Slashdot Mirror


Intel Launches Sandy Bridge-E Series Processors

MojoKid writes "Today marks the release of Intel's Sandy Bridge-E processor family and its companion X79 Express chipset. The first processor to arrive is the Core i7-3960X Extreme Edition, a six-core chip manufactured using Intel's 32nm process node that features roughly 2.27 billion transistors. The initial batch of Sandy Bridge-E CPUs will feature 6 active execution cores that can each process two threads simultaneously via Intel Hyper-Threading technology. Although, the chip's die actually has eight cores on board (two inactive), due to power and yield constraints, only six are active at this time. These processors will support up to 15MB of shared L3 Intel Smart Cache and feature integrated quad-channel memory controllers with official support for DDR3 memory at speeds up to 1600MHz, as well as 40 integrated PCI Express 3.0 compatible lanes. Performance-wise, Sandy Bridge-E pretty much crushes anything on the desktop currently, including AMD's pseudo 8-core FX-8150 processor."

204 comments

  1. taking all bets on the price by Anonymous Coward · · Score: 0

    given Intel's near monopoly on the cpu market, i'm guessing over 9000 dollars. fuck you intel

  2. missing option... by nonsensical · · Score: 0

    But there's no ~$300 priced k version.

    1. Re:missing option... by KingMotley · · Score: 1

      The $300 version is called i5.

    2. Re:missing option... by beelsebob · · Score: 2

      No, the i5 is $150 to $200. The around $300 versions are the i7 2600, 2600k, 2700, 2700k and 3820, the 2600k and 2700k have unlocked multipliers, the 3820 has a partially unlocked multiplier.

  3. Pseudosematics: "psuedo 8-core" by Anonymous Coward · · Score: 0

    No mojo for you.

  4. Re: Cough by slaker · · Score: 1

    The i7-2700k will have a launch MSRP of $331.

    I fully expect I'll be able to get one at Microcenter for $280 or so.

    --
    -- I wanna decide who lives and who dies - Crow T. Robot, MST3K
  5. Crushes anything on the desktop by CubicleView · · Score: 5, Funny

    So it's very heavy then?

    1. Re:Crushes anything on the desktop by houstonbofh · · Score: 3, Funny

      They are going for the "Big Iron" market.

    2. Re:Crushes anything on the desktop by Anonymous Coward · · Score: 0

      This is extremely serious news event, and you tell jokes today?

  6. Re: Cough by nonsensical · · Score: 1

    They were supposedly releasing the i7-3820 at $294, with "limited unlock" whatever that means. I was wanting to get the newer socket 2011 motherboard to future proof as much as possible.

  7. Socket by TheCRAIGGERS · · Score: 0

    Holy shit, did they actually manage to unveil a new CPU without forcing a new socket down our throats? I don't believe it.

    1. Re:Socket by Wagoo · · Score: 2

      Er, no.. they didn't. It's LGA 2011, a new socket. Again.

    2. Re:Socket by Zephiris · · Score: 1

      Herp. Derp. "After some delays, Sandy Bridge E is finally here. The platform is actually pretty simple to talk about. There's a new socket: LGA-2011, a new chipset Intel's X79 and of course the Sandy Bridge E CPU itself. We'll start at the CPU."

      --

      "A Goddess rarely smiles for she is forced by others to be an island unto herself." - Zephiris
    3. Re:Socket by TheCRAIGGERS · · Score: 1

      I didn't see that in the article, but I am sadly not surprised.

    4. Re:Socket by beelsebob · · Score: 5, Informative

      In this instance, no, this is a new socket, replacing LGA1366. Next year though intel will release new desktop CPUs based on the current LGA1155 for desktops.

      Intel actually don't release things on new sockets as much as people think. Every tick/tock pairing has one desktop socket, and one server socket, this is the server socket to go with LGA1155's desktop socket. The tock to come (Ivy Bridge) will also use LGA1155 for desktops and LGA2011 for servers.

      This is much the same as has happened before: Nehalem introduced LGA1156 and LGA1366, westmere reused them; Conroe introduced (properly) LGA775 and LGA771, Arandale reused them.

    5. Re:Socket by Anonymous Coward · · Score: 0

      It is on a new socket, wake up.

    6. Re:Socket by Kjella · · Score: 3, Insightful

      Intel actually don't release things on new sockets as much as people think. Every tick/tock pairing has one desktop socket, (..) Nehalem introduced LGA1156 and LGA1366, westmere reused them; Conroe introduced (properly) LGA775 and LGA771, Arandale reused them.

      Which means the motherboard is good for exactly one generation on upgrades, on the same die size. Who'd seriously upgrade their almost-new Nehalem system to Westmere, Conroe to Arandale or Sandy Bridge to Ivy Bridge? Then you'd do better putting all that money into one processor. Socket compatibility is only good if it lasts long enough there good reason to upgrade. With Intel, I assume that any new processor I buy will require a new motherboard, simple as that.

      --
      Live today, because you never know what tomorrow brings
    7. Re:Socket by beelsebob · · Score: 2

      No –it means that the motherboard is good for exactly one generation on upgrades at a smaller die size.

      It's also fairly common for people to buy a low end version of the tick, and then a high end version of the tock.

      Socket compatibility is only good if it lasts long enough there good reason to upgrade.

      Agreed, unfortunately, that's almost never, no matter who your CPU manufacturer is.

    8. Re:Socket by level_headed_midwest · · Score: 1

      You are a little off. 1. Nehalem introduced LGA1567 alongside LGA1156 and LGA1366. Westmere was never released for LGA1156, Clarkdale was but Clarkdale used a significantly different die than Westmere (two vs. six cores, no on-CPU-die memory controller). They pretty much only shared the 32 nm node and general microarchitecture. 2. Arrandale used Socket G1, not LGA115x. 3. Prescott was the first chip on LGA775, followed by Smithfield, followed by Cedar Mill/Presler. Conroe was the third or fifth generation of LGA775 chip depending how you look at it. 4. Dempsey (Xeon equivalent of Presler) was the first generation of LGA771 chip. Woodcrest (Conroe) came later. You could argue that Dempsey stank and that Woodcrest was the first chip on LGA771 worth buying, and you'd probably be right. LGA771 really didn't catch fire (pun intended) until Clovertown quad-core MCMs because AMD was notoriously late in bringing the Opteron quads to market.

      --
      Just "gittin-r-done," day after day.
    9. Re:Socket by unixisc · · Score: 3, Informative

      It ain't (necessarily) an Intel conspiracy to force you to fork out more cash for a new socket when an old one might have worked, since they'd simply be leaving the field open to AMD if that were the only reason (AMD's greatest selling point is the ability to leverage on previous generations). The reasons are technical - when a die undergoes die shrinks, there is also less area for the same number of signals, making it pad limited. Also, die shrinks need more power and ground pins even as the real estate available for such pins are reduced. As a result, the newer processors sometimes may have to undergo an I/O interface overhaul, in order to accommodate the same essential signals into fewer pins, and also drop any pins that can be eliminated.

      That is what one is seeing above - when Intel went from 1156 and 1366 to 775 and 771, reason for it was undoubtedly a reduction in the die size, and hence the number of pads that could be connected to the package. It doesn't make sense to retain the same old package, b'cos the extra 400-600 odd pins that are not being used simply add to package cost, which then percolates downstream. As a result of these changes, Intel is able to offer a complete system cost down compared to previous generations, albeit w/ different form factors. Of course, the price for that is that you can't insert a Conroe into a Nehelam socket.

      One may then wonder - how come the i7 comes in a larger package w/ 2011 bumps? Chances are that using 6 cores this time has increased the die size, eliminating the pad limitations, while the extra Vdd and Vss lines are needed, or that it is 6 different processor dies on a single multi-chip package. Either way, the larger package is required.

    10. Re:Socket by billcopc · · Score: 2

      1156 and 1366 to 775 and 771

      You got those backwards. 775 and 771 were the old sockets, spanning from late P4 all the way up to Core 2. Socket 1156 was for first-gen i3/i5, ans Socket 1366 was Nehalem (i7-9xx and extreme). They have never reduced the pin count, because each successive platform has introduced wider memory controllers and/or QPI/DMI/PCIe lanes. Sandy Bridge-E is no different, moving up to 4-channel DDR3, up from X58's 3 channels. That right there is an extra couple-hundred pins just for the memory.

      --
      -Billco, Fnarg.com
    11. Re:Socket by Grishnakh · · Score: 1

      This is Slashdot. If it isn't in the article summary, you can't expect anyone to know it.

    12. Re:Socket by unixisc · · Score: 1

      Thanks for correcting me. I wasn't sure about which preceded which, but I can see why they wouldn't reduce the pin count if they can keep the die size more or less similar by adding cores if the shrink happens to be major. In addition to the higher #channels, there's also the probability that they require more power and ground lines to allow for the increased frequencies, and sometimes, some chips may even require multiple independent power supplies (not saying that that's the case here, though, since I don't know Intel's design rules, and am just guessing based on Industry practices)

    13. Re:Socket by beelsebob · · Score: 1

      The i3, i5 and i7 on LGA1156 were nehalem too. They were the same microarch as i7 9xx, it, just like these chips, was just the server/workstation variant.

    14. Re:Socket by DragonWriter · · Score: 1

      Socket compatibility is only good if it lasts long enough there good reason to upgrade.

      Socket compatibility is good for system builders independent of whether it is ever useful to upgrade, since it increases the marketable lifespan of motherboards, increasing the time in which manufacturers can recoup the fixed cost of a particular motherboard model, decreasing motherboard prices.

    15. Re:Socket by Kjella · · Score: 1

      They have never reduced the pin count, because each successive platform has introduced wider memory controllers and/or QPI/DMI/PCIe lanes.

      Nitpick: Strictly speaking they went from LGA1156 to LGA1155, lol. And from leaked slides it's rumored that Haswell (2013) will get that down to LGA1150, because they're moving some of the voltage regulators on-die. But yes, the "big things" like memory channels means the number's been generally going up.

      --
      Live today, because you never know what tomorrow brings
    16. Re:Socket by smash · · Score: 1

      saves dealing with idiots who try and put it in a machine with a chipset that can't handle it.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
  8. $1000 processor vs the world by Anonymous Coward · · Score: 2, Informative

    Maybe I'm reading it wrong, but they are bench-marking a $1000 processor against a $300 processor?

    $1000 processor wins!

    1. Re:$1000 processor vs the world by hansamurai · · Score: 4, Insightful

      Because it actually provides context on how much better (or not how much better) that $1000 processor is. Plus, how many other desktop $1000 processors are out there to benchmark against? Certainly nothing from AMD.

    2. Re:$1000 processor vs the world by petermgreen · · Score: 1

      they are bench-marking a $1000 processor against a $300 processor?

      They are benchmarking the topen end desktop part of the current generation* against both thetop end desktop parts of the previous generation (990x), the upper-mainstream of the current generation (2700K) and the best chip the competitor can come up with (FX-8150). What else are they supposed to compare it with?

      It is a bit dissapointing is that they don't have the 3930K in the test, it should be only slightly slower thant the 3960x while being a lot cheaper. From what I can gather this is because intel didn't bother sending out samples of that chip to the reviewers.

      The fact it wins is as you say hardly surprising, what is more important is how much it wins by and in what tests. As expected it beats the 2700K by a wide margin in highly multithreaded tasks but has little advantage in thread count limited tasks. The 990x OTOH gets beaten by the 3960X in pretty much every test.

      * With this generation intel released the mainstream long before the top-end, this is the opposite of what they did with previous generations.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    3. Re:$1000 processor vs the world by Sloppy · · Score: 1

      $1000 processor wins!

      Not if you can build two computers out of $300 processors. ;-)

      --
      As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
    4. Re:$1000 processor vs the world by Shinobi · · Score: 1

      They sent it out to some reviewers. Sweclockers has it for example. Google translate it or something if you don't know swedish http://www.sweclockers.com/recension/14699-intel-core-i7-3930k-och-3960x-sandy-bridge-e

      I'm disappointed with its power consumption, same as with the Bulldozer. However, unlike with the Bulldozer, you actually get computational performance to match the power consumption.

    5. Re:$1000 processor vs the world by dbIII · · Score: 1

      Look at the article above this one about the 16 core AMD processors. If people are going to compare $1000 processors against $300 ones why not take a wander into server space? The goal posts have already been unfairly moved so why not give them a bit more of a nudge into Xeon and Opteron space?

    6. Re:$1000 processor vs the world by DurendalMac · · Score: 1

      Um...maybe because these aren't Xeons? They're high-end enthusiast chips. Xeons based on 2011 are due early next year.

    7. Re:$1000 processor vs the world by KingMotley · · Score: 1

      It's not a benchmark to see which is better, it's a benchmark to see how much better it is.

    8. Re:$1000 processor vs the world by Anonymous Coward · · Score: 0

      ... my original point being if they are only going to benchmark "extreme editions" ahead of the mainstream product release, they certainly will bias the bench-mark "articles" in their favour by the time the mainstream product is on the market. It's shady, but... it's Intel.

      (Mind you, no where does it say hint at my original point, but I was leaving that up to you.)

    9. Re:$1000 processor vs the world by billcopc · · Score: 1

      Given that the NDA was lifted just a few hours ago, I'm sure we'll see a proper 3930K review soon enough. Intel probably wants to give the Veblen types time to rush out and buy the X, before the K reviews start calling them suckers for paying $500 more for 100mhz.

      Myself, I'm waiting for the 8-core Xeon SKUs, so I can go dual-socket 16-core. You know, so I can do "make -j16" while multiboxing Crysis or something...

      --
      -Billco, Fnarg.com
    10. Re:$1000 processor vs the world by jon3k · · Score: 1

      The question isn't if it "wins", the question is how much better does it perform? Enough to justify the cost difference? Sometimes the answer is no.

    11. Re:$1000 processor vs the world by Anonymous Coward · · Score: 0

      Maybe I'm reading it wrong, but they are bench-marking a $1000 processor against a $300 processor? $1000 processor wins!

      I think -- more to the point -- that they're benchmarking Intel's flagship processor against AMD's flagship processor.

      This is a reasonable test to see how the two competitors are stacking up against each other in the enthusiast / high-performance market.

    12. Re:$1000 processor vs the world by Synerg1y · · Score: 1

      and lag equally on both in applications yet to be designed for this high of performance? :P

    13. Re:$1000 processor vs the world by Anonymous Coward · · Score: 0

      Myself, I'm waiting for the 8-core Xeon SKUs, so I can go dual-socket 16-core. You know, so I can do "make -j16" while multiboxing Crysis or something...

      "make -j32". 'Cause each core has 2 threads.

      But. There are already 10-core Xeon SKUs (Westmere EX). You could go grab yourself a 4-socket SGI UV 10 and do "make -j80":

      http://www.sgi.com/products/servers/uv/configs.html

      It might be a challenge to come up with a make job that could fully utilize a fully loaded out UV 1000 with 2560 cores (which for some reason is limited to 4096 threads instead of 5120, dunno why).

  9. A bit underwhelming by Kjella · · Score: 4, Informative

    Honestly, with a $500+ entry tag plus cooler which is not included plus expensive, low volume motherboard you might want to compare to a dual processor Xeon machine rather than other desktops for some alleged server/workstation stability too. Performance was as expected, 6 cores to 4 so it's faster in well-threaded workstation applications, not that different otherwise.

    What's disappointing is the platform, no USB 3.0, two SATA 6 Gbps ports, no SAS support, it seems like PCI express 3.0 made it in but no cards support it yet so there's nothing besides the processor that really screams high end. Well that and 8 memory slots if you feel 4x4GB isn't enough but there's alternatives like the old high end it replaces with 6 slots or 8 GB sticks that have been showing up lately - pricey but you can get 4x8GB for less than one of these CPUs. Don't get me wrong, it's the undisputed performance king but it's like the same car with a souped up engine and fuel system yet none of the features that say this is a $100k Ferrari.

    --
    Live today, because you never know what tomorrow brings
    1. Re:A bit underwhelming by slaker · · Score: 1

      This is being positioned as a hobbyist platform, same as LGA1366. The affordable E-series (i7-type) Xeons don't boot on consumer-class motherboards and don't have chipset support SMP though. These guys are the only game in town for people who want to stick three video cards in something and get a top notch CPU to go with it.

      --
      -- I wanna decide who lives and who dies - Crow T. Robot, MST3K
    2. Re:A bit underwhelming by Anonymous Coward · · Score: 0

      I don't care about USB 3.0 on a workstation, thunderbolt I do care about.

    3. Re:A bit underwhelming by unity100 · · Score: 1

      no hobbyist will spend $1000 for a cpu+cooler for these performances. pointless. leave aside lack of a lot of major stuff like usb 3 et al.

      and, not 3, but 4 video cards in crossfire or sli will not require this kind of computing power. even if you shove in 2 x 6990s in crossfire, which make 4 top-rate gpus put into 2 cards. apparently you dont know this enthusiast field, so dont bullshit about it.

    4. Re:A bit underwhelming by petermgreen · · Score: 1

      you might want to compare to a dual processor Xeon machine

      Well you can get 8-12 cores with one of those BUT they are previous-generation cores (the dual-socket variant of LGA2011 isn't out yet) and you have to pay through the nose to get a decent clockspeed. The only people I know of who have purchased dual xeon workstations have done so for the ram support.

      What's disappointing is the platform, no USB 3.0, two SATA 6 Gbps ports, no SAS support, it seems like PCI express 3.0 made it in but no cards support it yet

      But you have far more lanes. Afaict LGA2011 has 40 lanes from the processor. So even if PCIe3 doesn't pan out you can have two graphics cards running at 2.0 x16 and still have room for a nice LSI sas controller straight off the processor.

      so there's nothing besides the processor that really screams high end.

      The thing is with modern highly integrated platforms the processor is what defines the key features of the platform.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    5. Re:A bit underwhelming by beelsebob · · Score: 1

      no hobbyist will spend $1000 for a cpu+cooler for these performances. pointless. leave aside lack of a lot of major stuff like usb 3 et al.

      No, but someone who calls themselves a hobbyist, but is actually a moron will. These parts are intel going "look, we made the E5 Xeons, by the way, if you're a moron and want to hand us cash, please buy the desktop variants".

      The E5 Xeons released with these parts are really where the news is.

    6. Re:A bit underwhelming by Shinobi · · Score: 2

      Oh, you're astroturfing for AMD's marketing department again.

      Many hobbyists will easily spend $1k on this, it's a fact. Hell, averaged over a lifetime, computing is still a cheap hobby, compared to things like flying, amateur motor racing etc.

      As for the double cards, you may very well need the CPU to generate the data that you want to display over all those cards.

      My brother for example would love this CPU for his CAD work for his hobby(he designs and builds stuff as a hobby), and all the CUDA/OpenCL modules have had serious deficiencies.

    7. Re:A bit underwhelming by unity100 · · Score: 1

      head to overclock net and see if hobbyists will spend anything on this. you wont find anyone who breaks overclock records or does custom water cooling spending $1000 on this. only fanboys with brand loyalty. that is normal.

      your brother is better off with a dual socket solution and amd opterons if he is doing anything that serious. which could come even cheaper than this intel setup and provide multiples of performance. if he isnt doing that already, then he doesnt know shit, and your argument is null.

    8. Re:A bit underwhelming by DurendalMac · · Score: 1

      And you haven't looked at any benches of SLI/XFire on SB-E. It's surprising how much better it does over other platforms with 2-3 video cards.

    9. Re:A bit underwhelming by Shinobi · · Score: 2

      So that's AMD's marketing approach these days "Anyone who doesn't buy our systems is worthless, a fanboy and we'll deride them, and hope they buy our systems when we've insulted them enough"? And in regards to overclockers, do you really think they are the only hobbyists? Seriously? And even then, many overclockers who know what they are doing will buy them and make use of them.

      For his work, the 3930 will beat dual Opterons, because the computational tasks are not that easily parallellized, so you need strong per-core performance. Which these have... And AMD's current and last 2 generations haven't had.

      Personally, I moved away from Opterons in my workstation, because they simply couldn't keep up with Xeons. When my current workstation gets a bit too slow for what I'm doing, I'll carefully look at both options available, but in the current state, going with AMD would be unjustifiable and thus very unprofessional.

    10. Re:A bit underwhelming by mehrotra.akash · · Score: 1

      What about 4 6990's in quad crossfire
      I have no clue if thats even possible, but what would be the performance limitations of such a system?

    11. Re:A bit underwhelming by billcopc · · Score: 1

      You're right. No one in their right mind will buy this $1000 model when the $500 one is only 100mhz slower. Last year, when the 980X was launched, it was the only 6-core consumer CPU available, which made it worth $1000 to some people (*blink blink*). Had the 3960X been an 8-core bin, it might have made some waves, but at 6 cores it doesn't really stand out from its older siblings.

      As a result, those of us who want/need heavy parallelism have already switched to affordable dual Xeons like the E5645. Those who want the best single/dual thread performance simply overclock the 2500K/2600K. So right now, the 3960X is a chip without a real niche. It almost feels like a throwaway platofrm, much like the old Skulltrail.

      --
      -Billco, Fnarg.com
    12. Re:A bit underwhelming by Unequivocal · · Score: 1

      Interesting point and it makes me wonder if someone is going to figure out how to unlock the mysterious last two cores which are disabled? There must have been some reason why they didn't turn them on - maybe no MB support in the time they had to release?

    13. Re:A bit underwhelming by tompaulco · · Score: 1

      Honestly, with a $500+ entry tag plus cooler which is not included
      Kudos to Intel. Please don't include a stock cooler on your high end chips. Nobody buying a high end chip wants to use the stock cooler, and neither do they want to pay for a cooler that they are not going to use.

      --
      If you are not allowed to question your government then the government has answered your question.
    14. Re:A bit underwhelming by DragonWriter · · Score: 1

      no hobbyist will spend $1000 for a cpu+cooler for these performances.

      I've always kind of figurred that Intel's "Extreme" processors were targetted at "hobbyists" with more money than sense, and mostly served as a way to get that demographic to subsidize dev costs that are shared with the more mundane chips in the same generation.

    15. Re:A bit underwhelming by billcopc · · Score: 1

      It's probably a case where there is room on the die, but they haven't nailed the yields yet.

      --
      -Billco, Fnarg.com
    16. Re:A bit underwhelming by UnknownSoldier · · Score: 1

      > What about 4 6990's in quad crossfire I have no clue if thats even possible,

      Here's your answer ... just shy of 100 fps in BF:BC2 256x1600 4xAA

      $1499 each
      http://www.techpowerup.com/reviews/ASUS/Mars_II/

    17. Re:A bit underwhelming by rannmann · · Score: 1

      ...Well that and 8 memory slots if you feel 4x4GB isn't enough but there's alternatives like the old high end it replaces with 6 slots or 8 GB sticks that have been showing up lately - pricey but you can get 4x8GB for less than one of these CPUs

      This thing has 8 memory slots? It's about time, honestly. Since the 9xx series, the supported motherboards only have four RAM slots, which was a deal-breaker for me. I have 6x4gb sticks in my core i7 950 box (which turned out to be a little overkill for that processor), but the i7 3600k didn't support the amount of RAM I needed for the price. My choices were literally either: not enough CPU, or not enough RAM.

      Did it mention anywhere how the memory slots will be setup? Will be it dual-channel, or a fancy new quad-channel?

    18. Re:A bit underwhelming by Anonymous Coward · · Score: 0

      No, AMD's stuff is shit. It has horrible single thread performance, and it chews power like nothing else.

    19. Re:A bit underwhelming by smash · · Score: 1

      What exactly are you missing with the lack of USB 3? The LAST thing i want is a brain damaged high speed bus that needs my CPU to supervise it. USB is for cheap shitty peripherals. Trying to apply it to multi-gigabit connections is just stupid.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    20. Re:A bit underwhelming by smash · · Score: 1

      Nah, they'll just spend a grand on water cooling, power supplies, fans and electricity instead.

      Back in the real world, if someone needs the cpu power today, they will spend the money. If they NEED the cpu, they will not be going down the over clocking path, because if it is in any way used for anything serious all guarantees of correct cpu operation go out the window.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    21. Re:A bit underwhelming by Anonymous Coward · · Score: 0

      No, it has 8 cores because it has the same die as the upcoming Sandy Bridge-EP Xeons. Those will be available with 4, 6, or 8 cores enabled, but SB-E (the non-Xeon) will never go above 6 enabled.

      The reason why: 130W TDP and competition with Intel's own product line. i7-2600k has a 95W TDP for 4 cores and a graphics engine, figure maybe 15W for the graphics, so 80W for processors, or 20W per core. SB-E has no graphics, but 130W/8 = 16.25W per core, and I'm not even properly penalizing it for its other non-CPU power overheads either (double the memory channels of regular SB, QPI, 2.5x as many PCIe lanes, etc). So, 8-core clock speeds have to be significantly less than regular client SB, just due to power budget (and they will be on the 8-core Xeons).

      If Intel made an 8-core $1K flagship gamer chip this way, a ~$300 2600K with a much cheaper platform would easily beat it in almost all non-GPU-limited game benchmarks. Intel doesn't want to go there, so 6-core it is. You get about the same performance in most games, and more performance for parallel computing.

      This is a rebadging of a workstation/server product, which is what Intel's high end consumer i7s have always been.

      Speaking of which, you called Skulltrail a throwaway platform... but it really wasn't. It was just a 2-socket Xeon workstation motherboard that Intel tried to sell as an ultra high end gaming platform. All the engineering was done for another market, someone thought they could sell gamers on moar cores, they failed because it doesn't help frame rates at all, the end. Which is probably why this one feels the same to you, it's just that this time around they only needed to throw a single socket's worth of moar cores at gamers to make a system of questionable value to lots of its target market. Although people who want to run crazy multi-GPU setups will probably still like it for its 40 lanes of PCIe, since that's about the only thing regular Sandy Bridge is "bad" at (for values of bad which include "perfectly fine for dual GPU because 8 lanes per GPU is pretty much enough").

    22. Re:A bit underwhelming by unity100 · · Score: 1

      yes yes. thats why 1 amd supercomputer with bulldozers were ordered EVERY WEEK in the past 3 weeks. shut the fuck up if you dont know shit.

    23. Re:A bit underwhelming by unity100 · · Score: 1

      read overclock net. you dont know shit, yet you are still talking.

  10. Wait for Ivy Bridge. by wildstoo · · Score: 5, Informative

    That's nice and everything, but I'll wait for Ivy Bridge, which is due March 2012.

    According to Wikipedia:

    Ivy Bridge feature improvements from Sandy Bridge were expected to include:

    Tri-gate transistor technology (up to 50% less power consumption)
    PCI Express 3.0 support
    Max CPU multiplier of 63 (57 for Sandy Bridge)
    RAM support up to 2800MT/s in 200MHz increments
    Next Generation Intel HD Graphics with DirectX 11, OpenGL 3.1, and OpenCL 1.1 support
    The built-in GPU is believed to have up to 16 execution units (EUs), compared to Sandy Bridge's maximum of 12.
    The new random number generator and the RdRand instruction, which is codenamed Bull Mountain.
    Next Generation Intel Quick Sync Video
    DDR3 low voltage for mobile processors
    Multiple 4k video playback

    So yeah, just hang on for the die shrink if you care about performance and power consumption. My next system will definitely be Ivy Bridge based.

    1. Re:Wait for Ivy Bridge. by Anonymous Coward · · Score: 0

      That's nice and everything, but I'll wait for whatever's coming out after Ivy Bridge.

    2. Re:Wait for Ivy Bridge. by PIBM · · Score: 1

      Pffft.. you could have waited for whatever would come out last! Thats the only way you could not be overdone by those waiting for what's coming after whenever you decided to make your move :)

    3. Re:Wait for Ivy Bridge. by L4t3r4lu5 · · Score: 1

      I'm going to wait for Ivy Bridge too.

      I'm then going to buy an i5 2500K chip and Z68 motherboard for pennies, load it up with yesterday's memory, and have a system which will last me another 5 years. It's worked well with my existing nForce 680i / Core2Quad Q6600 setup.

      --
      Finally had enough. Come see us over at https://soylentnews.org/
    4. Re:Wait for Ivy Bridge. by L4t3r4lu5 · · Score: 4, Insightful

      16 EUs. Woah. For comparison: nVidia and AMD's current cards have over 1500 of them.

      So this is like comparing a consumer-class lawnmower to a Bugatti Veyron, graphics-wise.

      Why would anyone with such a extreme setup ever care for such shitty integrated graphics? And whatÃ(TM)s the point of DirectX 11 support, if you can't use it anyway.

      Because you can power off the graphics card and still do 1080p HD video decoding, or play some lower-end games. That'll shave a good 300w off of your total power usage. Not to be scoffed at.

      --
      Finally had enough. Come see us over at https://soylentnews.org/
    5. Re:Wait for Ivy Bridge. by ZenDragon · · Score: 1

      I think the biggest reason is possibly that the integrated graphics can function as your primary adapter when the the power of the dedicated cards is unnecessary, reducing power consumption and heat overall. Certainly its not going to win any graphic wars, but its more than enough to run Aero efficiently. I suppose power consumption is not on the top of the list of most peoples concerns when buying something like this, but its nice to have the option to do so.

    6. Re:Wait for Ivy Bridge. by Anonymous Coward · · Score: 0

      Ivy Bridge? Haha loser!

      You should wait for Haswell which is due 2013 and runs off the 22nm production line.

      Personally I'm waiting for Rockwell which will be 16nm and will totally destroy your Ivy Bridge piece of shit.

    7. Re:Wait for Ivy Bridge. by Kjella · · Score: 1

      16 EUs. Woah. For comparison: nVidia and AMD's current cards have over 1500 of them.

      An EU ~= 2 shaders so 32 to 1500, but yes...

      Why would anyone with such a extreme setup ever care for such shitty integrated graphics?

      Ivy Bridge isn't what Intel considers "extreme", it's their mainstream processor. As such it'll go into plenty corporate desktops and other places that want CPU power but not to play games or for casual games. Less and less people get a discrete graphics card, because the Intels don't suck quite as bad as they used to, their market share is now about 60%. But yes, for this discussion it wasn't very good selling points. OTOH, if you game the Sandy Bridge-E doesn't actually deliver any more than a 2600K, put that money in graphics cards instead...

      --
      Live today, because you never know what tomorrow brings
    8. Re:Wait for Ivy Bridge. by Sloppy · · Score: 1

      Careful, this kind of thinking may keep you from ever getting any upgrade at all. There's always something niftier coming in half a year.

      --
      As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
    9. Re:Wait for Ivy Bridge. by Anonymous Coward · · Score: 0

      Because you can power off the graphics card and still do 1080p HD video decoding, or play some lower-end games. That'll shave a good 300w off of your total power usage. Not to be scoffed at.

      But how many people purchasing this particular CPU would ever actually do that?

      The people buying these hobbyist CPU's are not interested in shaving power. Integrated graphics is all well and good for the mobile market, but I bet you'll have a hard time finding anyone running integrated graphics in a home built machine, certainly not one running a top-tier CPU in it.

      It's like selling a car with an optional pedal-drive. Why?

    10. Re:Wait for Ivy Bridge. by beelsebob · · Score: 1

      Actually, nVidia's currently have 512 of them per chip, and AMD currently have around 400. AMD like to claim they have 1600 by counting the number of scalar units instead of the number of vector units.

      Intel's current (Sandy Bridge) IGP is pretty damn good –it holds its own against integrated offerings by AMD and nVidia. Ivy Bridge is expected to put them right in the mix.

    11. Re:Wait for Ivy Bridge. by ksd1337 · · Score: 1

      I for one welcome our Terminator overlords!

    12. Re:Wait for Ivy Bridge. by NitroWolf · · Score: 2

      That's nice and everything, but I'll wait for Ivy Bridge, which is due March 2012.

      So yeah, just hang on for the die shrink if you care about performance and power consumption. My next system will definitely be Ivy Bridge based.

      That's nice and everything, but I would just wait for Haswell, which is due in 2013.

      So yeah, just hang on for the die shrink from Sandy Bridge if you care about performance and power consumption. My next system will definitely be Haswell based.

      That's nice and everything, but I would just wait for Haswell, which is due in 2013.

      So yeah, just hang on for the die shrink from Sandy Bridge if you care about performance and power consumption. My next system will definitely be Haswell based.

      That's nice and everything, but I would just wait for Broadwell, which is due in 2014.

      So yeah, just hang on for the die shrink if you care about performance and power consumption. My next system will definitely be Broadwell based.

    13. Re:Wait for Ivy Bridge. by bfields · · Score: 2

      "I bet you'll have a hard time finding anyone running integrated graphics in a home built machine, certainly not one running a top-tier CPU in it."

      As a kernel developer I have scripts running all day that compile new kernels, boot VM's to them, and run some tests. The faster the better. I wouldn't say no to lower utility bills either even if it's not the first thing I look at.

      But in any case, for graphics, the integrated stuff is fine. And Intel's involvement in Linux graphics development has meant their stuff has been among the best supported for a while. (Maybe that's changed lately, I haven't kept up). I usually build my own machines (partly exactly because I seem to recall it being hard to find pre-built desktops with higher performance that could get usable graphics without the need for proprietary drivers. Which may not be a factor for others but I want to be able to debug problems on the latest upstream kernel without worrying about what some binary driver might be doing.)

      So there's one small counterexample, to your first statement if not the second. (I usually get a better CPU but nothing "extreme".)

    14. Re:Wait for Ivy Bridge. by NitroWolf · · Score: 1

      Integrated graphics is all well and good for the mobile market, but I bet you'll have a hard time finding anyone running integrated graphics in a home built machine

      I have a number of machines running integrated graphics for numerous reasons. Least of which is the kids machines - you sure as hell don't need a discrete graphics card to play stupid little flash games and do homework.

      Machines that are for checking email and surfing the web... don't really need discrete graphics either.

      Plenty of reasons to use integrated graphics.

    15. Re:Wait for Ivy Bridge. by KingMotley · · Score: 1

      If I am paying $1500-$2000 for an enthusiast computer, do you think I care whether my electric bill is $88 or $89 this month?

    16. Re:Wait for Ivy Bridge. by petermgreen · · Score: 1

      their market share is now about 60%

      Intel now integrates a GPU in all their current gen mainstream processors. So many many people will end up owning an intel GPU without ever using it.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    17. Re:Wait for Ivy Bridge. by m50d · · Score: 1

      Intel's involvement in Linux graphics development has meant their stuff has been among the best supported for a while. (Maybe that's changed lately, I haven't kept up).

      I've heard this claim repeatedly, but I've used three different intel graphics chipsets and three different nvidia ones under linux, and all the nvidia ones worked better (no random corruption, working S3 suspend) than all the intel ones.

      --
      I am trolling
    18. Re:Wait for Ivy Bridge. by thue · · Score: 1

      Cute :).

      But IMO the jump from Sandy Bridge to Ivy Bridge is especially large, particularly in power consumption, so it makes sense to make Ivy Bridge the CPU to wait for.

    19. Re:Wait for Ivy Bridge. by bfields · · Score: 1

      I've heard this claim repeatedly, but I've used three different intel graphics chipsets and three different nvidia ones under linux, and all the nvidia ones worked better (no random corruption, working S3 suspend) than all the intel ones.

      Is that using open-source drivers or nvidia's proprietary drivers? (Again, for my case I'm only interested in the free drivers.)

    20. Re:Wait for Ivy Bridge. by JoeMerchant · · Score: 1

      Because you can power off the graphics card and still do 1080p HD video decoding, or play some lower-end games. That'll shave a good 300w off of your total power usage. Not to be scoffed at.

      Damn, if you just buy a $300 nettop, the whole system only consumes 30w while doing 1080p HD video decoding.

      Last time I had a graphics card that consumed 300W was... well, never. Live 2 years behind the curve and you can save a lot of CO2 emissions.

    21. Re:Wait for Ivy Bridge. by m50d · · Score: 1

      Nvidia's proprietary drivers. (I'd prefer free ones, but pragmatically they get suspend right, their dual-monitor support works better, the opengl performance is great and vdpau is fantastic)

      --
      I am trolling
    22. Re:Wait for Ivy Bridge. by Shinobi · · Score: 2

      Actually, right now, 2 years behind the curve uses more electricity than being on the edge

    23. Re:Wait for Ivy Bridge. by edxwelch · · Score: 1

      > Because you can power off the graphics card and still do 1080p HD video decoding, or play some lower-end games.

      I believe that only works with the right CPU / GPU combination and the right OS support

    24. Re:Wait for Ivy Bridge. by Khith · · Score: 1

      Of course something nifty is coming out! I hear that the upcoming Osborne system is going to be pretty good! I've already cancelled my order for their existing computer...

    25. Re:Wait for Ivy Bridge. by AmiMoJo · · Score: 1

      It wouldn't be so bad if Intel stuck to one socket. With AMD at least if you buy a current AM3+ board you are likely to have an easy CPU upgrade path for years, instead of having to throw out the mobo and RAM when all you really want is a better processor.

      --
      const int one = 65536; (Silvermoon, Texture.cs)
      SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
  11. Re: Cough by Anonymous Coward · · Score: 2, Informative

    Trying to future proof an Intel based motherboard is pointless. Considering this new socket replaces the LGA 1155, which is less than a year old, in an ever decreasing release interval, I would estimate this socket will be obsolete by next spring, summer at the latest.

  12. Re: Cough by slaker · · Score: 1

    As someone with a decent investment in LGA1366 stuff, I'd rather play it smart and keep everything on the "mainstream" LGA1155 for anything but the six core CPUs. The motherboards are harder to find and substantially more expensive for the dubious value of having some extra PCIe lanes and a couple extra DIMM slots.

    I'm in the process now of selling off my LGA1366 machines while they still have value and replacing them with Xeon E-series equipment.

    --
    -- I wanna decide who lives and who dies - Crow T. Robot, MST3K
  13. Re: Cough by beelsebob · · Score: 4, Informative

    It doesn't replace LGA1155 it replaces LGA1366, which is 3 years old. These chips are server and workstation level chips.

    Intel's next desktop architecture is called Ivy Bridge, will be released in the first 3 months of next year, and will be using LGA1155. Ivy Bridge E will use LGA 2011. Only in about 20 months time (by which point LGA1155 will be 3 years old) will Haswell come out on a newer socket.

  14. more sources by neersign · · Score: 1
    In case you want more than just hothardware, here's a decent selection
  15. 6 or 8 cores by spaceman375 · · Score: 5, Funny
    The chip has eight cores, they all work, but you can only use six. The other two are reserved for the DHS and cronies. I, for one, do NOT welcome our dual-core overlords.

    (I've always wanted to start a conspiracy theory.)

    --
    On the one hand you take life too seriously, and on the other, you do not take playful existence seriously enough. Seth
    1. Re:6 or 8 cores by L4t3r4lu5 · · Score: 1

      They're for processing the most efficient flight lanes to use when dusting us with their "contrail" mind control chemicals.

      No, the pollen filter in your car won't protect you; They pre-seeded that.

      --
      Finally had enough. Come see us over at https://soylentnews.org/
    2. Re:6 or 8 cores by Anonymous Coward · · Score: 0

      the problem with mind control drugs is that they build a resistance...
      the mind is only so fast anyways put too much mind control drugs and it locks up hard, emulating a system crash.
      i should know, i am paranoid schizophrenia and without the drugs it's rather a lot like someone replacing my mind with rainbows, which doesn't do me much good, the mind control drugs fix that because having rainbows on the brain is far worse than the mind control.

      perhaps it was because my nicknames on the internet are all related to deletion and i've dared use darik's boot and nuke.

    3. Re:6 or 8 cores by Narnie · · Score: 1

      6 cores to perform the Users' work.
      2 cores to to watch it done.
      and 1 unite them and to bind them in the dark.

      --
      greed@All_Evils:~#
  16. DOH ! by unity100 · · Score: 1, Insightful

    All the new generation, ALL that high price, and it still comes up close with amd's new cpus in multithreaded performance ?

    http://www.hardocp.com/article/2011/11/14/intel_core_i73960x_sandy_bridge_e_processor_review/6

    no wonder there have been 3 opteron (bulldozer) supercomputer orders in the last 3 weeks.

    1. Re:DOH ! by PIBM · · Score: 1

      From 46 FPS (AMD 8150, 8 'cores') to 72 FPS (3960x, 6 cores) is close for you ? Wrong link ? Trolled ?

    2. Re:DOH ! by unity100 · · Score: 1

      MULTITHREADED performance it says. not fps. there are cheaper amd chips which do fps as well with cheaper prices.

    3. Re:DOH ! by PIBM · · Score: 2

      That's the number of frame per second that the MULTITHREADED VIDEO ENCODER was being able to encode. All of the provided benchmark on that page were about video encoding using multithreaded encoders, and the new proc was beating all others easily. That's why I asked if it was the wrong link perhaps?

    4. Re:DOH ! by Yvan256 · · Score: 1

      So the target market is those filthy movie pirates?

      Next up: the MPAA demands 50% of Intel's profits!

    5. Re:DOH ! by PIBM · · Score: 1

      No kids ? No wife ? No video camera perhaps ? Or just not getting out of your basement ? Encoding 1080p video takes a while, and I should soon start encoding dual stream (3d) 1080p.. That kind of power would be welcome for me (sporting an i7-960 on the computer running the video tools). Anyway, that's the original poster link target..

    6. Re:DOH ! by makomk · · Score: 1

      x264 used to be quite heavily optimized for Intel processors, sometimes at the expense of performance on AMD chips. It's not quite as bad as it once was but...

    7. Re:DOH ! by unity100 · · Score: 1

      oh yeah. instead of encoding in the time that it takes for me to go get a cup of tea and back, now my encoding will be complete by i reach the door to wc while passing through the corridor from the kitchen to the living room. yes. that totally justifies shelling out $900 to encode my home videos.

    8. Re:DOH ! by PIBM · · Score: 1

      Some people are not bothered by that 900$. Some people will buy the outrageously priced video cards for the limited benefits. You don't need that 80" LCD TV either, but it's nice to have. Shaving 10-15 minutes off my encoding when I want something is nice to have too. Beside, there's much more to be done on a computer that will benefit from this processor. Think about compiling & linking, where the time saved directly correlate to money in the pocket. Not that I would buy this processor, but there's definitely value in it.

    9. Re:DOH ! by Shinobi · · Score: 1

      Ignore Unity100, he's astroturfing for AMD's marketing department

    10. Re:DOH ! by PIBM · · Score: 1

      I was actually starting to wonder, and I looked up his recent posts. Definitely the case. Anyway, he's bad at it!

    11. Re:DOH ! by kesuki · · Score: 1

      you sir are incorrect, i haven't found the specs online, but only blu-ray devices support 1080p and i was pretty sure 3-d didn't need dual progressive scans but rather two interlaced streams that form a 1080p image. but the wiki on it claimed it uses 50% more overhead suggesting dual 1080p streams. but as the 3ds shows us 3-d doesn't have to mean high definition.

      besides 3d is a fad, with 18% of the population unable to watch 3d (seizures) and many more complaining of headaches it's not likely to be widespread.

    12. Re:DOH ! by unity100 · · Score: 1

      Some people are not bothered by that 900$. Some people will buy the outrageously priced video cards for the limited benefits.

      i am one of those 'some people', and i participate in communities that are populated by those people, and noone will buy something that provides only 15-20% performance, but burns a small oven and is priced at a fucking $900.

    13. Re:DOH ! by DurendalMac · · Score: 1

      So close = beats BD silly? Because I'm seeing it whomping BD in those benches, beating it by over 50% in some cases! Did you even read your own link? And you can astroturf all you want. It doesn't change the fact that BD's single-threaded performance suck dead goat ass.

    14. Re:DOH ! by Shinobi · · Score: 1

      Look at how he responded to one of my posts.

    15. Re:DOH ! by Anonymous Coward · · Score: 0

      I wasn't aware of these details so decided to research them in response to this post. Please keep in mind that single-threaded performance is very different than multithreaded performance. From http://www.spec.org:

      Opteron SPECINT - Maxes out at 26.1 for a single thread vs 55 for a Xeon
      Opteron SPECFP - Maxes out at 34.2 for a single thread vs 73 for a Xeon

      To compare them for multiprocessing ability, the numbers are much closer. A 96-core Opteron is able to produce 1310 SPECFP-Rate, whereas an 80-core Xeon is able to produce 1380. If you keep in mind that AMD's chips tend to be cheaper per core, you can see why some people prefer AMD.

    16. Re:DOH ! by KingMotley · · Score: 1

      Click the link and read please before commenting. The FPS is multithreaded performance. It's refering to how many frames per second it can encode video, and PIBM is right. There is a huge difference between the two processors in those tests, eye balling it, it looks like approximately 42-50% improvement on each of them.

    17. Re:DOH ! by PIBM · · Score: 1

      Side By Side (SBS) encoding is often done with two full quality streams. You can pipe it to two projectors (with polarizing filters) and I know that some TVs can accept the stream too (mine do, at least). I'm not sure on what you meant about the fact that only blu ray devices support 1080p, but I've been running a triplehead setup of 2560x1600 screens since 2007, which makes it 6 times higher in resolution than a 1080p stream. Screen caps are huge, so much that I am not recording much of them anymore.

      It's sure that it does not have to be that high resolution for the 3d streams, but it's sure nice to be able to have them. For the headaches & such, the type of projection plays an important role. I have not setup myself with dual projectors, but I believe it's where I'd like to go. Think IMAX at home.

    18. Re:DOH ! by Anonymous Coward · · Score: 0

      That page shows intel is about 40% better than anything amd had to offer, with 25% fewer cores and only 5% more mhz

    19. Re:DOH ! by KingMotley · · Score: 1

      Apparently, you are NOT one of those 'some people'. Some of my friends have sold their 3 and 4 way NVidia 480 cards to replace them with 3 and 4 way 580's for the 8% performance increase -- at a cost of $900-$1200.

    20. Re:DOH ! by KingMotley · · Score: 1

      My encodes take hours, even on the best of machines today. Encoding high quality video at full HD (or QHD) resolution takes some time and horsepower.

  17. Mac Pro? by CokeBear · · Score: 1

    Could this be the processor for a new Mac Pro? Or will Apple wait for Ivy Bridge.

    I've had a Mac Pro (or its predecessor) under my desk for over 10 years now; upgrading regularly. Even if its not the top selling Apple Product, its still the machine that Pros are looking for.

    --
    Reality has a liberal bias
    1. Re:Mac Pro? by Yvan256 · · Score: 1

      It's it's a 10-years-old "Mac Pro", chances are that it's a Power Mac G4. Even a 2010 Mac mini can easily beat that, apart from the hard drives.

      If you go from using a PowerMac G4 to even a current day Mac Pro, even the low-end Quad-Core model is going to seem ludicrously fast by comparison.

    2. Re:Mac Pro? by CokeBear · · Score: 1

      I meant that I've had some flavor of Apple Pro product under my desk for 10 years, upgrading every 1-3 years. Not the same machine all that time

      --
      Reality has a liberal bias
    3. Re:Mac Pro? by petermgreen · · Score: 1

      Could this be the processor for a new Mac Pro? Or will Apple wait for Ivy Bridge.

      Early next year intel will be coming out with dual socket capable LGA2011 xeons. Those are what I would expect apple to put in the mac pro.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
  18. Intel compiler & Windows OS in benchmarks by coder111 · · Score: 3, Interesting

    This is all nice and well, but are there any sites that actually benchmark this CPU under Linux, running some stuff not compiled with intel compiler? AFAIK most of the benchmark software is running on windows is compiled with ICC, and ICC cheats- it disables most optimizations on non-intel CPUs.

    How about some linux developer workload? Compile times? IDE performance? Java performance? PHP, Apache, PostgreSQL, MySQL performance? KDE/Gnome performance? CAD/CAM? Matlab or Octave? Bzip2/gzip/SSL/zip under Linux? I know some of these workloads depend on IO/graphics more than on CPU, but I'd like to see results anyway. And I'm sick and tired of reviews that run some Intel compiled synthetic benchmarks and then some games that primarily use GPU anyway. Phoronix is guilty of that as well- they should have more WORK workloads and less FPS counts for games. But at least they are trying- and Bulldozer performance under Linux/GCC isn't that bad compared to Intel CPUs as it is under Windows/ICC.

    --Coder

    1. Re:Intel compiler & Windows OS in benchmarks by Anonymous Coward · · Score: 0

      Why would anyone use icc to compile these programs when msvc is much better? I don't think icc has any measurable market share on windows.

    2. Re:Intel compiler & Windows OS in benchmarks by coder111 · · Score: 1

      http://en.wikipedia.org/wiki/Intel_C%2B%2B_Compiler#Criticism

      As far as I remember, some benchmark software (I think Passmark family) is compiled with ICC. If not benchmarks themselves, then a lot of windows system libraries that are used by benchmarks are compiled by ICC. I haven't verified this myself, as I'm not that interested in synthetic benchmarks most review sites use. I should get my hands on some of benchmarking tools and verify what compiler was used to build them- compilers usually leave some strings in executable files.

      --Coder

    3. Re:Intel compiler & Windows OS in benchmarks by Anonymous Coward · · Score: 0

      Additionally, the driver support for the built in Sandy Bridge driver aren't very good at all. With Ubuntu 11.04, if I change the refresh rate I crash the video driver (maybe even the whole OS, I haven't checked). Also, the external displayport doesn't work with 1920x1200 monitors. Intel's documentation is a bit iffy - lacking register indexes and missing essential registers entirely. I wish that some vendor of video cards would take Linux seriously.

    4. Re:Intel compiler & Windows OS in benchmarks by Anonymous Coward · · Score: 0

      This is all nice and well, but are there any sites that actually benchmark this CPU under Linux, running some stuff not compiled with intel compiler? AFAIK most of the benchmark software is running on windows is compiled with ICC, and ICC cheats- it disables most optimizations on non-intel CPUs.

      How about some linux developer workload? Compile times? IDE performance? Java performance? PHP, Apache, PostgreSQL, MySQL performance? KDE/Gnome performance? CAD/CAM? Matlab or Octave? Bzip2/gzip/SSL/zip under Linux? I know some of these workloads depend on IO/graphics more than on CPU, but I'd like to see results anyway. And I'm sick and tired of reviews that run some Intel compiled synthetic benchmarks and then some games that primarily use GPU anyway. Phoronix is guilty of that as well- they should have more WORK workloads and less FPS counts for games. But at least they are trying- and Bulldozer performance under Linux/GCC isn't that bad compared to Intel CPUs as it is under Windows/ICC.

      --Coder

      You could check Phoronix.com and ask for a review comparing both. They do have a Sandy Bridge bench with the kernel 3.2 vs 3.1, maybe if you ask, they might do it.

      http://www.phoronix.com/scan.php?page=news_item&px=MTAxNTQ

  19. Re: Cough by beelsebob · · Score: 5, Insightful

    When buying hardware, trying to future proof is dumb. You could try to "future proof" now and buy a $1500 system. In 3 years it'll be shit though.

    Alternatively, you could buy a $600 mediocre system now, and another $600 system in 2 years that'll be faster than the above $1500 one. The result will be that you've spent $300 less, you've got machines that are reasonably current for 4 years, and the system you get out at the end is faster.

  20. But does it crush AMD's new 16 core processor ? by unity100 · · Score: 2, Interesting

    http://venturebeat.com/2011/11/13/amd-introduces-worlds-first-16-core-pc-microprocessor/

    no it doesnt. not when it comes only close with amd's 8 cores in multithreaded apps. 16 cores , becomes unmatchable.

    1. Re:But does it crush AMD's new 16 core processor ? by DurendalMac · · Score: 1

      That's a server/workstation CPU. Intel will be pushing out 8 core SB-E Xeons early next year. Then you can start making comparisons.

    2. Re:But does it crush AMD's new 16 core processor ? by unity100 · · Score: 1

      early next year, 12-16 core desktop dozers are out. every year, new dozers are out with more cores. that is the amd strategy.

    3. Re:But does it crush AMD's new 16 core processor ? by Anonymous Coward · · Score: 0

      Please go take your AMD trolling somewhere else, this story has nothing to do with a server grade CPU which is what the story you posted is about, the Sandy Bridge-E is a desktop CPU. There is no point in trying to compare them.

    4. Re:But does it crush AMD's new 16 core processor ? by Anonymous Coward · · Score: 0

      You 're comparing a server processor (optimized for servers) with a desktop processor (intel sandy bridge - e), apples to oranges.

      For the opterons parts, Xeon parts should be compared against it.

    5. Re:But does it crush AMD's new 16 core processor ? by KingMotley · · Score: 1

      If you take the benchmarks on x264 encoding, first pass that YOU linked, the 3960x scored 230.11 and the 8150 scored 141.39. I'm not sure that I would call that close. However, let's go ahead and extrapolate that out (with a ton of assumptions of course, of which many are probably inaccurate).

      The 3960x is producing 230.11 over 6 cores, therefore over 8 cores, you might expect 230.11/6*8=306.81.
      The 8150 is producing 141.39 over 8 cores, therefore over 16 cores, you might expect 141.39/8*16=282.98.

      Seems the 3960x (a desktop chip) should be 8.4% faster, if everything else remains equal, and both CPUs scale linearly. Not only that, but the 6282 SE is more expensive, hotter, and uses a more expensive motherboard than the 3960x. I don't see any category in which AMD competes -- it loses in price, performance, AND heat all at the same time.

    6. Re:But does it crush AMD's new 16 core processor ? by DurendalMac · · Score: 1

      Early next year? Give me a break. AMD might have those out by the middle of next year. And yes, AMD's strategy is to keep adding more cores (which will be largely useless on desktop systems) because their per-core performance sucks ass. BD was a huge failure in that arena. What the hell is the point of a 12-core desktop CPU when most desktop users aren't going to significantly stress more than four of them (if that) at any given time?

    7. Re:But does it crush AMD's new 16 core processor ? by smash · · Score: 1

      You know 10 core xeons have been out for like... ages now. And given that 4 intel cores can beat 8 AMD cores, then 10 should quite handily beat an AMD 16 core machine.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    8. Re:But does it crush AMD's new 16 core processor ? by Anonymous Coward · · Score: 0

      However, let's go ahead and extrapolate that out (with a ton of assumptions of course, of which many are probably inaccurate).

      The 3960x is producing 230.11 over 6 cores, therefore over 8 cores, you might expect 230.11/6*8=306.81.
      The 8150 is producing 141.39 over 8 cores, therefore over 16 cores, you might expect 141.39/8*16=282.98.

      Here's one inaccurate assumption: that clocks stay the same as you add cores. For example, the fastest 16-core Bulldozer is the 6282SE, at 2.6 GHz base clock speed. (The FX-8150 base clock speed is 3.6 GHz.) I don't remember what the respective turbo speeds are but I doubt that the 16-core BD has hugely aggressive turbo given how power-limite it is.

      Intel won't be able to stay at the same clocks when they go from 6 to 8 cores either, though it won't be as bad as AMD's dropoff. But it hardly matters given that 141.4 * (16/8) * (2600/3600) = 204.2, meaning there's good reason to expect that Intel doesn't even need 8 cores to destroy 16-core BD in this benchmark.

    9. Re:But does it crush AMD's new 16 core processor ? by Anonymous Coward · · Score: 0

      It probably will. Have you seen AMD's own benchmarks of its 8- and 12-core against Intel Xeon's 6? E.g., doing FEA:
              http://www.amd.com/us/products/server/processors/opteron/Pages/ls-dyna-performance-two-socket-servers.aspx
      Intel's 6 is faster, by a slim margin, than AMD's 12. Since the SBE are even faster, by a decent margin, than current SB, I would expect the margin of SBE over AMD's 12 to be even wider and perhaps even close to AMD's 16 ... Which is as much as you can say right now, since we only have numbers for Intel's SBE, not AMD's 16.
      If, on the other hand, you're price conscious, then Intel's 2700 series (also SB but not E) retails for about $300 (minus $20 in current offer, probably in view of the launch of the first member of the next iteration), i.e., 30% (or 28%) of the price for about 80% of the performance of the SBE ... I would certainly expect this to be close (80-90%) to the AMD's 12 currently retailing at $1440 (496% ~ 5x the price!) Which, again, is as much as can be said at the moment since there doesn't seem to be any actual (as in "I want to buy now") pricing information for the AMD 16 yet.
      So nope, sadly these days you can't go with AMD, neither for raw performance or bang-for-the-buck ... I wish you could, even if only to put some pressure on Intel to release fully functional i7 8-cores, or even their next-gen Ivy Bridge, but nope, Intel's current offerings sweep the floor with anything in AMD's lineup, why run when a leisurely walk will do?

    10. Re:But does it crush AMD's new 16 core processor ? by unity100 · · Score: 1

      no it wont 'probably' do anything. dont make arguments out of your ass with probabilities.

      while intel banked on increasing yield from existing cores, amd banked on increasing cores. bulldozer is a technology that allows slapping cores without new research. this is why it took so long and delayed so long.

      so, regardless of the yield from cores from intel, amd will be able to slap more cores and make them work - as opposed to intel putting 8 core sandy bridge e but making only 6 core work at one time like now.

      when intel comes up with 10 cores which allow 8 cores to work at the same time next year, amd will be putting 20-24 core server dozers.

      you cannot double the yield from a core nomatter how you do. even sandy bridge e's top performer gives 10-15% more performance over 2600. therefore, it is a foregone conclusion.

      intel seems to have lost the core race. this means, they will lose server market. they may stay strong in desktop market. however, server market is much much more profitable and bigger. in addition, in time when yields increase, amd's current chips will catch up to intel eventually in terms of clocks and single thread performance. but intel wont be slapping extra cores that easy.

      there.

  21. That's how it has long worked by Sycraft-fu · · Score: 1

    These are Intel's "enthusiast" parts which generally means "people with too much money". Some people want the highest end performance, price is no issue. Intel is happy to stick a hose in their pockets and siphon out the cash.

    That's also why these came after the regular SB parts. Intel full and well knows that for 99.99% of people a standard SB is more than plenty and they'd like to have something economical.

    In terms of the other things you'd like to see, USB3 and more SATA 3 will probably be coming but Intel is going to need to replace their DMI interconnect with something faster. They'll just run out of bandwidth between chipset and CPU otherwise. Remember that things like that take time to properly design.

      SAS is probably never coming to consumer boards. Too much additional expense for not enough gain. Remember that SAS is actually more complex from a logic standpoint (hence why SAS controllers can handle SATA drives but not vice versa). There just aren't many SAS drives out there in use in consumer systems, and there'll only be less as SSDs take off. After all, what do you really want for a high performance desktop: A 300GB 15k SAS drive for $400 that you have to deal with the noise and cooling, or a 250ishGB SSD for around $400, knowing that the SSD beats the 15k cold in transfer rate, IOPS, seek time and so on?

    Only thing I could really see this being useful for, maybe, is GPU compute systems. Reason is that as stated the regular SB only has 16 lanes of PCIe for graphics, the E series has twice that (the others are for other slots). If you knock 4 cards in a system, that means that you'd be splitting the bandwidth down to 4x per card. While chips like the NF200 can help by providing 16x to each card, they still have only 16x to the CPU and thus only a single card can get full bandwidth at any one time. Presuming you had problem sets that required a lot of swapping to system RAM, this could help. However, I find that to be a rather outside possibility, as to use GPUs to their full processing efficiency, you need shit to fit in the RAM on the card.

    At any rate, stick with Intel's consumer shit unless you have a real reason.

    1. Re:That's how it has long worked by Wesley+Felter · · Score: 1

      SAS is probably never coming to consumer boards. Too much additional expense for not enough gain.

      The X79 (Patsburg) chipset already has SAS; it's just disabled.

  22. they should have a lower cost CPU that works X79 by Joe_Dragon · · Score: 1

    As that is where AMD wins as you can get lower end cpu and use the chipset with a lot of PCI-e lanes.

    Intels other CPU have to few pci-e lanes just X16 + the chipset link is to few. When 1 video card can take up X8-16 pci-e lanes.

    As things like sata 6, usb 3.0, PCI-e based SSD cards, cable card tuners, Thunderbolt, need the pci-e lanes as well.

    Intel pushes the X1 slots, networking, sound, usb, sata, all over that X4 based chipset link.

    At least have 20 pci-e + chipset link so you can have 1X16 or 2X8 + a X4 slot or have Thunderbolt in place of the lanes for the X4 slot.

  23. Don't count on it by Sycraft-fu · · Score: 1

    The Z68 isn't likely to go down in price any, it is the chipset for the IB. The IBs are drop-in replacements for SB processors, same board, same chipset, and all that. In terms of RAM, same deal. They'll still use DDR3. Now that doesn't mean you'll pay much for it, DDR3 is dirt cheap, like $100 or less for 16GB of high quality RAM, but it won't be any cheaper on account of new RAM coming out (RAM is also cheapest when it is in the most production, not because of new tech).

    In terms of the CPU... Maybe. Thing is Intel doesn't tend to reduce prices on older parts. A Q9550 is still $300 from Amazon. For that price, you can get an i7-2600. this will probably continue and you'll find that the IB chips will be no more or less than the SB chips they replace. If you buy used you can get one cheaper, of course, but not new.

    1. Re:Don't count on it by Yvan256 · · Score: 2

      I've got a Z80 right here. I can't imagine how fast a Z68 must be!

      Did they finally break that 65536 bytes barrier?

    2. Re:Don't count on it by tgeek · · Score: 1

      I've got a Z80 right here. I can't imagine how fast a Z68 must be!

      Did they finally break that 65536 bytes barrier?

      Obviously the Z68 is 12 Zs slower than the Z80 . . .

    3. Re:Don't count on it by Yvan256 · · Score: 1

      I read that as Z86 because I'm used to read Z80...

      Has any company ever used decreasing numbers to indicate a new product?

    4. Re:Don't count on it by ravenshrike · · Score: 1

      If we assume that Z chipsets are neutrinos and that Z is a form of geometric energy input, then the Z68 is 12 Zs faster than the Z80.

  24. Are we looking at the same chart? by Sycraft-fu · · Score: 4, Insightful

    What I see is the 2600k, the 4 core $300 chip, matching or beating the Bulldozer, and the 3960X beating everything by a decent margin.

    You are correct in that the Bulldozer doesn't have much to worry about from the new E series as they are much higher priced and compete in a different market. What it does have to worry about is the regular SB chips, which are killing it. Even when things are stacked in what should be its favour: Heavily threaded tasks, the SB does as good or better. Then if you take many other tasks that are not as multithreaded, the SB pulls way ahead.

    THAT is the BD's big problem... Well that and the fact that the Ivy Bridge comes out in a few months. The E series is just for people with too much money. In the consumer market, the regular SB is an amazing performer.

    1. Re:Are we looking at the same chart? by unity100 · · Score: 0

      well you are seeing wrong.

      in various important applications like photoshop cs5, dozer bests intel's existing sbs, all.

      http://www.overclock.net/hardware-news/1150995-extremetech-analyzing-bulldozer-why-amd-s-25.html#post15475610

      moreover, you are able to overclock bulldozer to ramp up your performance. and its possible to go to comfortable 5ghz with air coolers.

    2. Re:Are we looking at the same chart? by DurendalMac · · Score: 2

      Cool cherry picking, bro. Too bad the 2600k spanks it silly in the vast majority of other tests. Keep clinging to one or two benches as proof of performance while the rest of us laugh at you. And you can't hit 5ghz on air with BD in typical cases. It'll eat enough power to run a small town. A 2600k can do that well enough without blowing your power bill through the roof.

      It's always funny seeing AMD fanboys desperately clinging to Faildozer. Most of them had sense enough to realize what a joke it was.

    3. Re:Are we looking at the same chart? by unity100 · · Score: 0

      are you looking down on fucking photoshop cs5 ..... bro ? that is one program that many people working in graphics, web, or publishing have to work with. most of the other benches there, are stuff that percentages of population only use for specific needs. noone will universally require you to use them.

      and, you can hit 5ghz on air with bd, and you can hit it good. search youtube. everyone is doing it.

    4. Re:Are we looking at the same chart? by KingMotley · · Score: 2

      I find it funny that you posted a link to another forum where people are calling you a whiner in the very first post.

    5. Re:Are we looking at the same chart? by DurendalMac · · Score: 1

      Nice dodge, dumbass. I'm pointing out that CS5 is ONE application. Did you see the rest of the CS5 suite? The i7 2600k soundly beats BD. Are you looking down on the rest of CS5? Yeah, no one ever uses Premiere, After Effects, or hell, any other software out there as the i7 tends to beat BD in 95% of it. Do you think that Photoshop is the only application people will ever use?

      I hope for your sake that you're just a troll or a paid AMD astroturfer. Otherwise you have some painful deficiencies...

    6. Re:Are we looking at the same chart? by mobby_6kl · · Score: 1

      >Do you think that Photoshop is the only application people will ever use?

      Well, maybe he's just an Mac user.

    7. Re:Are we looking at the same chart? by smash · · Score: 1
      Yup. Bulldozer's big problem is that AMD kinda fucked up. 3x as many transistors and intel, for what exactly? A nice big heap of fail.

      I'm sure they must be scratching their heads wondering what happened - and i'm equally sure that the successor to bulldozer will fix whatever glitch is currently hobbling the machine.

      Well, at least they better do, or AMD is going to go broke. They haven't had a CPU worth shit for years now - ever since the core2 came out, intel has spanked them quite handily.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    8. Re:Are we looking at the same chart? by DurendalMac · · Score: 1

      If he's a Mac user then he wouldn't be an AMD fanboy.

  25. What else are they supposed to bench it against? by Sycraft-fu · · Score: 5, Informative

    There are only three things that you can bench it against usefully:

    1) The 2500/2600k CPUs that are the high end for the consumer boards. The question there is "What do I get moving up to the much more expensive E series?"

    2) The top of the line AMD Bulldozer. The question there is "How much faster is Intel's high end than AMD's high end?"

    3) The previous Intel high end, the i7-990X. The question there is "How much faster would it be if I upgraded?"

    In all cases, you are talking a very high priced, over spec'd part. There are no other chips in its category really. It is for people who demand the max performance and aren't concerned with the stiff price premium to have it.

  26. Why not put QPI in all CPU's and open it up? by Joe_Dragon · · Score: 1

    So you can have more chipset choice like AMD?

    1. Re:Why not put QPI in all CPU's and open it up? by petermgreen · · Score: 1

      The vast majority of systems have either no graphics card at all or one graphics card. Intel optimises their mainstream CPUs for that case whether hobbyists like it or not. AMD are also moving in that direction (look up socket FM1).

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
  27. Probalby not by Sycraft-fu · · Score: 1

    Apple uses workstation class components for their Mac Pros. In terms of Xeon CPUs, I'm not sure how Intel is going to handle it. They've had SB Xeons for awhile now, the E7-8830 is an example. They have some features you see in the new SB-E chips, some that you see in the normal SB chips. They are also a different socket from either. I don't know if they plan a separate Xeon line or not.

    At any rate, Apple is likely to stick with Xeons, that is just how they do things for better or worse. When will they do it? Who knows? Could be soon, could be never. They seem less and less interested in the pro market all the time.

    1. Re:Probalby not by petermgreen · · Score: 1

      They've had SB Xeons for awhile now, the E7-8830 is an example.

      At least according to wikipedia that chip isn't actually SB based.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    2. Re:Probalby not by Anonymous Coward · · Score: 0

      They've had SB Xeons for awhile now, the E7-8830 is an example.

      Incorrect. The E7-8xxx series is based on Westmere and is for multiprocessor configurations. The only Sandy Bridge Xeons are the E3-12xx series but it is uniprocessor only. Dual and multiple processor Sandy Bridge Xeons are not expected until Q1 2012.

  28. Re: Cough by Anonymous Coward · · Score: 5, Insightful

    The result will be that you've spent $300 less, you've got machines that are reasonably current for 4 years, and the system you get out at the end is faster.

    You're also sending twice as much to the garbage pile.

    I know this isn't a consideration for most, and it's all but encouraged through the new "disposable electronics" thing that's crept up over the last decade, but at some point we need to consider that some considerations extend beyond the financial, even when talking about buying consumer goods.

    For instance, I know people that buy a new printer every time their starter ink runs out because it's still cheaper than buying replacement ink cartridges. Three times a year they're throwing a perfectly good printer into the trash. Yeah, it saves them money, but does that really make it right to throw it in a landfill? I have a hard time saying yes.

    Maybe if we required manufacturers to subsidize the disposal of their goods when such goods are non-biodegradable it would help do something to eliminate the whole "designed for the dump" phenomenon?

  29. Re: Cough by beelsebob · · Score: 2

    You're also sending twice as much to the garbage pile.

    Not that your point in general is wrong, but you're actually sending about 150% to the garbage pile, not 200%. One machine every 2 years, instead of one machine every 3.

  30. Re:they should have a lower cost CPU that works X7 by beelsebob · · Score: 1

    As things like sata 6, usb 3.0, PCI-e based SSD cards, cable card tuners, Thunderbolt, need the pci-e lanes as well.

    1) SATA 6 and USB3 is typically on the mobo not using up PCIe anyway
    2) The thing is, the number of machines not owned by enthusiasts that actually use these other things is tiny. Are you seriously going to spend thousands on a PCIe SSD, but not be willing to shop out for a decent CPU too? I have no idea who actually uses a TV tuner any more with on-demand services being infinitely more convenient, and thunderbolt is a) something that should be integrated into the graphics card anyway b) something which only really benefits laptops, as it's pretty much the generic docking station and nothing else.

  31. 10% != "well ahead" IMHO by sgt+scrub · · Score: 1

    The Core i7-3960X Extreme Edition finished well ahead of the second-place Core i7-990X

    I don't see any benchmark placing the 3960 more than 10% faster than the 990. How can 10%, and under, be "well ahead"? The FPS tests are all under 2% in favor of the 3960. $1,000 + Motherboard upgrade for 2%? With the Icy Bridge you will get a die reduction. This means at least you will get a power consumption drop.

    --
    Having to work for a living is the root of all evil.
  32. Re: Cough by PopeRatzo · · Score: 1

    You're also sending twice as much to the garbage pile.

    Blame Intel for their cockamamie upgrade path that makes every component obsolete every 12 months.

    Why even bother buying motherboards and processors separately if they're both going to become obsolete at the same time thanks to Intel?

    --
    You are welcome on my lawn.
  33. Re: Cough by Mal-2 · · Score: 3, Informative

    The result will be that you've spent $300 less, you've got machines that are reasonably current for 4 years, and the system you get out at the end is faster.

    You're also sending twice as much to the garbage pile.

    I know this isn't a consideration for most, and it's all but encouraged through the new "disposable electronics" thing that's crept up over the last decade, but at some point we need to consider that some considerations extend beyond the financial, even when talking about buying consumer goods.

    For instance, I know people that buy a new printer every time their starter ink runs out because it's still cheaper than buying replacement ink cartridges. Three times a year they're throwing a perfectly good printer into the trash. Yeah, it saves them money, but does that really make it right to throw it in a landfill? I have a hard time saying yes.

    Maybe if we required manufacturers to subsidize the disposal of their goods when such goods are non-biodegradable it would help do something to eliminate the whole "designed for the dump" phenomenon?

    A machine that was bleeding-edge two years ago is still quite powerful today for the majority of people out there. Also, not every server in the rack has to be equal. There are plenty of less-demanding but still important roles that two year old machine can fill when it is kicked down a notch. I'm sure the "weakest link" hardware can be put to good use elsewhere when upgrade time rolls around. I know I consider the mobo-CPU combo as a unit now, rather than thinking "I can upgrade the CPU later". Maybe I can and maybe I can't, but it doesn't matter that much. So long as I have room to boost RAM and storage, I can extend the useful life of the hardware a great deal. It just may not be my fastest, l33t3st system any more. At worst, I can give the machine away -- my 3 year old secondhand hardware is generally as good as most people would buy off the shelf new, and I already have a good idea what it does best.

    --
    How is the Riemann zeta function like Trump rallies? Both have an endless number of trivial zeros.
  34. Re: Cough by TheRaven64 · · Score: 3, Insightful

    You're also sending twice as much to the garbage pile.

    Only if you actually throw the machines away. If anything, it's the reverse. If you upgrade components of a machine, it's much harder to find a use for the bits you remove than if you replace the whole machine. A two year old machine may be underpowered for you, but there are lots of people who can use it for another few years.

    --
    I am TheRaven on Soylent News
  35. Re: Cough by DurendalMac · · Score: 1

    What kind of moron would throw a still semi-decent computer in the trash? Sell it for $250 to recoup some of the loss. And who says it would end up in a landfill? I don't know about you, but my state has a recycling program. I can drop it off at Goodwill and it gets recycled.

  36. Re: Cough by RyuuzakiTetsuya · · Score: 1

    I bought an PS3. 400 bucks and I haven't had to upgrade since. Maybe I might buy a slick SSD and throw it in there for giggles.

    --
    Non impediti ratione cogitationus.
  37. Re: Cough by UnknownSoldier · · Score: 2

    > When buying hardware, trying to future proof is dumb. You could try to "future proof" now and buy a $1500 system. In 3 years it'll be shit though.

    That's total nonsense.

    a) The gaming rig I built in 2001 lasted until 2008 -- upgrading the video card from a GeForce 2 to GeForce 4 to GeForce 6600GT kept it alive much, much longer.

    b) I just priced out a complete gaming rig for a friend based on what I have. For $1300 you can build a gaming rig that WILL be perfectly fine for gaming in 3 years. By that time, you can upgrade the video card and it will play all the latest games.

    You _do_ know that MOST games are GPU bound at 1920x1080, not CPU bound right?

    Here are the benchmarks to back that claim up:

    Sabertooth 990 FX
    http://www.guru3d.com/article/asus-sabertooth-990fx-review/17

    And proof that GPU's are the bottlenecks in the latest games ...
    http://www.tomshardware.com/reviews/battlefield-3-graphics-performance,3063.html

    I could go on, but there would be little point.

    So go ahead, and blow $500 on that latest i7-2600K -- meanwhile us budget AMD guys will be putting money that we saved on buying the cheap 4-core system towards a high end GPU like the 6970, because you are forgetting one tiny, but important fact. Almost ALL the big PC games are designed to run on the 5 year old console hardware -- the PS3 and XBox 360 only have ~6 core and ~2 cores respectively, which means modern CPU's are NOT the bottleneck -- the GPU's are.

  38. Re: Cough by level_headed_midwest · · Score: 2

    LGA2011 really replaces LGA1567 for the Xeon MPs. Intel's on-again, off-again planned replacement for LGA1366 is actually LGA1356. However, it seems like Intel may somewhat simplify their socket lineup and just have LGA1155 for general uniprocessor platforms and LGA2011 for performance UP and all multprocessor applications.

    --
    Just "gittin-r-done," day after day.
  39. Re:they should have a lower cost CPU that works X7 by 9jack9 · · Score: 1

    I have no idea who actually uses a TV tuner any more with on-demand services being infinitely more convenient

    I use a TV tuner. Not very often, but it's part of my disaster plan. When disaster strikes, such as a power outage, or traveling, it's still possible to plug in the TV tuner and get digital broadcast TV. That way my wife doesn't miss DWTS. It's all about continuity of service, dude.

  40. Re: Cough by beelsebob · · Score: 1

    So go ahead, and blow $500 on that latest i7-2600K -- meanwhile us budget AMD guys will be putting money that we saved on buying the cheap 4-core system towards a high end GPU like the 6970

    1) the i7 2600k is $300, not $500
    2) You seem to have completely misread my post – I was advocating not buying insane CPUs and instead buying something middle of the range.

    That said –I wouldn't be buying anything AMD for this build –the Phenom II X4s are beaten by the i3s on price and performance, and the X6s are beaten by the i5s on price and performance.

  41. Re:they should have a lower cost CPU that works X7 by BlackSnake112 · · Score: 1

    I have 3 TV tuners in a system. I record all the TV I want to watch. Then watch it when I want to. I have found the on demand services to be lacking for the show that I want to watch. They also forced me to watch on their schedule. The show would go away after a while. I record them and watch when I want to. it is great to not have to wait for all the 2 part TV shows unless it is a season ending cliff hanger.

  42. Version update for those ruby slippers? by sgt_doom · · Score: 1

    I thought they were for the entrance and exit routines for Oz???? (That Wiz can be pretty tricky, ya know?)

  43. using your Z80 by Anonymous Coward · · Score: 2, Funny

    If you posted that using your Z80, THEN I'm impressed. ;)

  44. Wasted cores, poorer cooling performance? by MobyDisk · · Score: 1

    ...the chip's die actually has eight cores on board (two inactive), due to power and yield constraints...

    So that means that some of these chips might have 7 or 8 cores functioning, but they still won't enable them because it would probably exceed their TDP. I wonder if it is possible to find out which cores are working and enable them? This might mean that once the yields go up they could start binning them and sell 7 and 8 core versions. The other interesting thing is that this chip won't perform as well as a 6-core chip with only 6 physical cores, because it will be harder to dissipate the heat when there are 2 cores just sitting there acting as an insulator.

  45. Re:What else are they supposed to bench it against by JoeMerchant · · Score: 1

    It is for people who demand the max performance and aren't concerned with the stiff price premium to have it.

    For my, personal taste, I prefer Atoms - they draw less power, I can leave 3 of them them booted and running 24/7 for the cost of a single "barnburner" like the latest i7 running 8 hours a day, and they do what I need.

    However, if you need the performance, increasing the cost of a processor from $150 to $1150 is only going to increase your total system cost by a factor of 3 (2 if you pay for a copy of Windows...), for a performance increase of 4x or more....

    The value proposition is there, much better than the (warning: bad car analogy ahead) Ferrari to Honda performance vs. value proposition.

    The question is: do you really need it? If all you're doing is booting to a web-browser 30 seconds faster, buy an Atom and leave it booted up - hell, buy an Atom for every room in the house and save yourself the time spent walking to/from the PC.

  46. Re: Cough by Anonymous Coward · · Score: 1

    If you upgrade components of a machine, it's much harder to find a use for the bits you remove than if you replace the whole machine.

    I've got some of that problem going on. The leftovers aren't a whole computer; they're really a motherboard+RAM+CPU. It's not like I need to upgrade the 10-year-old-ATX case or 5-year-old power supply. So I need to find someone who wants computer "guts" which aren't a whole usable computer. That rules out anyone who isn't a computer-dork. But computer-dorks don't want my ancient electronics. All that's left is to BUY MORE EQUIPMENT to turn the obsolete guts into an obsolete system, just so it'll be usable enough to have sufficient appeal that some non-dork will be willing to take it off my hands for free.

    It's not worth the trouble; at some point you just have to left it go, and allow it to poison some poor Chinese solder recycler and a few generations of people who live whereever he works. There isn't a good secondary market for motherboards. And I sort of understand that; I don't like to buy used motherboards, either. I don't trust 'em.

  47. Re: Cough by billcopc · · Score: 1

    Sure, that's the common sense approach which works for most normal users, but there are some of us who directly benefit from having a faster machine NOW, typically because we have a real business need for it. I could do my work on a "mediocre" system, but it would slow me down significantly, and after a month or two, the time (=money) lost waiting for the slower PC exceeds the cost delta of the faster machine.

    The problem with Intel's latest processor is it isn't that big of an improvement over last year's 6-core line-up, putting the 3960X in a strange niche that only interests a small pocket of wealthy e-peeners. Those who want more cores would go for a dual Xeon, and those who want gaming performance would overclock the tits out of a 2600K. Four-slot SLI has never caught on, it's just an experimental curiosity. Even 2x dual-GPU rigs are notoriously unreliable thanks to crappy driver support, so you stand to gain very little from this X79 platform at this point in time.

    --
    -Billco, Fnarg.com
  48. still has HyperTransport by Joe_Dragon · · Score: 1

    still has HyperTransport

  49. Re: Cough by UnknownSoldier · · Score: 1

    > 1) the i7 2600k is $300, not $500

    i7 = Mobo+CPU = $500
    AMD = Mobo+CPU $250-$300 (depending if you want Crossfire or not)

  50. Re:What else are they supposed to bench it against by Sycraft-fu · · Score: 1

    A SB-E? No I don't need it. A regular Sandy Bridge? Yes, and thus I have one.

  51. Re: Cough by beelsebob · · Score: 1

    Uhhh, the FX-8150 is $250 alone, where are you getting this magical $0 motherboard from?

    Other than that, your figure for the i7 is way off. Good quality Z68 boards can be had for $100, if you want to drop crossfire as you suggest, more like $70, so you're looking at $370 for a system that blows any AMD system out of the water.

    Scale back to the i5 2500 and you're looking at $270 for the mobo and CPU, only $20 more than the bulldozer CPU on its own.

    Scale back further to the i3 2100 that can beat the 4 core Phenom II X4 you're talking about and you're looking at $180 for motherboard and RAM, which is less than the Phenom II X4 system costs.

  52. Free by Anonymous Coward · · Score: 0

    http://tinyurl.com/free-satilite-tv

  53. Re:What else are they supposed to bench it against by Synerg1y · · Score: 1

    There are plenty of valid business scenarios that constitute increased ROI from using such a chip as opposed to a lesser chip. This thing should be getting benched against the i7 extreme, not the amd proc which can't compete w the extreme to start w (all due respect to AMD it wasn't meant to article is just a little dirka dur).

    Also remember, you can OC... this processor, so if your willing to take the jump, you can get even more out of it, and the proc isn't limited by the software your running, it's simply designed to run a lot of software at once whether the software is designed to run on multi core arch or not. If you have a lot of apps you need open at once as required with a lot of dev / graphics positions, for a company the increased productivity might steep the cost of the proc, of course not every IT person can take advantage of it either.

  54. Re: Cough by ravenshrike · · Score: 1

    Unless you're running massive amounts of video/audio encoding, the 2500k currently beats everything on the market price/performance wise. As you overclock it gets even closer to the 2600k in terms of performance in everything but audio/video encoding as well.

  55. Re:What else are they supposed to bench it against by ravenshrike · · Score: 1

    If you're a gaming enthusiast the answer is jack all. The 2500ks clock up to 4.8 at similar rates, and moreover can do so on(admittedly very good) air cooling, and while the extra L3 cache will help, it won't help enough to justify the extra $100, assuming they don't drop the price of the 2500k even further.

  56. Good post Vigile, thanks... apk by Anonymous Coward · · Score: 0

    Great reads, nice to see things are advancing on the CPU front... which is great - as I only buy once every 3-5 yrs. typically (between this & the advances in video too? Great gaming + work rigs too that will probably DOUBLE what I have now (Intel I7 Core 920 + NVidia GeForce 470 GTX).

    APK

    P.S.=> I wonder how fast + how many cores we'll see on silicon based CPU's? I heard 4-5ghz is the "limit" ghz-wise on it, until they "change the game" & go to something like gallium arsenide (which I heard is risky & poisonous to work with plus isn't as plentiful or economically 'efficient' to produce)? apk

  57. Re: Cough by smash · · Score: 1

    Also, we're presuming that my system is not going going to be handed down/on to someone who isn't as demanding on the machine. I go through work machines every 12 months or so, it doesn't mean they go in the bin - they just get handed onto users who don't need anything even remotely special.

    --
    I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
  58. Re:they should have a lower cost CPU that works X7 by smash · · Score: 1

    You're assuming there is a benefit to running video over 16 pci lanes. News flash: PCI x16 is still slow for video. Get a video card with an appropriate amount of memory and you should not be hitting the PCI bus enough to matter. If you're hitting the PCI bus constantly your performance is already fucked, whether it is x4, x8, or x16. AMD may win the PCI bus race, but they lose in CPU speed, heat, stability and power consumption.

    --
    I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
  59. sure. by unity100 · · Score: 1

    thats probably why all the 3 supercomputer orders in the last 3 weeks have been placed by amd bulldozers. because 10 core xeons 'quite handily' beat an amd 16 core machine ..

    go fuck off.

  60. Re: Cough by cheekyboy · · Score: 1

    where in brazill?

    You can get MBs for $99 for intel, and the AMD idles at higher watts so over 4 years, your power bill will double your cpu cost.

    --
    Liberty freedom are no1, not dicks in suits.
  61. You would put a computer in the garbage? by Grog6 · · Score: 1

    There's always somewhere to use last generations' computer...

    My p4 is used for programming my car, and the K6-II 450 is my TV computer for my bedroom; it will run mythTV.

    All the later one will still play games, so they're loaded with whatever was current at the time.

    Oh, I guess using more power is bad too. Oh well. :)

    --
    Truth isn't Truth - Guliani
  62. Re: Cough by DarthVain · · Score: 1

    Or if you are like me, you don't throw anything away, only re-purpose them as web servers, file servers, media pc's, bookshelves, small tables, dust protectors, ancient artifacts, and curious images of days of yor.