about the
NeXT that someone would care to spew on the net? :-)
Mach was certainly designed to aide and abet distribution of work over
a network (which may very well live in one box, employ shared memory
and other nifty stuff).
Seems to me that one should probably build a bang up uniprocessor
first .... but I've been wrong before (and will be again, no doubt).
--
Keith H. Bierman |*My thoughts are my own. !! kbierman@Eng.Sun.COM
It's Not My Fault | MTS --Only my work belongs to Sun* kbierman%eng@sun.com
I Voted for Bill & | Advanced Languages/Floating Point Group
Opus | "When the going gets Weird .. the Weird turn PRO"
"There is NO defense against the attack of the KILLER MICROS!"
Eugene Brooksyam@nttmhs.ntt.jp (Toshihiko YAMAKAMI) (02/08/90)
From article <4791@helios.ee.lbl.gov>, by antony@lbl-csam.arpa (Antony A. Courtney): > More relevant to the discussion of the NeXT is probably the comparison of the > Lisa to the Mac. The Lisa was slow, overpriced, and uncompetetive. That > wasn't of much importance. The machine was important because it was a machine > which people at apple could do R&D for. The Macintosh embodied the design > concepts of the Lisa, but it was very clear that the fundamental mistakes the > engineers made were not repeated in the Mac. If you look at the NeXT as a > Lisa of sorts, then it is a very good machine. I agree. When I was a computer science student in University of Tokyo, Japan, I heard from one of assistant professors, "In 20th century, American people were surprised three times. Hiroshima, Sputonik, Lisa. Hiroshima was the first atomic bomb. Sputonik, the first sattellite into space. Lisa, the first workstation with $10,000. It can write, draw, calculate, communicate, and schedule. That's all what we need in business." It was 1983 maybe before Macintosh appeared (in Japan?). Now it is 1990. I am surprised at the power of Interface Builder. So, I hope it can be executed faster in next NeXT, as done Lisa things in Mac. -- Toshihiko YAMAKAMI Toshihiko YAMAKAMI NTT Telecommunication Networks Laboratories Telephone: +81-468-59-3781 FAX: +81-468-59-2546 junet: yam@nttmhs.ntt.jp CSNET: yam%nttmhs.ntt.jp@relay.cs.net snail-mail: Take 1-2356-523A, Yokosuka, Kanagawa 238-03 JAPAN
antony@lbl-csam.arpa (Antony A. Courtney) (02/08/90)
In article <160@zds-ux.UUCP} gerry@zds-ux.UUCP (Gerry Gleason) writes: }In article <8905@portia.Stanford.EDU} underdog@portia.Stanford.EDU (Dwight Joe) writes: }}In article <7341@pdn.paradyne.com} alan@oz.paradyne.com (Alan Lovejoy) writes: }}|<NEXT can only be saved if Steve Jobs replaces the 680X0 }}|<with RISC processor like the Sparc chip. In all compute }}|<intensive applications, the Sparcstation I beats the NEXT }}|<timewise. Worse, NEXT costs MORE than a Sparstation I. } }}|Well, he has lots of choices here. } }}Agreed. But Sparc now has a huge software base that could }}easily be modified to fit the I/O of the NEXT, which BADLY }}needs software, if NEXT were to incorporate the SPARC chipset. } }Huh? SPARC has a larger software base than 680x0? Where have you }been? If your talking about stuff for SunOS, sure, but NEXT was }never intended to leverage from that software base, or they could }be doing it now. IMHO, software is NEXT's main problem. Job's }decided to invend yet another wizzy system interface, and at a time }when the whole industry wants system software standards. When the }MAC first came out there wasn't anything comparable, now there is, }so it is unlikely that the NEXT environment will have sufficient }software written for it. } }}So, Jobs was squeezed out of the higher education market. The }}one thing that Jobs didn't count on back in mid 1980 was the }}rise of the RISC machines. What hurt him was the very long }}product development time of NEXT. He expected that back in mid-1980 }}(when he conceived of NEXT) there would be nothing like RISC. } This is a bit off from the truth. I am looking at an article about Bill Joy in the SF Chronicle from 1985 or 86. The article claims that Sun is well on the way to trying to make SPARC a standard, and then 3-4 paragraphs later, says: ...and Bechtolsheim recently began work on the design of a low-cost educational computer that Sun employees jokingly refer to as the "Last" machine in a pointed reference to its principal competitor, Steve Jobs' NeXT Inc. Joy says the two computers will have slightly different focuses. He refers to Jobs' computer, which is reported to have stunning sound and graphics features, as the "ultimate dorm room machine for yuppie puppies." } }Has anyone seen a 68040? I thought not. You are comparing }a chip that won't ship until this summer with one that is in }a machine that has been in production for some time. This }occurs over and over in the RISC/CISC debate, but that doesn't }seem to keep people from making these silly comparison's. } ... }I expected RISC processors to get to commodity prices very quickly }(i.e. prices based almost completely on the cost to make to chip). People also seem to be ignoring something here: A key word in SPARC is Scaleable. It is, according to Sun, "designed from the outset to yield succesively faster chips". I'm SURE that by the time a 68040 based machine is shipping, Sun will have a comparably priced SPARC based machine with equal or greater performance. }Gerry Gleason antony -- ******************************************************************************* Antony A. Courtney antony@lbl.gov Advanced Development Group ucbvax!lbl-csam.arpa!antony Lawrence Berkeley Laboratory AACourtney@lbl.gov
antony@lbl-csam.arpa (Antony A. Courtney) (02/08/90)
In article <8dmK02o886EM01@amdahl.uts.amdahl.com> terry@amdahl.uts.amdahl.com (Lewis T. Flynn) writes: >In article <4791@helios.ee.lbl.gov> antony@lbl-csam.arpa (Antony A. Courtney) writes: > >[some interesting points about computer systems evolution] > >>is also what made UNIX so spiffy. Researchers wrote Multics. It sucked. But >>people learned an awful lot about what should and shouldn't be in an OS and >>how to implement OSs. Then people scrapped it and wrote UNIX based on things > >I couldn't let this pass. Multics may not have been a commercial success, but >it hardly sucked. Its main problem may have been that Honeywell (who inherited >it when they bought the GE computer division) was never more than lukewarm >about it and tried to kill it umpteen times over the years. They seem to have >finally succeeded. but as late as '87 or so there were still over a hundred >systems running it. Only something with a great deal to offer could generate >such loyalty to orphan software (and hardware, if I remember correctly). > You and several other people posted messages with a similar tone. My appologies if I offended any old-time Multics users out there. When posting the message, I was a bit hesitant in saying flatly that "Multics sucks". I had hoped people would see my point and just ignore that little out of context statement. I have, I will admit, never used Multics. Whether or not it sucked TECHNICALLY SPEAKING is not really relevant to the point I was trying to make. The point I was ateempting to make was that Multics, for WHATEVER REASONS(and several of you argue marketing), was not a succesful operating system. The point of my posting was that people learned from Multics and then started over. On a similar token, I argued in the same posting that the Apple Lisa "sucked". In actuality, it had some very good design principles. And it was the first commercial machine to use windows and a mouse. But for whatever reasons, it was not succesful. So engineers took what they learned and started over. My point is just that it is important and valuable to learn something from one venture into a new research field, abandon those efforts, and redesign using what has been learned. Whether or not Multics was technically sound is neither here nor there in this discussion. antony -- ******************************************************************************* Antony A. Courtney antony@lbl.gov Advanced Development Group ucbvax!lbl-csam.arpa!antony Lawrence Berkeley Laboratory AACourtney@lbl.gov
alan@oz.nm.paradyne.com (Alan Lovejoy) (02/08/90)
In article <2100@crdos1.crd.ge.COM> davidsen@crdos1.crd.ge.com (bill davidsen) writes: > Having explained why vendors were interested in the 860, and what it >offers relative to SPARC, would someone tell me why DG thinks the 88k is >better than SPARC? Serious question, what has the 88k got that SPARC >doesn't? Significantly better performance. A better supplier. The 88k Open Consortium with proven BCS. Cache coherency and FPU built-in. Hardware multiply with a latency of only four cycles, and the ability to start a new multiply thru the pipe each clock cycle. Super fast task swithching. Indexed addressing mode. Highly useful "mak" (shift bitfield left) and "ext" (shift bitfield right) instructions. But that was a strange question to ask. The natural question would seem to be "What has SPARC got that the 88k, and the Rx000, don't?" ____"Congress shall have the power to prohibit speech offensive to Congress"____ Alan Lovejoy; alan@pdn; 813-530-2211; AT&T Paradyne: 8550 Ulmerton, Largo, FL. Disclaimer: I do not speak for AT&T Paradyne. They do not speak for me. Mottos: << Many are cold, but few are frozen. >> << Frigido, ergo sum. >>
ca@cbnewsi.ATT.COM (christopher.arnone) (02/09/90)
Initial information about the 040 indicate that it will be faster than the SPARC at 25Mhz. Of course, this remains to be seen. However, given Motorola's track record, I believe they feel they can still deliver the type of performance required from a state-of-the-art microprocessor from the 680x0 CISC base. I know that HP has announced the use of the 040 in one of their workstations, and it almost a sure bet that NeXT will do the same. I would bet that 040 based workstations will be comparable to the Sun 4 machines. I've heard that Motorola is planning an 050 also, so CISC is far from dead.
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (02/09/90)
In article <5190@convex.convex.com> swarren@convex.com (Steve Warren) writes: | I have seen assertions in several places that the '040 is faster than | SPARC at equal clock rates. Does anyone have 'real' knowledge about | this? The only comparison which counts is the performance at the highest production clockrate. When I buy a workstation I don't care about RISC vs. CISC, or "clock rate efficiency" (there's a marketing term if I ever heard one), I care about how many clock seconds it takes to run my job. If the job is CPU bound that maps one to one into CPU sec. Arguing about benchmarks between a production chip and an engineering sample chip is interesting but not informative. Let's talk about the speed of SPARC in 200MHz GAs or ECL. I hear there are cooling problems and the chip can become litterally vaporware ;-) Let's not take this too seriously, I will be very interested in benchmarks of production 040 machines against production SPARC machines, when both exist at the same time. I suspect that the chips would run about the same speed based on use of SPARC and reading of the same dribble of tech info I've seen (ie. benchmarks will be less than 2:1 different, and not all will show the same chip faster). -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "Stupidity, like virtue, is its own reward" -me
rogerk@mips.COM (Roger B.A. Klorese) (02/09/90)
In article <4801@helios.ee.lbl.gov> antony@lbl-csam.arpa (Antony A. Courtney) writes: >The point of my >posting was that people learned from Multics and then started over. But you're simply wrong. People wrote UNIX to deal with the fact that they just couldn't get sufficient access to the big iron it ran on. UNIX was one approach; Prime's PRIMOS (and subsequently in many ways, Apollo's DOMAIN and Stratus' VOS) were attempts to keep more Multics functionality (Prime used to refer to PRIMOS as "Multics in a matchbox") on smaller boxes. But were they "learning" from the Multics experience, or merely dealing with the fact that computers were not yet powerful enough to meet their needs? -- ROGER B.A. KLORESE MIPS Computer Systems, Inc. phone: +1 408 720-2939 MS 4-02 928 E. Arques Ave. Sunnyvale, CA 94086 rogerk@mips.COM {ames,decwrl,pyramid}!mips!rogerk "Two guys, one cart, fresh pasta... *you* figure it out." -- Suzanne Sugarbaker
baum@Apple.COM (Allen J. Baum) (02/09/90)
[] >In article <486@taniwha.UUCP> paul@taniwha.UUCP (Paul Campbell) writes: > >Look closely at the architecture (try the Byte article), the NeXT board >multiplexes data/addresses from the '030 to make it look like the '040 >(or NuBus) - this means that their silicon is probably already set up >for the '040, my bet is that Steve is just waiting for the parts in >quantities/prices that he can afford to ship them. BTW I think he lucked >out on the '040 bus interface, my guess is that the reason that it looks >so much like NuBus is because Apple buys the most 68ks. Um, I hate to break it to you, but the '040 has separate 32 bit address and data busses. They are not mulitplexed. They do not resemble the multiplexed address/data of Nubus. -- baum@apple.com (408)974-3385 {decwrl,hplabs}!amdahl!apple!baum
srg@quick.COM (Spencer Garrett) (02/09/90)
In article <486@taniwha.UUCP>, paul@taniwha.UUCP (Paul Campbell) writes: > Look closely at the architecture (try the Byte article), the NeXT board > multiplexes data/addresses from the '030 to make it look like the '040 The '040 does not multiplex address and data lines. The signals coming out of that chip look very much like those of the '030. The 88000, on the other hand, does multiplex addresses and data on its M bus. I'm not familiar with NuBus, so I won't speculate whether this has any bearing on that.
wbeebe@rtmvax.UUCP (Bill Beebe) (02/09/90)
In article <8859@portia.Stanford.EDU> underdog@portia.Stanford.EDU (Dwight Joe) writes: >My suspicions are confirmed. The NEXT computer is in trouble. Yea, yea, sure, sure. I've read the same articles in the various trade rags including, of all things, the Wall Street Journal. I don't care particularly for Steve Jobs (who I consider to be a high-tech snake-oil salesman). But the thing that bothers me is the very short-sighted attitude that we now *all* seem to have; if it isn't an instant financial success, then there must be something wrong with it. We've grown addicted to hyper growth and excessive greed. NeXT's problems are a symptom of an even greater problem with the great Shakers and Movers of American Technological Industry. >NEXT can only be saved if Steve Jobs replaces the 680X0 >with RISC processor like the Sparc chip. In all compute >intensive applications, the Sparcstation I beats the NEXT >timewise. Worse, NEXT costs MORE than a Sparstation I. Replacing the 68030 with a Sparc chipset is *your* personal view. Mine tends towards the 88000 (because of the Moto roots), the i860 (I like a long shot), and even the Transputer (an even longer shoot). Replacing the 68030 with the 68040 and making the necessary changes in the motherboard to support its architecture makes a lot more sense anyway. Besides, haven't you seen Moto's 40 ads? The 68040 runs 20 mips at 25 Mhz! The Sparc only pulls a measly 18. Ain't science wunnerful? :-). >Too, the extra gadetry (like the DSP chip) on the NEXT is >unlikely to be used by engineers doing compute-intensive >applications. The DSP might help out in making >a realistic video game; otherwise, its deadweight. >What difference does it make if you can play Beethoven's >fifth on the NEXT? This engineer likes the classics, and being able to play them on a computer is very appealing. What's even more appealing is doing sophisticated composition, human voice tract modeling for speech synthesis, and who knows what else? The DSP doesn't get in the way of "compute-intensive" work; it stays put until called into service. I don't know what set you off on a tangent with regards to the NeXT, but it sure beats the hell out of contemporary offerings from Apple, IBM, and yes, dare I say it? Even glamorous Sun, which in spite of it's best efforts at licensing to cloners and other chip makers has failed to sweep the other inferior chip designs into the ashbin of history. The much-maligned IBM PC, with all its warts, has one overwhelming redeeming quality; it's damn cheap. And the aps the run under the Son of CP/M, otherwise slurred as Mess-Dos, have a very high quality user interface and equally high utility. I know I aught to stop because I'm sounding like a jerk, but hey, you can't do it any better than here on the net! Give me a Sparc-based machine that sells for $1899 complete (read that with sufficient memory AND disk space to do useful work) with tools of equal quality, depth and breath that I have grown to expect on PCs, and I'll close my little mouth..
mac@ra.cs.Virginia.EDU (M. Alex Colvin) (02/09/90)
since this is the arch group, i'll suggest that multics required computers with lots'a hardware to run, where unics ran on programmable data processors. perhaps the analogy could be to the xerox star, a nifty box that cost too much, and sold too little. followup to comp.os.nostalgia
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (02/10/90)
In article <7356@pdn.paradyne.com> alan@oz.paradyne.com (Alan Lovejoy) writes: | But that was a strange question to ask. The natural question would seem | to be "What has SPARC got that the 88k, and the Rx000, don't?" You may like Sun, or hate them, but it seems unwise to ignore them. By going SPARC a vendor gains access to a very large number of applications already ported to run on the CPU. SPARC has software, and (as MS-DOS proves) a CPU and o/s will be popular is there's enough good software for it, even if the o/s and CPU are at least a decade behind state of the art. The fastest production versions of SPARC seem to beat the fastest production versions of CISC, and SunOS is reasonably up to date as a unix version, so there are obvious benefits from SPARC which the 88k didn't offer when it was selected. Oh, and SPARC is multi-source, too. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "Stupidity, like virtue, is its own reward" -me
bcase@cup.portal.com (Brian bcase Case) (02/10/90)
>>Look closely at the architecture (try the Byte article), the NeXT board >>multiplexes data/addresses from the '030 to make it look like the '040 >>(or NuBus) - this means that their silicon is probably already set up >>for the '040 > >Um, I hate to break it to you, but the '040 has separate 32 bit address and >data busses. They are not mulitplexed. They do not resemble the multiplexed >address/data of Nubus. Ah, but there is a mode bit (or is this one of the things that is set at reset time like on the R2000/R3000...) that allows the 040 address and data bus to be literally tied together to form a muxed bus. Still, I seriously doubt this had anything to do with the system design of the NeXT. Also interesting about the 040, the output impedance of the output drivers on the chip is selectable at either 4 or 30 Ohms (I might have the numbers slightly wrong). The idea is that you can have speed at the cost of terminating properly or you can have simplicity (no termination) at the cost of speed.
alan@oz.nm.paradyne.com (Alan Lovejoy) (02/10/90)
In article <2113@crdos1.crd.ge.COM> davidsen@crdos1.crd.ge.com (bill davidsen) writes: >In article <7356@pdn.paradyne.com> alan@oz.paradyne.com (Alan Lovejoy) writes: >| But that was a strange question to ask. The natural question would seem >| to be "What has SPARC got that the 88k, and the Rx000, don't?" > > You may like Sun, or hate them, but it seems unwise to ignore them. True. >By going SPARC a vendor gains access to a very large number of >applications already ported to run on the CPU. SPARC has software, and >(as MS-DOS proves) a CPU and o/s will be popular is there's enough good >software for it, even if the o/s and CPU are at least a decade behind >state of the art. And one also gains a very tough competitor--you know, those people we agreed it would be unwise to ignore? Perhaps this explains the crowd of Sun-clones curently flooding the market... > The fastest production versions of SPARC seem to beat the fastest >production versions of CISC, and SunOS is reasonably up to date as a >unix version, so there are obvious benefits from SPARC which the 88k >didn't offer when it was selected. Oh, and SPARC is multi-source, too. Well, I guess customers would feel more secure if the 88k were second-sourced. After all, Motorola just might go belly up one of these days :-). And if the competition drives the SPARC price down to firesale prices, where will the R&D money to produce the next generation come from? Oh, SPARC is scaleable? That's nice. It hasn't seemed to provide it with any particular clock rate advantage so far (both the SPARC and the 88k are in production at 33MHz, for instance, in spite of the fact that the SPARC started out with approximately a years lead (perhaps more) over the 88k in introduction schedule). Commoditization is the kiss of death in the long run for any product which must quickly evolve over time, since it leads to stagnation because the cost of improvement can't be justified--and is in fact resisted by the customers who have grown dependent on its current form. I'm sure we can all think of examples of this in tech history. But each MPU has its advantages and disadvantages. None of them are "perfect," and can't be, since that's relative anyway. ____"Congress shall have the power to prohibit speech offensive to Congress"____ Alan Lovejoy; alan@pdn; 813-530-2211; AT&T Paradyne: 8550 Ulmerton, Largo, FL. Disclaimer: I do not speak for AT&T Paradyne. They do not speak for me. Mottos: << Many are cold, but few are frozen. >> << Frigido, ergo sum. >>
zenith-steven@CS.Yale.EDU (Steven Ericsson Zenith) (02/11/90)
In article 14253 Alan Lovejoy (alan@oz.nm.paradyne.com) writes >Commoditization is the kiss of death in the long run for any product which >must quickly evolve over time, since it leads to stagnation because the >cost of improvement can't be justified--and is in fact resisted by the >customers who have grown dependent on its current form. I'm sure we can >all think of examples of this in tech history. I can't see the historical evidence to support this at all. My recent experience at INMOS and observation of the success of INTEL, and recently released designs from that company show otherwise. Even though subsequent design evolutions of the Z80 did not succeed, I wonder if anyone at Zilog would blame the commoditization of the Z80 for susequent mediocrity? Regards, Steven . Steven Ericsson Zenith * email: zenith@cs.yale.edu Department of Computer Science | voice: (203) 432 1278 Yale University 51 Prospect Street New Haven CT 06520 USA. "All can know beauty as beauty only because there is ugliness"
paul@taniwha.UUCP (Paul Campbell) (02/12/90)
In article <38464@apple.Apple.COM> baum@apple.UUCP (Allen Baum) writes: >>In article <486@taniwha.UUCP> paul@taniwha.UUCP (Paul Campbell) writes: >>Look closely at the architecture (try the Byte article), the NeXT board >>multiplexes data/addresses from the '030 to make it look like the '040 > >Um, I hate to break it to you, but the '040 has separate 32 bit address and >data busses. They are not mulitplexed. They do not resemble the multiplexed >address/data of Nubus. I know they are not physically multiplexed ... but look at the (synchronous) timings of TS (start) and TA (ack) (compared with the asynchronous ones used on previous 68K chips) and how much they are like the equivalent NuBus lines. You can actually put the '040 into a mode where you can physically tie the address/data lines together (and run them as a multiplexed bus). I sure there are also other reasons for NeXT using a multiplexed bus (lower pin counts on their VLSI for example) but I think that the idea of multiplexing address/data lines on a workstation (something that is rather counter-intuitive, not something one would usually do deliberately) is really rather a bright one! and a really good match for the NuBus too (since you have to do it anyway). Paul -- Paul Campbell UUCP: ..!mtxinu!taniwha!paul AppleLink: CAMPBELL.P You know there's something wrong when 100,000 people marching in Moscow makes page 1 and 400,000 in Washington doesn't .....
richf@adiron.UUCP (Rick Fanta) (02/12/90)
There's been a LOT of speculation in this series about Jobs' next move. I have been reading a variety of magazines over the past few months toward this end too, and offer a few hopefully interesting insights. First, the next NeXT is likely to offer a kick ass color graphics system (probally as an option). Remember, Jobs is a part owner in Pixar, and he needs to find a niche market for his machine (just as the Mac succeed, at least in part due to the fact that it answered the needs of the desktop publishing). It has been a very poorly kept secret that color is in the works. In article in EE times (last September?) when the Renderman standard was announced, Jobs raved of the possibilities of 3D desktop movies and the attributes of the Intel i860. Commitments to Motorola may prevent him from using it though (I don't know a damn thing about this, but if he's getting a cut-rate deal on MC micro- processors, they probably wouldn't take too kindly to his going to the oppo- sition for a RISC to do his ray-tracing). I think I remember an article speculating that a MC96000 32-bit DSP chip would be used to boost the vidieo, but it I don't think it would have enough muscle for Renderman (then again, what do I know). Second, the next NeXT is likely to use the MC68040 instead of a RISC. Although it may make you nauseous, think like a marketing person. If he switches to a RISC, compatibility with the small but promising software base that's already out there is toast! This pisses off users and software developers too; not good for a small company trying to build a user base. Besides, if NeXT was going to build a machine around the MC88000, they could have done so already. The fact that they scheduled the announcement of the new machine for sometime in the first quarter (me thinks) of this year coincides strangely with the recent (expected) announcement of the new MC68040. Performance is a problem, but there maybe a two prong solution (my speculation) to this problem within the next year or two. The slow optical disk access is mainly due to the multiple passes that are needed. I am hopeful (as is Jobs too, I'm sure) that this problem will be solved in the near future. I'm sure that the 40Mb hard disk buffers will help this problem in the short term though. The other solution to the performance problem may be some sort of Postscript coprocessor (could the same RISC being used for the kick-ass graphics be used for this as well???). From what I hear, Postscript puts quite a burden on any system, so this seems like a good move if it is possible. Comments (and more news) from more knowledgeable sources would be much appreciated. ----------------------------------------------------------------------------- Rick Fanta, PGSC, New Hartford, New Yawk
melling@psuvax1.cs.psu.edu (Michael D Mellinger) (02/12/90)
In article <430@adiron.UUCP> richf@adiron.UUCP (Rick Fanta) writes:
Second, the next NeXT is likely to use the MC68040 instead of a RISC. Although
it may make you nauseous, think like a marketing person. If he switches to a
RISC, compatibility with the small but promising software base that's already
out there is toast! This pisses off users and software developers too; not
good for a small company trying to build a user base. Besides, if NeXT was
going to build a machine around the MC88000, they could have done so already.
The fact that they scheduled the announcement of the new machine for sometime
in the first quarter (me thinks) of this year coincides strangely with the
recent (expected) announcement of the new MC68040.
How much work would it be for software companies to support a RISC version
of their NeXT software? I would think that most of their software is written
in C, or another high level language, and would need nothing more than
a recompile. How much assembly level optimization is done in most commmercial
software?
-Mikebaum@Apple.COM (Allen J. Baum) (02/13/90)
[] I wrote: >>Um, I hate to break it to you, but the '040 has separate 32 bit address and >>data busses. They are not mulitplexed. They do not resemble the multiplexed >>address/data of Nubus. > In article <xxx> bcase@cup.portal.com (Brian bcase Case) writes: >Ah, but there is a mode bit (or is this one of the things that is set >at reset time like on the R2000/R3000...) that allows the 040 address >and data bus to be literally tied together to form a muxed bus. Still, >I seriously doubt this had anything to do with the system design of the >NeXT. > >Also interesting about the 040, the output impedance of the output >drivers on the chip is selectable at either 4 or 30 Ohms (I might >have the numbers slightly wrong). The idea is that you can have >speed at the cost of terminating properly or you can have simplicity >(no termination) at the cost of speed. I looked it up, and yup, they do it. There are 3 mode lines which are read during reset for configuration. One changes the timing so address and data are never driven at the same time; thus, you can short address and data busses together to form a multiplexed bus. Another is the 4/30 ohm driver selection, and the third is Data Latch Enable" mode, where read data can be latched by the DLE signal instead of the clock. They give examples of where this is advantagous. -- baum@apple.com (408)974-3385 {decwrl,hplabs}!amdahl!apple!baum
freek@fwi.uva.nl (Freek Wiedijk) (02/13/90)
In article <C:t?h_2@cs.psu.edu> melling@psuvax1.cs.psu.edu (Michael D Mellinger) writes: >How much work would it be for software companies to support a RISC version >of their NeXT software? I would think that most of their software is written >in C, or another high level language, and would need nothing more than >a recompile. How much assembly level optimization is done in most commmercial >software? As far as I know WriteNow, the default word processor on the NeXT, is completely coded in assembly. There's glory for you! -- Freek "the Pistol Major" Wiedijk Path: uunet!fwi.uva.nl!freek #P:+/ = #+/P?*+/ = i<<*+/P?*+/ = +/i<<**P?*+/ = +/(i<<*P?)*+/ = +/+/(i<<*P?)**
vanroy@vega (Peter Van Roy) (02/13/90)
We bought a NeXT machine for use in our research group, over a Sparcstation.
We use it partly for doing large simulations and for generating traces.
The NeXT is not very fast, but it definitely has its advantages.
Two reasons that were very important for us:
1. The NeXT's operating system is Mach, so swap space is limited
only by available disk space.
2. The removal optical disk. Wonderful for storing large traces and
for doing backups.
Peter Van Roy
vanroy@ernie.berkeley.edu