Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
Intel Hardware

Intel V8 Octa-Core System, Full Performance Tests 123

Posted by Zonk from the friday-afternoon-spec-oggling dept.
MojoKid writes "In the April time frame, details of Intel's dual-socket 8-core system dubbed 'V8' became available but only preliminary performance numbers were shown. The platform consists of quad-core Xeon processors in an Intel 5000X chipset-based motherboard, along with FBDIMM (Fully Buffered DIMM) serial memory. A follow-on article at HotHardWare goes into significantly more detail on the platform and showcases many more performance metrics on a Windows Vista 64-bit installation. The POV-Ray and Cinebench 95 benchmark numbers alone are something to smile about. 'Intel's V8 isn't about promoting a platform as much as it is a show of strength and a glimpse of things to come. What V8 and QuadFX show is that both Intel and AMD are on a path to offering true, enthusiast-class, dual-socket platforms. And that's a good thing. Perhaps AMD is a little further down the path thanks to a more tweaker-friendly motherboard in the QuadFX-compatible Asus L1N64-SLI WS, but until consumers have more motherboards to choose from and perhaps quad-core processors from AMD, we can't very well declare that the time for QuadFX has arrived. One motherboard does not a platform make.'"
This discussion has been archived. No new comments can be posted.

Intel V8 Octa-Core System, Full Performance Tests More

Intel V8 Octa-Core System, Full Performance Tests

Comments Filter:
  • Now, when will SuperMicro come out with a workstation based on this platform for me to savor the hum of under my desk?
  • FB-DIMMS suck for gameing and the next chip may let you use DDR2 ECC and have more pci-e lanes and maybe SLI / CROSS FIRE.
    Right now you can get a 2-4 cpus amd system with 2 high end video cards and hard RAID and still have pci-e lanes left and that system maybe better at high end video work.

    AMD systems have the pci-e lanes for 2 full pci-e x16 lanes ,2 x4 lanes, and lanes for on board sas raid cards and pci-x at the cost of 4 pci-e lanes.

    It's too bad the macpro uses the same chip set the lack of pci-e lanes
    • I think this is a classic argument by someone distracted by specs without actually getting concerned with real world differences or really going through the numbers.

      You've been saying that a lot, but basically, an x16 slot is not going to net twice the graphics performance as the same card in an x8 slot. The benchmarks I've seen show about a 2% difference, which few people would notice.

      I suspect that the same would be true about your other other points. A single PCIe lane has the bandwidth to handle five

      • Re: (Score:3, Interesting)

        by Joe The Dragon ( 967727 )
        the build in ports are on the pci-e bus as the 5000x chip set uses pci-e lanes for part of the chipset to chipset link and most pci-e hardware raid cards are x4 with x8 for cards with more ports http://www.3ware.com/ [3ware.com].

        also the 8800 cards are slowed down by a x8 pci-e slot.

        pci-e video in cards may use 1-4 lanes also you may want a pci-e based firewire bus.

    • Re: (Score:2)

      by aliquis ( 678370 )
      Uhm, so the 4 pci-express x16 slots isn't really x16 or what? Because if they are how can the chipset be low on pci-express lanes? Sounds like there are plenty?
  • I don't get it. Why have there been so many reviews of these systems lately? They have been available for many months. I ordered one in February. Anybody with a large PC case has had access to these dual-quad systems in the EATX form factor since the beginning of 2007 at least.

    • Re: (Score:2, Interesting)

      by mistermark ( 646060 )
      indeed! in my job (I work at a small OEM in the Netherlands), I already sold several dual quad-core systems the last couple of months, build to order. This week I sold a similar system (with this exact same motherboard) but with 16GB of memory instead of 4. Luckily it's not going to run Vista ;-)

      • Re:Why now? (Score:5, Funny)

        by andy314159pi ( 787550 ) on Friday June 15, 2007 @02:13PM (#19523059) Journal

        I already sold several dual quad-core systems the last couple of months, build to order. This week I sold a similar system (with this exact same motherboard) but with 16GB of memory instead of 4. Luckily it's not going to run Vista ;-)
        But it appears to be one of the few machines that supports the minimum hardware requirements of Vista.

        • Re: (Score:1)

          by RedElf ( 249078 )
          If that meets the minimum, then what do we need, a beowulf cluster of those to run the recommended system requirements for Vista?

        • Re: (Score:3, Funny)

          by sgt scrub ( 869860 )
          And Quake2 STILL stutters every three seconds.
        • Yeah, but I'll wait. I don't want to buy a PC like this and find out that it requires serious hardware upgrades just to boot Windows Vienna [wikipedia.org]

    • Intel is trying to pre-empt AMD marketing Barcelona as the one true 8 core workstation processor. AMD is going to have a somewhat more appealing offering, especially as a gaming platform what with non-registered RAM, but simply by hyping their server platform as a workstation Intel gets to get to an 8 core enthusiast platform first. And with multi-threaded applications so far behind, the performance difference between two native quad core processors communicating via hypertransport and two dual-dual core pr

  • April is not a time frame (Score:3, Insightful)

    by Flying pig ( 925874 ) on Friday June 15, 2007 @01:59PM (#19522829)
    The original report came out in April, which is the name of a month and a time period, _not_ the "April time frame". Adding verbiage does not make your submission look more impressive or indeed add any meaning whatsoever.

    Moving forwards from this present moment in time, I think we should take on board the suggestion that redundant verbiage be deep-sixed, or at least run the concept up the flagpole and see if anybody salutes.

    That off my chest, calling this thing a V8 is just as annoying as it presumably does not have two angled banks of 4 cores running off a common crankshaft.

    Yes, if you must use stupid analogies I will prod them till they break.

  • I think the writer's text editor isn't quite multi-core ready, looks like a race condition to me... or maybe he should learn to type :P

  • Naysayers R US (Score:3, Insightful)

    by andy314159pi ( 787550 ) on Friday June 15, 2007 @02:06PM (#19522935) Journal
    Figuring out how to redesign a program to run in parallel is a terribly difficult thing to do, for the most part. There are sometimes linear algebra problems that appear in science applications that lend themselves to parallel coding, but those aren't things that most users are trying to implement. *They cannot give up on making sequential processing faster.* Making a platform as massively parallel as this is (for a personal computer) will never accomplish what improving other facets of the architecture like memory latency, cache size, and of course the chip frequency. So we have been using machines with four processors for 11 years, and for the most part only one processor gets utilized even after extensive efforts to make our applications run in parallel. The overhead for farming out work is worthwhile only when you have very large chunks of computing that doesn't have to be sequentially processed. I really see this multicore processing stuff as a bit of a cop out.

    • Re: (Score:3, Insightful)

      by Jeffrey Baker ( 6191 )
      How is it a "cop out"? The current crop of Xeons have 3.0GHz clock speeds, huge caches, and excellent per-clock performance including single-cycle 128-bit packed operations. They are by any measure the finest x86 processors ever offered on the market. The fact that you get four of them per socket is just a bonus.

      Also, I can think of one general-purpose workload that is easy to parallelize: sorting. Tons of applications require fast sorting, from word processors to mail programs and web browsers all the

      • Re: (Score:3, Interesting)

        by CastrTroy ( 595695 )
        The question is, how many items do you have to sort, and does it even give a noticeable increase in speed once you distribute all the data to the seperate processors and gather it back up again. From my parallel programming course, I remembered you could do a sort in O(1) time, but that you had to have N processors, and that doesn't even count distribution and gathering time. The best parallel algothims get sorting time of O(n log(n)), which is the same a quicksort, but you can parallelize it, so on 4 pro
        • The best parallel algothims get sorting time of O(n log(n))

          Wrong. Radix and bucket sort algorithms sort in O(n). And, you can parallellize them using divide and conquer method. Thus, you can get O(n/m) + some overhead (which turns out to be O(n/m) as well) for merge where m is the number of parallel cores.

          • Re: (Score:1)

            by Krakhan ( 784021 )
            That's true, but that assumes you can actually use Radix and Bucket sort on your data. I think the OP might have been referring to the fact that if you only use comparison between two elements, than the best you can do for sorting is Omega(n log n).

    • Re: (Score:2, Interesting)

      by KillerCow ( 213458 )

      Figuring out how to redesign a program to run in parallel is a terribly difficult thing to do, for the most part.


      Not on a server. Forking on the accept call is embarrassingly parallel.
    • This is so not true. Have you ever heard of OpenMP [openmp.org]? It lets you trivially parallelize any for loops in your program (i.e. let each CPU handle a point in the loops. It is embarrassingly easy to implement. How would it benefit the user? Umm, how about searching your terabytes of images with face recognition software that will soon become available? How about all your photoshop processing, or for finance nuts, or updating your gigabytes of spreadsheet data? Those all run on for loops, you can bet.

      Sorry? You
      • Everything you've written is in Java/.NET and takes 2 seconds just to bring up a window on a modern CPU? Then maybe it's time to stop dissing lower level languages.

        Uh, what? .NET is not Java. GUI apps written in .NET run very fast, unless they're written by bad coders.

        It is also very easy to implement multithreaded apps in. By design, instead of using some hokey automated quasi-multithreading.

        • Fast and .net? (Score:2, Informative)

          by anss123 ( 985305 )
          Excepting some tech demos I've yet to see a GUI app written in .net that I'd consider "very fast". Hell, write me a .net app that can read in a 20MB BMP file faster than IrfanView can have it saved back as a jpeg and I'll applaud you. It must be possible, but I suspect one have to resort to unsafe code.

          Java have somewhat of a bad rep, but it's every bit as fast as managed .net (Windows.Forms call out into native code). It might even have more of a future in our multiprocessing tomorrow thanks to Sun's push
    • It is if you are using the wrong tools. Check here to get Tim Sweeney's (unreal engine) thoughts on how he envisions programming his next game engine:

      http://www.st.cs.uni-sb.de/edu/seminare/2005/advan ced-fp/docs/sweeny.pdf [uni-sb.de]

      (Hint: It involves switching from C++ to a Haskell like language)

    • Multi-core and multi-threaded processors are innately different to program for than compute clusters or even multi-socket SMP systems. On a Core2Duo, the delay for communicating between threads is the same as a level 1 cache miss - which isn't that big a deal at all. On an UltraSparc T1 it's the same as a level 1 cache *hit* when the two threads are running on the same core. This means it starts to be feasible to spawn "asynchronous subroutines" for even reasonably trivial operations.

      Once you start thinkin

    • Just as you complain that the chip makers aren't concerned with your problems, you don't even pretend to understand their problems. Their problem is that the number of transistors you fit on a chip is doubling every year and a half, and they have to do SOMETHING with those transistors. There is only so much you can use transistors to make a chip faster before you've long passed anything resembling a cost/benefit ratio.

      Maybe if Intel and AMD weren't so concerned with power consumption, they would release 5
    • Comment removed based on user account deletion
    • I 'Nay' too. Most of the time we have processing power to spare. Therfore I think the road will go elsewhere.
      I think we will see much more small, cheap but powerfull cpu's in everyday equipment.
      Which will be able to communicate and form spontanious clusters.
      Think RFID and Sun's SPOT.

      Intel and AMD will make 8 core cpu's and discover that its a small market.
      After that they will use the *same* techologie to make single-core processors which can be dynamically coupled together with something like bluetooth. And

    • Figuring out how to redesign a program to run in parallel is a terribly difficult thing to do, for the most part. ...

      But that is where a fundamental mistake is, you can't redesign it; you need to scrap it and start over by first understanding what it is you want to accomplish from a high level. Then break it down to it constituent parts and dependencies independent of old procedural programing methods. The second big issue is you need to find designers that can think parallel, few can. Same with progra

  • I smell a lawsuit from a juice company. :)
    • Doubt it I have one in my Ford, my wife has one in her Volvo, theres one in my old Suburban and GTO. V8 is pretty common.
  • The all new 'V8 Octa-Core'!

    However, as much as I'm drawn to this, I'm prompted to hold out for the 'V32 Duo Quintuple-Core' System - now that sounds like a real hair-on-your-back piece of technology - whoa!
  • From a recent article:

    Microsoft executive Ty Carlson spoke about the future of Windows recently during a panel discussion at the Future in Review 2007 conference held in San Diego, California. Carlson said that future versions of Windows would have to be "fundamentally different" in order to take full advantage of future CPUs that will contain many processing cores.

    "You're going to see in excess of eight, 16, 64 and beyond processors on your client computer," said Carlson, whose job title is director of tec

    • How does OSX and Linux handle eight processors?

      Just fine, thank you.
      Of course it depends on what you are using it for, though. If you are using it to compile software, run web apps, batch-encode music and/or run software written in Erlang you can easily max out more cores than that. If you use it to play games or rip one DVD at a time it won't be much faster than if you had only one or two cores.

      • If you use it to play games or rip one DVD at a time it won't be much faster than if you had only one or two cores.

        But that's just because the software isn't written to take advantage of more cores yet. Video transcoding can be parallelized pretty well, and games definitely can use all the processing power they can get.

    • So, if Windows is only designed for two or four processors, why even consider eight? Of course, that's Microsoft... How does OSX and Linux handle eight processors?

      Thanks for answering your own first question.

      Partial answer to the second question: Nobody knows how OSX runs on more than four processors, because so far you can only run it on four.

      Half-made-up answer to the rest: From what I have heard, Linux will handle up to 8 (some say 16) quite well, but tends to taper off after that.

    • So, if Windows is only designed for two or four processors, why even consider eight?

      Best not to listen to marketing dweebs for technical information. Windows NT ("Vista") is - and always has been - designed from the ground up to work very well with multiple CPUs. It's heavily multithreaded, fully re-entrant, kernel locking is very fine-grained, etc, etc.

      I have no idea what this person thinks they're saying, but Windows NT4 was available for machines with 8 CPUs a decade ago and Windows 2000 has been running on 64-CPU machines for years. It's possibly some sort of incredibly poorly communicated misunderstanding about how modern machines are more likely to find multiple cores on a single package, rather than discrete CPUs, but even that would only require scheduler tweaking and certainly nothing "fundamentally different". It may also be a reference to Singularity [wikipedia.org].

      What is clear, is that "Microsoft executive Ty Wilson" has NFI what he's talking about and needs to be whacked with a clue-by-four (and probably was). There's nothing at all wrong with Windows' SMP support, especially in the context of the hardware it typically runs on.

      Of course, that's Microsoft... How does OSX and Linux handle eight processors?

      OSX, not very well. They've only moved away from a single big kernel lock relatively recently - although Leopard is supposed to have some significant improvements in this area - and there's lots of work that needs to be done. Linux's SMP support is excellent (almost certainly better than Windows') and it's been running on machines with quite large CPU counts for years.

  • Top Ten Uses For Your New Cores (Score:5, Funny)

    by DysenteryInTheRanks ( 902824 ) on Friday June 15, 2007 @02:36PM (#19523377) Homepage
    From the Home Office in Bangalore India!

    Top Ten Uses For Your New Processor Cores:

    10. Vista (Starter, Ultimate Turbo Champion, etc). If this applies to you, stop reading list here, all your new cores are belong to Microsoft.

    9. Time to install Web 2.1, baby!!

    8. Full-screen full-motion porn on all three of your 30-inch computer monitors [time.com] while running global warming computer model in background

    7. Terrorism.

    6. Receiving chocolate cake over the Internet.

    5. As a tool to help you personally become a more productive worker, engaged citizen and attentive spouse and parent, rather than as a weird techno-fetishistic ends unto itself. Ha ha, just kidding!! LOLzzz.

    4. Dedicated core for Safari installs/updates.

    3. Department Homeland Security monitoring/spyware (federal statutory requirement)

    2. AT&T Broadband/RIAA monitoring/spyware (in EULA)

    1. Wife's monitoring/spyware (in the vows)

  • Power consumption (Score:4, Interesting)

    by A Friendly Troll ( 1017492 ) on Friday June 15, 2007 @03:26PM (#19524155)
    FTFA [hothardware.com]:

    Our testing showed the V8 ssytem consumping much more power than anything else while idling at the Windows desktop; almost 50W more than QuadFX and over 100W more than the QX6800. With the processors operating under full load, however, the tables turned somewhat.

    Yeah, the tables did turn. Under full load, the QX6800 - which is already power-hungry - uses 319W. The V8 and the QuadFX are at 474W and 498W, respectively. That's an extra 155-179W... For what?!

    Is this a continuation of the P4 Prescotts, which used 130W+, IIRC? These beasts use *even more* juice.

    Yeah, such CPUs have their place, but if this is an indication of the future of desktop computers, fu*k it. The V8 uses more power over a QX6800 (50W) while idling than what my CPU (E4300) uses at full load. Are we going to be able to buy 50W CPUs in five years, or are we going to have to deal with insane cooling solutions for 200W CPU monsters?


    • Your numbers need correcting. First, this is platform power, which includes a lot more than CPU (the memory subsystem can have a large impact too). Second, look at the power per core:

      QX6800: 319W / 4 = 79.8 W
      QuadFX: 498W / 4 = 124.5 W
      V8 : 474W / 8 = 59.5 W

      So your conclusion is reversed: this is the kind of trend we want for power: downward.

      • Your numbers need correcting. First, this is platform power, which includes a lot more than CPU (the memory subsystem can have a large impact too).

        I know. And the GPU is also very power-hungry, despite idling. It's still a shitload of power just for the CPU. The QX6800 should be using around 110W by itself, so the V8 and the QuadFX are roughly double that (slightly lower, though).

        Second, look at the power per core:
        QX6800: 319W / 4 = 79.8 W
        QuadFX: 498W / 4 = 124.5 W
        V8 : 474W / 8 = 59.5 W
        So your conclusion is

        • Don't be silly. Intel and AMD aren't going to take away options for low-power computing. Even if they did, there's still lots of other options, such as VIA's CPUs. For one thing, notebooks can't use anything too high-power because batteries can't store that much energy, and even if they did, cooling would be a serious problem (a system this hot would burn your lap). And look how popular Intel's Core series was (and still is) when it was released; people really didn't like the P4 series of inefficient pr

          • And look how popular Intel's Core series was (and still is) when it was released; people really didn't like the P4 series of inefficient processors.

            Do you remember when Pentium M came out? It ran faster than most of the desktop offerings and used less power. However, it used a different socket, so even if you were prepared to pay five times more money for it, compared to an equivalent desktop CPU, you couldn't have used it. It took like a year to get a few Socket 479 motherboards "on the market" - they were

            • I remember that quite well. There was a huge demand for that chip on the desktop, but because of the problems you mention, very few people actually bought it. But Intel finally realized that the P4 was a sinking ship, and adapted the Pentium-M architecture for the desktop, which became what we now call "Core".

              Personally, I think their former CEO Craig Barrett was to blame for the P4 debacle (and many other bad moves as well), and their current CEO Paul Otellini deserves kudos for some good decisions like

        • How about a 16-core CPU with 35W per core, which by the numbers above would indicate a trend downward? 16*35W=560W. Whoops!

          Maybe a 32-core CPU with 20W per core? 32*20=640W. Double whoops!

          Why are these all Whoops? This 32-core CPU will now replace 16 or 8 blades, which are far less efficient?

          However, my point still stands: what if in five years we CANNOT buy a 50W CPU? What if Intel and AMD will be manufacturing only eight-core CPUs at the low end, with a record-breaking 15W per core? That's still 120W, wh
    • Keep in mind that the "V8" is an eight-core system (or quad-dual core system). It looks like the AMD QuadFX is still "just" a dual dual core system.

      Also, this is a very extreme system, and like quad-SLI, is probably only for those that are willing to throw money away to no extra benefit to almost all home users. Anyone that truly needs the power should just get a workstation system instead, that way, you don't get the pimp-style marketing rubbish, and probably get a cheaper system too.

    • Re:Power consumption (Score:4, Informative)

      by Chandon Seldon ( 43083 ) on Friday June 15, 2007 @04:04PM (#19524709) Homepage

      Are we going to be able to buy 50W CPUs in five years

      Sure. You can even get a 1W CPU today if you want. There's just an energy / performance tradeoff, and the V8 goes for all-performance. "Normal" desktop processors today have design power usages of either 65 or 90 watts. Low power 45W desktop processors are available, and you can go to notebook / specialty processors below that.

    • LOL! I just substituted "Watts" for "Gasoline" in your piece and kept thinking about "V8's".
      Man did I get a big smile on my face.
  • "One motherboard does not a platform make."

    This is the type of remark that makes me want to smash the teeth in on your average power user / paid review writer.

    Ok, the Commodore came with a single motherboard design for years. Is that not a platform?
    The 2600 was not a platform either?

    See sir, when you use extra words in a vain attempt to sound witty, you end up making the whole article something I'd rather just skip than risk being bamboozled by more, equally stupid remarks, which tend to only be equalized b
  • will still not get any geeks laid.

  • May I recommend this setup. Its very stable, the Xeon class server and workstation boards are very solid (using a Supermicro X7DAL-E), and the performance for software development is simply unmatched. Everything on a good Intel 5000X based board is supported by Linux as well. Run a 64-bit distribution and pack it full of RAM (FB-DIMMs are interesting beasts - various ways to deal with ranked memory and exploit the parallel nature of the serial memory bus and trading latency for bandwidth and vice versa)

    The

    • Its designed as a very boring very well performing workstation.

      Translation: It is designed for rock-solid stability (yes, even if you run Windows) and reliability.

      If you want a tweakable board, check out Asus, Tyan, or *shudder* Abit, but the BIOS on those systems will likely be a bit more buggy, be a little less supported by Linux, and not have chassis designed specifically for each board. Sure, ATX is ATX, but Supermicro puts a lot of extras in: selectable fan profiles (set it for a server, the fans will

Slashdot Top Deals

Solutions are obvious if one only has the optical power to observe them over the horizon. -- K.A. Arsdall

Close