Say hello to Intel's latest Socket - LGA1155

Say hello to Intel's latest Socket - LGA1155

Summary: Details have emerged about Intel's next-generation platform. Get ready for yet another socket change!

SHARE:

Details have emerged about Intel's next-generation platform. Get ready for yet another socket change!

The new platform, along with the "Sandy Bridge" CPUs, are scheduled to hit PCs in 2011.

Bit-tech.net has the details:

Sandy Bridge CPUs will be based on the current 32nm, second generation High-k metal gate manufacturing process, but will include new architectural features such as AVX - Advanced Vector Extensions - and AES - Advanced Encryption Standard - acceleration and will arrive in Q1 next year, with possibly very limited products popping up late Q4, although depending on who we talk to the story changes.

They will be split into two markets: mainstream and enthusiast. The mainstream models will replace the current LGA1156 'H1' Clarkdale and Lynnfield CPUs (Core i3 and Core i5) and will use LGA1155 'H2' packaging.

There's only one pin difference between the new LGA1155 socket and the existing LGA1156 used by the Core i3 and i5 CPUs, but there's zero chance that current motherboards will work. Why? Because the pin-outs are different, designed to incorporate the integrated GPU built-in to all Sandy Bridge processors.

We've still got a bit of a wait for these CPUs, but I get the feeling that they'll seriously change the processor landscape. These pieces will be game-changes.

Topics: Networking, Hardware, Intel, Processors

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

22 comments
Log in or register to join the discussion
  • Double FAIL!

    Not only is the article a fail on Intel, but ZDnet's ad partner is now serving malware links.

    I got forwarded off to this website:

    h_t_t_p://91.213.157.40/index.php?q=a1c22cf1b047669035a53c30c1c075079110220

    (http portion broken intentionally for your own safety - try this website if you dare)

    This is the second time in as many days that this has happened to me - on two different computers. ZDnet better kill this ad partnership and switch to a company that actually filters out malicious ad space buyers, otherwise I'll quit coming here.
    Joe_Raby
    • How awful. A 404 page.

      Does it only serve malware to IE user-agents?
      AzuMao
      • No, using Opera I got a warning

        It's one of those 'Fake AV' things, which have been in error quite a few times recently.

        Heck, I got one of those warnings on my own site, and there was no ads on it and no malware.
        Lerianis10
      • Your ISP is likely censoring your internet

        I've tried in every browser and it comes up.

        Are you behind the Great Firewall of China or something?
        Joe_Raby
        • Weird, tor and anonymouse also give 404. And China has more infected PCs..

          ..than any other country. The Golden Shield Project censors protesters, not malware distributors.
          AzuMao
  • RE: Intel Socket FAIL

    I'll be laughing when the Phenom II X6 processors come out, and I'll be able to upgrade not only myself, but many happy customers that bought Athlon II X2, X4, and Phenom II X2 and X4 processors in 890GX chipset systems that we're currently shipping out.

    Intel is just crap now. They're all over the map, and they're taking differentiation and market segmentation to extreme levels of insanity. It's just bad for the industry.

    When they got out of marketing platforms like Viiv and Centrino, I knew it was the end of the era for Intel. AMD is only too grateful to take up the task with Vision.

    Intel really needs to go back to the drawing board. This is just pathetic. LGA 775 was on the market for a long time, and they had at least a respectable life cycle with upgrade support. They need to go back and look at their previous designs.
    Joe_Raby
    • I'll be laughing..

      ..when Intel's octo-cores come out, and I don't have to replace my socket 1366 motherboard for them.


      Does it really matter that their netbook/laptop lineup is incompatible with their desktop/server lineup?
      AzuMao
      • You're wrong actually

        Intel is planning a socket change for 8-core CPU's too because the chipset won't support Sandy Bridge anyway. So that $300 motherboard that you bought will be scrap. LGA 1366 has always been a short-lived socket, just like Skulltrail. Nehalem EX is using LGA 1567, just so you know. Sandy Bridge was originally designed for LGA 1156, but now it's 1155. The same shift is happening with 1366.

        Also, your second question doesn't fit with what I'm arguing, which is that the problem is that Intel will now have 3 separate sockets - [b]just for desktops[/b]. Likewise, all of the new LGA 1156 socket boards don't even have SATA 6Gbps and USB 3.0, and all manufacturers carrying 890GX chipset boards are already offering them with both options - AND they'll run 6-core processors at a reasonable price, and certainly a better value than Intel's 6-core mortgage-repeater.
        Joe_Raby
        • Damn you, Intel. :( NT

          [b] [/b]
          AzuMao
    • Modders are a very small segment of computer users

      You realize that most computer users don't do
      chip upgrades - they buy all new PC's.

      I work in the industry. Not once have I ever
      upgraded a chip in a server - by the time the
      processor is obsolete, it's well past time to
      just scrap the box and start over.

      Saying that AMD has some sort of advantage here
      by sticking with a particular socket over a long
      period of time is stupid.
      franklin.brown@...
      • Re: Modders are a very small segment of computer users

        Maybe in the US. But in other areas in the world, such as China, Malaysia, the Philippines, DIY PC are the norm with a lot of shops assembling PCs that are much cheaper than branded products.
        aaarrrnnn
      • No. Throwing away an entire computer and buying a new one from scratch..

        ..when only one component needs upgraded is stupid.
        AzuMao
      • In the middle

        Some IT guys do swap out chips. You can realistically get away with this with AMD.

        Agreed on Desktops this doesn't make that much sense unless you are trying to be on the bleading edge all the time. Most users won't do it.

        But in the server world where a rack may be more expensive, switching from a 2 core to 4 core or a 4 core to an 6-8 core makes a lot of sense. Our guys did it because it gave them instant performance improvement for certain server loads. We do a lot of virtualization and the cores really help there.
        DevGuy_z
    • Return to days of yore

      [i]Intel is just crap now. They're all over the map, and they're taking differentiation and market segmentation to extreme levels of insanity.[/i]

      Wouldn't go that far in dissing the wave of progress they've made with their Core initiative, but it does appear Intel is reverting to what we saw in the 90's+ now that AMD is crawling back in the game. In doing so, AMD is forcing the boys at Intel to keep the price of their processors lower than they'd obviously prefer to.

      So in turn they're choosing to stick it to their client base via the socket interface in hopes of making up the difference - a la sweeter margins - on chipsets, which is where the real money stands to be made at this point.

      You also can't say that there isn't at least some degree of interoperability between Pentium, Celeron, Core i3 and Core i5 desktop models with the Socket T replacement. However with additional and varied interfaces, it'll be the enthusiast class, the ones who build and rebuild PCs frequently and rarely keep their setups intact for long, who stand to lose the most.

      It's a return to the kind of maddening interface diversification we saw years back, a road few should want to travel again. By contrast, the kind of consistency and durability we saw with LGA 775 was like a breath of fresh air indeed.
      klumper
      • ????

        Isn't the whole point of charging less for their stuff to make the client base
        prefer them over AMD's offers? If they lower the prices but make them unappealing in
        some other way, doesn't that defeat the whole point? Why not just save themselves
        the trouble and not do anything?
        AzuMao
        • Answers?

          [i]Isn't the whole point of charging less for their stuff to make the client base prefer them over AMD's offers? [/i]

          Yes. Pricing, like performance, are compelling selling points. It doesn't take hardware enthusiasts and overclocking diehards alone to take notice of that magic bullet.

          [i]If they lower the prices but make them unappealing in some other way, doesn't that defeat the whole point? [/i]

          Sure, but who's to know and fret other than hardware hobbyists? The big OEMs needn't trifle since they cookie cutter everything. But considering the overall numbers and sales reach of the DIY crowd, I guess it's not enough for Intel to lose sleep over in the long run.

          From their standpoint, that base has already served its critical role, in getting the salutary word out at the time of the Core architecture introduction (Nahalem simply rides on its compelling heels). That same vital DIY group - along with hardware reviewers - always helps to drive sales at the outset by creating widespread appeal with their collective [b]Good Computerkeeping Seal of Approval[/b].

          [Same basic formula applies to new OS releases and power users btw.] After that, DIY hobbyist types basically fade into increasing unimportance until the next game-changing architecture is introduced (when hopefully newly enthused trumpets will sound).

          [i]Why not just save themselves the trouble and not do anything?[/i]

          Considering Intel chipsets on mobos always come at a premium, more mainboard sales = more $ chips out the door. Hence improved margins.

          That's not to say there couldn't be wholly technical reasons behind this new course they've chosen [dropping any pretense of cynicism for a moment]. LGA 775 did, after all, offer better power distribution to the processor over the S478 when it was introduced.

          But when you consider that everything from P4 Prescotts and dual core Preslers right up thru the Core C2D and C2Q series all utilized the same rugged L775 array interface, this newfangled socket variegation does raise an eyebrow or two, does it not? You tell me.
          klumper
          • Well they've just got themselves one less DIYer to trumpet their next..

            ..release. Fool me once, shame on them.
            AzuMao
          • I don't like it either

            but what choice is there, short of jumping off their gravy train? One more reason to always appreciate CHOICE.

            Only wish I knew the ultimate reasoning behind this new socket maze they've introduced. Is it mostly geared to improve profit margins, or improve power distribution + performance? Or alas, a little of both?

            Either way, the DIY crowd ends up biting extra dust from the fallout. :(
            klumper
  • ZDNET tooo slow

    I got hit with a redirect. Come on guys get this fixed.
    dave01234
  • Two sides to the story

    1 - different packaging stops people you know no better
    putting the wrong chip in the wrong place.

    2 - new technology and changes to existing tech. require
    changes to the format.

    Think on these and maybe it is not just a money grab, but
    a requirement to advance technology.
    geoffrey.seymour