Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×

Sun Cancels UltraSPARC IIIi+ 97

Posted by CowboyNeal from the belated-goodbyes dept.
Doctor Memory writes "El Reg is reporting that Sun has finally come clean and admitted that they have killed the UltraSPARC IIIi+ chip. According to John Fowler, Sun's server chief, 'We canceled it last fiscal year to focus on the ramp (up) of UltraSPARC IV+, Niagara and Niagara 2.' Sun has had great success with its new Niagara line, and with it's line of AMD-based systems."
This discussion has been archived. No new comments can be posted.

Sun Cancels UltraSPARC IIIi+ More

Sun Cancels UltraSPARC IIIi+

Comments Filter:

  • UltraSPARC IV is the replacement for UltraSPARCIII (Score:5, Informative)

    by IYagami ( 136831 ) on Friday September 01, 2006 @07:29AM (#16023035)
    I think that Sun UltraSPARC IV can replace UltraSPARC III smoothly (they share the same socket, Sun can provide help doing this, c...) According to: http://www.sun.com/processors/UltraSPARC-IV/index. xml [sun.com] "Executing on Sun's Throughput Computing strategy, the dual-thread UltraSPARC IV processor marks the first milestone in Sun's Chip Multithreading (CMT) roadmap (...) It protects customers' investments through 100 percent application binary compatibility, and can provide an upgrade path for current UltraSPARC III processor-based systems " It's like replacing an Athlon Socket AM2 Single Core with a Dual Core (I think...)
    • Except the smallest USIV box they make is the 490, 4RU and non-hs power supplies. Get me a 240 replacement with a USIV and I'll be all over it. Or better yet figure out how to add HS dual power and the USIV to the 210.

      • Re: (Score:3, Informative)

        by IYagami ( 136831 )
        I'm afraid you're right (and I'm wrong).

        According to
        http://www.sun.com/servers/entry/v240/specs.xml [sun.com]
        or
        http://www.sun.com/servers/entry/v210/specs.xml [sun.com]
        There are no processor upgrades avalaible to the Sun V240 or Sun V210 Server. I suppose that the upgrade is only avalaible to the Big Iron Systems.

        • Re: (Score:1, Informative)

          by Anonymous Coward
          I'm under NDA so I'm posting as AC, but apart from the T1000/T2000 (which are nice) there are 3 new boxes about to be announced to modernise the low end range. I've deleted the presentation, but from memory US IV+ CPUs and modern busses are the main benefits. I think they were talking about a 15-30% performance improvement.

          Expect the model number to be incremented by 5.

      • Re: (Score:1)

        by Big Jason ( 1556 )
        The V490 is 5U, not 4.

      • Re: (Score:2, Informative)

        by eclectus ( 209883 )
        uh, the V490 does have hotswap power supplies. As a matter of fact, all servers that sun makes that have dual power supplies are hotswappable.

        • Re: (Score:2, Informative)

          by htd2 ( 854946 )
          Not quite, the V210 and X2100, the X2200 and the T1000 do not have dual hot swapable PSU's.

          They are all 1U boxes and most 1U boxes don't have dual hot pluggable PSU's.

          • Re: (Score:2, Insightful)

            by setantae ( 103317 )
            Well the V210 doesn't have dual power supplies, so you are in direct agreement with the post you replied to, which said: "As a matter of fact, all servers that sun makes that have dual power supplies are hotswappable."

      • Re: (Score:3, Informative)

        by illumin8 ( 148082 )

        Except the smallest USIV box they make is the 490, 4RU and non-hs power supplies. Get me a 240 replacement with a USIV and I'll be all over it. Or better yet figure out how to add HS dual power and the USIV to the 210.

        This is FUD plain and simple. The V490 was based on the V480 and both of these products have had HS power supplies since their release. The 440 came out later which has decent processor speed but a little less memory bandwidth than the full-fledged V490.

        From this page [sun.com]:

        RAS features include ho

        • Whhops, sorry about that, I had just been looking at the ass-end of a 490 and noted the PS's didn't have a release on the rear, but now looking through sunsolve I see the the ac inputs are fixed, but the guts of the PS's are HS. My bad

      • Re: (Score:2)

        by himself ( 66589 )
        And the V490 has some kind of fruity European 208V power supplies, which only the V890 seems to share. Gaaah!
    • You are refering to a different family of processors.

      The UltraSPARC III/IV/IV+ use the Gigaplane Interconnect which is designed to support >4 Modules, the UltraSPARC IIIi uses JBus which is designed to support 4 CPU's. The two are not compatible so you cannot drop a US IV+ into a IIIi slot.
    • USIV is the replacement for the USIII, upgrading an USIII system is as easy as calling Sun and getting them to swap in the USIVs (very similar to replacing a single core athlon with a same-socket dual core athlon, except the motherboards get swapped out, not just the processors)

      Unfortunately, the article is talking about the USIIIi+ being cancelled, which was the upgrade for the USIIIi

      In the server space, Niagara based systems are the replacement for the USIIIi, but there is not really a replacement o
      • Only in the mid to high end boxes. the 6800 and up (except the 10k and 12k) can mix and match USIII and USIV uniboards but we're talking about under 12 RU chasis.
    • The UltraSPARC IIIi (not III) uses a different socket and different interconnection bus. It is JBus on the IIIi, and something else (don't remember, or never knew) on the III and IV. The IIIi an III are not socket compatible. That said, Sun should be able to make a motherboard with IV chips fit in the IIIi enclosures if ther eis a market for it. I doubt there is. The Niagara and AMD boxes pretty much have the volume servr market covered between them, and Dual Niagara II boxes will quite likely outperform th

  • who cares() (Score:4, Funny)

    by Anonymous Coward on Friday September 01, 2006 @07:34AM (#16023053)
    {
        return NOBODY;
    }

  • Of course they've been having great success.... (Score:5, Funny)

    by Chineseyes ( 691744 ) on Friday September 01, 2006 @07:34AM (#16023054)
    Spammers having been letting me know about Niagara for years

    • Re: (Score:3, Funny)

      by Shaper_pmp ( 825142 )
      Dear sir,

      My name is Mtubutu Balawe, and I beg your help. A wealthy government official has recently died, and I require to move large amounts of erectile dysfunction pills from my home to the USA...
      • Good day dear Customer,

          Achieve astounding results in bed with these products designed to make any man a winner

          Get a several month supply of everything you need - in seconds

          Hot deals on stuff produced by well-known brands from worldwide.

        Cjialis from 3, 75 $
        Njiagra from 3, 35 $
        Valjium from 1, 25 $ ...
    • "Niagara Falls.... Step by step.... Inch by inch...." ;-)
  • You can't delay something that was cancelled. Sun was just delaying the expectation for people clamoring for the new UltraSparc IV. I can't wait to get my hands on some at work. We could consolidate some of our Java servers together in our testing environments at a lower cost than buying more of the BIG IRON systems. I'll be curious to see what systems they're going to wrap around these processors. Are the current lines going to get faster CPUs? Are they going to create whole new lines of Servers?

    • Re: (Score:1)

      by htd2 ( 854946 )
      Sun has just released a faster US IV+ (1.8 Ghz) that will probably appear in the 490, 890 and above.

      If the article is correct then there will be no replacement for the US IIIi

      The T2 will replace the T1 but this will be a system upgrade rather than a module change. Sun tends to provide module upgrades for the mid range and high end SPARC units as well as for the AMD based boxes (generally).

      There will be a whole new server platform in 2008 based on the Rock CPU called Supernova this will be a complet
      • I am really hoping for a T2-based laptop. The T1 was designed for low power consumption, and so would fit nicely in a laptop, if not for the anaemic FPU performance. The T2 should address that. 90% of the time when I am mobile, I could live happily with a single-core T1/2, and if you can turn off the other 7 then this could make a really lean chip for mobile use.

        Sun's current laptop offerings are really depressing. They are big, heavy, and offer very poor price/performce ratios (in my experience, Ult

  • We looked at the name (Score:5, Funny)

    by Colin Smith ( 2679 ) on Friday September 01, 2006 @07:45AM (#16023087)
    "And thought, damn, nobody's going to buy that, people aren't stupid. Everyone knows that IIIi+ is almost 4, it has to be 3.8 at least so we just decided to rename it 4. Sorry... IV... IV+... Would another plus be overkill do you think?" said a Sun spokesman.

     
    • You know, that's the second time I've seen that website, and I'm still not impressed.

      Class action lawsuit due to patent infringement of Kodak's patents, related Java.
      There was a patent lawsuit. I don't know where you get "class action" part from. Sun also settled and licensed the technology immediately after the judge decided they were infringing. So you proved... how responsible Sun is?

      Sun has done some questionable things, but those aren't it.
      • It seems applix7 has a great many "check out X's wrongdoing" posts to various threads. It seems to just be a (hobby|quest|purpose in life|self-appointed duty) of (his|hers) to point out these things.

        I'm sure it helps to have people keep tabs on big companies. Sun's about the least of my worries when it comes to companies misbehaving, though.

        Hey, speaking of Sun and different lines of chips, what about the clockless chip I used to hear about? Is that still in development, or has it been canned?

        • It seems applix7 has a great many "check out X's wrongdoing" posts to various threads.

          Yep, that's why I decided to speak up. His goal is sound enough, but his research is severely lacking. His site makes no mention of many of Sun's earlier intentional and unintentional misdeeds. For example, Jini was not a technology that just appeared out of the blue. It was a complicated aquisition that ended up with accusations of technology theft and poor morals.

          That's all in the past now, but it's quite a bit more dam

        • Re: (Score:2)

          by IvyKing ( 732111 )
          Hey, speaking of Sun and different lines of chips, what about the clockless chip I used to hear about? Is that still in development, or has it been canned?


          The US-IIIi is supposed to have some asynchronous (i.e. clockless) logic in it.

      • Sun has done some questionable things, but those aren't it.
        Their handling of sun4m and hardware thereof in not-so OpenSolaris 10 seems to be a bit (more?) questionable given that there was a workable, buildable version of it- and I doubt svn goes back 3-4+ years to s10_22. It'd at least be better to have them put it back in, along with all the hardware eol'ed along the way(and fully documented). Then Sun can just close the book on that platform for good once there is something of a "reference build" that wo
        • I just read through your thread on osol-discuss (read it here [opensolaris.org]). I think Casper's and Bryan's replies to you explain everything. I am not quite sure what your beef is. Just because OpenSolaris does not compile on sun4m anymore? Most of the Solaris 10/OpenSolaris features do not support sun4m. Of course, if you want to, you can always contribute code to the project to enable sun4m support. Good luck with that.
          • OK, I just read the thread, and my only remaining question is: was support for line breaks in text messages introduced after sun4m?

        • Re: (Score:1)

          by turgid ( 580780 )

          Their handling of sun4m and hardware thereof in not-so OpenSolaris 10 seems to be a bit (more?) questionable given that there was a workable, buildable version of it- and I doubt svn goes back 3-4+ years to s10_22. It'd at least be better to have them put it back in, along with all the hardware eol'ed along the way(and fully documented).

          Nonsense. Solaris 10 is too big for sun4m machines. Run NetBSD or Linux insted. Let it lie. Solaris 10 is the future. sun4m is the (distant) past.

    • Wow that site is full of Green Party rants. Waste of time, unless you're looking for a good laugh. Most of the companies are on there for the "misdeed" of outsourcing labor. Why can't people accept that some guy in India could do your job for half the cost? Nobody *owes* you a job.

      --
      If you want to accuse me of being a Republican, then you'd be way off. If there were a Pragmatic Party, I'd join it.

  • Ho Hum (Score:4, Interesting)

    by gentimjs ( 930934 ) on Friday September 01, 2006 @07:57AM (#16023131) Journal
    Is anyone surprised? USIV and T1 are by far Suns big cash cows. While I'm typing this at an USIIIi powered workstation, and would love a simple upgrade, I cant say that this upsets me in the least. I'd far rather Sun's R&D go into thier higher-end stuff than entry-level stuff, since it will push the current "high end" crap down to the level that us mere mortals will be able to get it! :-D

    • Re:Ho Hum (Score:5, Insightful)

      by P Fayers ( 470140 ) on Friday September 01, 2006 @08:26AM (#16023243) Homepage
      The concentration of R&D on the high end and then deriving a cheap version of the chip is what has been causing Sun trouble for a while. Intel tried the same thing with Itanium and it didn't work for them either.

      Sun's current method, introduce the low end chip - Niagara - first and then build up to the high end stuff (the Rock CPU) seems to be a much better idea. Produce the high volume stuff first and use the revenue from that to produce the high end, high margin stuff.

      • Re: (Score:3, Insightful)

        by Konster ( 252488 )
        Yes, but Intel has a long history of chips besides the Itanic that have done well with the top down method of development (marketing).

        Xeon anyone? Don't laser cut a few pins needed for SMP functionality, sell the same CPU as being better, therefore, then profit.

        Opteron? Oo...how is an Opteron very much different than a regular A64...really?

        All CPU makers have a marketing department that had sold the same bit under different names for different amounts of cash, with varying bits of cache and power management

        • Re: (Score:1)

          by htd2 ( 854946 )
          This all depends on which SAPRC chip you are refering to. T1 for the right workloads is pretty much on it own and you generally need 4 cores to catch it. US IV+ is less of a stellar performer but then people buy very big Sun's for a lot of reasons only one of which might be single CPU throughput.

        • Re: (Score:2)

          by Glock27 ( 446276 )
          Opteron? Oo...how is an Opteron very much different than a regular A64...really?

          Opteron has one more Hypertransport line for interprocessor communication. That was the reason for Socket 940 vs. Socket 939. I need to read up a bit to completely understand Socket AM2 vs. Socket F (great names there, AMD marketing types :P). Socket F apparently supports fully buffered DIMMs for one thing.

          Otherwise, though, as you point out they're very similar, and may well come off the same wafers.

          • As I recall, all AMD x86-64 chips of a particular generation have the same core design. If there is a manufacturing defect in any of the cache, then it is marked as 1MB-cache Opteron, a 512KB Athlon 64 or a 128/256KB Sempron[1], depending on how severe the defect is. If only one of the HT busses works, then it is marked as either an Athlon 64 or a 1-2 way Opteron. If more work, then it is sold as a 4- or 8-way Opteron. If there is a defect in one of the cores, then it is sold as a single-core Athlon.

            Th

    • Re: (Score:3, Insightful)

      by Nutria ( 679911 )
      I'd far rather Sun's R&D go into thier higher-end stuff than entry-level stuff, since it will push the current "high end"...

      Are they even going to have more than a token low-end HS anymore, or relegate the low-end to AMD64, with overlap in the middle?

      • That's a possibility. When you look at the upcoming AMD stuff (Quad core, Newisys Horus chipset) you could potentially build a 128 core opteron, the current biggest boy on the sparc side is the 144 core USIV+ 25000. You give me DSD's and a supermassive opteron and we can start talking about sun killing sparc.

    • Re: (Score:2)

      by himself ( 66589 )
      gentimjs wtrote:
      >
      > USIV and T1 are by far Suns big cash cows. While I'm typing this at an USIIIi powered workstation, and would
      > love a simple upgrade, I cant say that this upsets me in the least. I'd far rather Sun's R&D go into thier higher-end
      > stuff than entry-level stuff, since it will push the current "high end" crap down to the level that us mere mortals
      > will be able to get it!
      >

      Right, well, I buy for a university, not for personal use, so I have to disagre
      • I seem to recall that sun still pitches that "100% binary compatibility" thing ?

        • Re: (Score:2)

          by himself ( 66589 )
          gentimjs asked:
          >
          > I seem to recall that sun still pitches that "100% binary compatibility" thing ?
          >
          Good point, but that's only their own stuff, you know?

          They blew a lot of smoke about a partneership with Oracle, too, which I take with a grain of salt: until, for example, Oracle Enterprise Manager is certified on Solaris 10 and Sun x86, I will have to run it on Solaris 8 or 9, SPARC. Which, obviously, is lame, since development on these products trailed of
  • I hope niagara and UltraSparc IV to have prices cut with this move.
    • Probably not. The machines which were due to get UltraSPARC-IIIi+ chips are better replaced by Sun's Opteron range (at least until Niagara 2 arrives) which are already cheaper, faster and more power efficient than the UltraSPARC-IIIi based systems.

      There isn't a direct competitor for the v440 (4 CPU sockets, 4U chassis) but it's so expensive you could buy a small x4600 for the same money.

      • Re: (Score:1)

        by htd2 ( 854946 )
        In fact you could buy a large X4600 for less money than a V440. A X4600 with 4 x Dual Core 885 Opterons and 32 GB of RAM costs $39K, a 4 way V440 with 32GB of RAM costs $40K. The X4600 gives you well over 2x the performance, its more expandible and it costs less.

  • What the i Stands For (Score:4, Funny)

    by CastrTroy ( 595695 ) on Friday September 01, 2006 @08:27AM (#16023245)
    I guess we know what the "i" stands for in UltraSpark IIIi. It's for imaginary. Like in math.

    • Re: (Score:2)

      by mnmn ( 145599 )
      The i means a little weaker than the regular.

      But the e standard for exceptionally lousy, as in Ultrasparc IIe

      Dont buy it.
    • No no... The UltraSPARK processor is the KDE version. For reasons unknown, Sun is moving forward with the (Gnome based) JDS, so you'll have more luck in tracking the UltraSPARG.

    • Re: (Score:2)

      by IvyKing ( 732111 )
      The IIIi is quite real (I'm replying on a IIIi machine), it is the IIIi+ that's been imaginary.

    • Re: (Score:2)

      by turgid ( 580780 )

      But.. but... imaginary numbers are no more imaginary than real ones.

      /me ducks.

  • "Its" is possessive. I know it's no big deal but I just couldn't resist reminding everyone that virtually every story posted here is written sloppily and goes entirely unedited these days. =/

  • Run Windows on an AMD based SUN box and I bet Hell has frozen over! At least that is what I would have thought 10 years ago!

    • Nah, I preferred running Windows on a SPARC based Sun box using an i386 daughter board. And that was 10 years ago!

  • Office Space (Score:3, Informative)

    by Double Mint Len ( 998680 ) on Friday September 01, 2006 @09:53AM (#16023620) Journal
    In August of 2004, I was in Sillicon Valley, and I visited the Sun headquarters.

    It was quite possibly the most disorganized, un-coordinated place on earth. If Disney World was run by a single quadrapalegic, I think it would still be more effecient.

    We took a tour, and the guide honestly said that the only really interesting thing they had going on, was that they could bring their dogs to work....uh thanks man. really technologically advanced
  • but I don't want to have a noisey rackmount system. I think it would be interesting if Sun made an inexpensive "developer's kit" for the UltraSparc T1. Like Mac Mini form factor (or mini tower), and under $1500. Actually they could just make it an ATX motherboard and let the developer figure out power supply, case, ram, etc. As much as I adore Sun's cases, they are probably very expensive to produce.

    Tempted to just get a SunFire T1000 for $3500 anyways just to experiment on. It's a pretty good price for a s
    • As would I, but I suspect that the volume isn't there, and due to the single floating point unit, the Niagara I problem doesn't make a very good desktop box. On the other hand, a dual Niagara II box with a 16 lane PCI-e slot would rock as a personal workstation, but I suspect if Sun builds it, it will be priced out of reach for the masses.
      • If people can make dev boards for FPGAs for under $100 (http://digilentinc.com/) then Sun ought to be able to make a handful of affordable multiprocessor boards for us to play with.

        I don't see much value in floating point. I'm not arguing to make it a desktop, that would be pointless since there are zero real games for Sparc. I just think it would be nice for experimentation and number crunching. (real number crunching doesn't use highly inprecise floats).

        I think a pair of dual core Athlons would run cir
  • I seem to recall McNealy talking about this before he stepped down. I don't know of anyone who expected the IIIi to get anything more than maybe a speed bump, and even that was questionable. This is utterly not news.

    Oh but wait--it's The Register. Any product that gets cancelled is worth talking about, if you can spin it as a death knell for the company du jour.
  • Really. I want to know. I've wondered that for years, and now with Sun's outlook kinda shakey, why do they still do it? Kill the SPARC, and just go with something from Intel or AMD? What advantage does the SPARC have that the offerings from the Intel and AMD don't?

Slashdot Top Deals

"The four building blocks of the universe are fire, water, gravel and vinyl." -- Dave Barry

Close