Intel moving to six-cores this year; What will you do with them?

Intel moving to six-cores this year; What will you do with them?

Summary: Intel outlined plans to produce six-core chips in the fourth quarter. The move is that latest salvo in a game to out-core AMD on everything.


Intel outlined plans to produce six-core chips in the fourth quarter. The move is that latest salvo in a game to out-core AMD on everything. AMD finally gets its quad-core chip out the door and Intel moves to six. The future looks like this: Four cores, six cores, eight cores, 10 cores, 12 cores. Oh my. Who wouldn't want all those cores?

core1.pngPat Gelsinger (see statement and PDF of presentation), Intel's senior vice president and general manager of the chip giant's Digital Enterprise Group, gave some details about the latest six-core chip called "Dunnington" and a new Itanium processor called "Tukwila." Some reports went for the six-core news hook. Others, like's Tom Krazit, focused on software developers.

But like that annoying smart arse in the back of the room, I'd like to pose a question: What exactly are we supposed to do with six cores when we have barely figured out what to do with four? Yeah, we know. Smaller and faster (gallery right). If Intel wants to really get fancy, it'll talk you in circles about multithreaded applications and parallelism. Just imagine what these multithreaded apps could do. The problem: Few software vendors have multithreaded apps. Microsoft is looking into it.

Simply put, the core wars just don't bring that much to the table right now in terms of killer apps--unless you count virtualization as the thing that will get the masses enthusiastic. Intel's server chip roadmap assault is really about running circles around AMD, tick tocking to new cores, and most of all preserving if not improving the average selling price of chips. All you have to do is buy more cores and figure out what they'll enable later. Intel needs more cores to thump AMD. You may not need them.

This economic prism on the core wars gets a little annoying because it really puts the kibosh on great headlines. Consider the following:

Intel says:

Intel's current 7300 chipset based platform combined with the Quad-Core Xeon 7300 processor is the industry's virtualization platform of choice for MP servers. Dunnington is socket-compatible with the Caneland platform and will be available in the second half of 2008. Dunnington is the first IA (Intel Architecture) processor with 6-cores, is based on the 45nm high-k process technology, and has large shared caches. Another supported feature is FlexMigration technology, which allows a single compatible virtualization pool that supports live VM (Virtual Machine) migration across both 65nm and 45nm high-k Intel Core microarchitecture-based servers and 45nm-based servers. This provides investment protection as well as the option to choose the right server platform to best optimize performance, cost, power and reliability.

In English: Intel is moving to six cores to make AMD's Barcelona irrelevant. On deck: The price squeeze.

Intel says:

Tukwila is Intel's next-generation Itanium processor with four cores, 30MB total cache, QuickPath Interconnect, dual Integrated Memory Controller and mainframe-class RAS features. It is the world's first 2 billion transistor microprocessor and is projected to deliver more than double the performance of the current generation Itanium processor.

In English: You can figure out what to do with this performance later. Tukwila sounds pretty damn good eh?

Intel says:

Nehalem will provide dramatic performance and energy improvements to Intel's current industry-leading microprocessors. Nehalem is scalable with future versions having anywhere from 2 to 8 cores, with Simultaneous Multi-threading, resulting in 4 to 16 thread capability. Nehalem will deliver 4 times the memory bandwidth compared to today's highest-performance Intel Xeon processor-based systems. With up to 8 MB level-3 cache, 731 million transistors, Quickpath interconnects (up to 25.6GB per second), integrated memory controller and optional integrated graphics, Nehalem will eventually scale from notebooks to high-performance servers. Other features discussed include support for DDR3-800, 1066, and 1333 memory, SSE4.2 instructions, 32KB instruction cache, 32KB Data Cache, 256K L2 data and instruction low-latency cache per core and new 2-level TLB (Translation Lookaside Buffer) hierarchy. These technical improvements will result in performance improvements as well as flexibility for a wide range of eventual products based on the Nehalem architecture. Gelsinger also discussed the new Tylersburg platform, which can be configured for both one socket High End Desktop (HEDT) and two socket (HPC and dual processing server) operation.

In English: Intel is trying to make you drool. One issue: None of us have seen the power of multi-threading in a software application we use every day. There's no "I gotta have it" moment. Maybe Microsoft will figure something out, but by most accounts the company is just starting to get its head around multi-threaded apps. In the meantime, running Office faster isn't much of a sales pitch.

Bottom line: Next time you get lured by multiple cores, think about Intel's average selling prices. In the end, that's what the core wars are about. UBS analyst Uche Orji is no dummy. In a research note, Orji isn't drooling over six cores, eight cores or any other multiple other than Intel's stock price.

Orji writes:

As Intel's tick-tock execution strategy is based on building upon its prior generations, its upcoming Nehalem architecture builds on its current performance leading Core family. Most important with Nehalem is our belief that Intel will be able to raise average selling prices on its server processors by 9% in 2009 (vs flat in '08) based on a ~ 50% performance increase. Adding the secular trends to faster speed grades & lower power could result in ASP upside.

That riff roughly translates into more performance at a slightly higher price. It's up to you to figure out what exactly you'll do with more power.

Topics: Servers, CXO, Hardware, Intel, Processors, IT Employment

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


Log in or register to join the discussion
  • Oh please..

    "..But like that annoying smart arse in the back of the room I?d like to pose a question: What exactly are we supposed to do with six cores when.."

    Here we are 25 years later and pundits are still questioning why we need more performance. Dude, I'd hardly call you a "smart" arse.

    The part on "what Intel really means" is equally nostalgic. Again, it's the old "any product Intel produces is to kill AMD" and "any product AMD builds is to save human kind with innovation". Oh please. You guys are starting to look like theinquirer,net.
    • That wasn't the question

      "Here we are 25 years later and pundits are still questioning why we need more performance."
      The question wasn't "Why do we need more performance", the question was "What do more cores do for performance right now." The answer to the 'real' question is that they don't do all that much yet. Not beyond two cores anyway, as OS's, unless they're tuned (and who tunes thier OS? Not my folks...) don't take advantage of more cores as readily as they should yet. Will they? I think so, but it'll be a while.
      • Same answer

        "what do more cores do for performance right now"?

        Well, same answer to "what does more MIPS do for performance right now". Same old same old.

        More cores is just a new technique to deliver more performance than more cycles on one core. The objective is the same and the usual "why do we need more.." is also the same.
        • not exactly the same

          <<Well, same answer to "what does more MIPS do for performance right now".>>

          That is not exactly true, as the basic question by Larry may also spell out as "for the same row processing power is it best/ok to have multiple cores"

          The answer to that depends on the time of use you have. For any multi tasking environment with each individual task requiring less that row power offered by each core, having more core is a good deal as it reduce latency involved with task switching, and provide native multitasking, not emulated one. If some tasks require more row power than that provided by a single core, then if these applciation are not multi threaded having multiple core is a bad deal.

          We are now in a situation that is a mix of the above situation. We have plenty of small apps running that do not require that much cpu, but a waiting to data; but on top of that, ensentially on the desktop arena, we run a few cpu ungry applications which are not yet multi threaded; as a result, multiple cores lead to a better user experience ( in the desktop arena ), but restrict the usage we can have of available row power.
        • Still doesn't answer the question

          I am running 2 complete applications. I need no more applications running now nor do I in the near future. What does having 6 cores do for me beyond using the first 2 than 4 cores or even 2 cores. Just how many applications will you be running at any given time that would make having more cores available give you a significant gain. My guess is not many.

          Then you have the problem of the users trying to remember what is where and how do I get to it. Makes no sense to me. Make the existing cores faster and you gain something. Adding things you will never use doesn't sound like a big gain to me at all, just more complexity and probably a loss in what you already have working.
          • Sorry - explanation on prior post

            In the first paragraph the 2 complete applications was meant as an example, not as my current circumstances.
          • Hardware/Software Balance

            CPU's have greatly exceeded most users needs. A word processor, the internet and a few multimedia apps are not capable of tieing up the dual core CPU's unless you go out of your way to do so.

            Software needs to be designed to take advantage of the processing power. Even so the old bottlenecks (Ram and hard drive speed) become an issue before CPU performance.

            There are always one or two aps that can actually use the existing performance, but such a small percentage of the users require this speed.

            Software still has a few hurdles left to break. Voice communication and thinking are two examples. When an enticing piece of software emerges that requires CPU brute force I will be on board like most of the public. Until then I will take advantage of lower prices.

            My rule of hardware purchasing has always been to buy what you need today not what you think you need tomorrow.

            Laptops with dual CPU's can now be purchased for $500 or less. There is no need to pay more than that unless you have extrodinary needs. Desktops are even cheaper. If you want to make a media center out of a laptop, you can still do so for less than $500.

            This was a pipe dream five years ago. Look at the hardware trends and make your own decisions, but computers will continue to decrease in price and get faster. Let your software determine your hardware purchases. In a few years hardware will be cheaper and better, and hopefully the software will make use of the speeds (or a chip maker will concentrate on usable performance upgrades to secure the competitive edge)
      • Say what you will.. I have a need for them

        I plan on saving the world and curing cancer.

        Fold On!!!
        • Uh....right....sure dude.

          And after that? What are you a friggin' slacker?
      • 4800 to Q6600 ... yawn

        I moved from a dual-core AMD 4800 to a quad Q6600 for the reason I need VT stuff to test Hyper-V.

        No real difference in performance.
        • I am a believer in well; 4800 AMD and 9850. anymore is not reality.

          My VT is runtime (Java)Environment, correct me if I'm wrong but things are moving slowly.
    • yea right...

      its acctually guys like you, "prognosticator" that we have all this useless technology at our disposal. Dude we seriously dont need so many cores for the simple reason that the article mentioned. We dont have any apps to go with it. Why upgrade something that is not yet broken and my friend going the multithreaded approch is marching into dangerous waters because its not at all easy to write programs for multithreaded...and now we have 6! ha! please go and get your self those 6 cores and show off to your friends that your system is a weeeeeeeeeee bit faster then theirs.
      • All my apps are threaded - yours are too.

        Not sure what you are running but all of my key applications are threaded. If you use Windows XP Pro, <crtl><alt<del> and inspect the "threads" to see that apps can be issued to resources - and they are.

        As far as "...tell your friends...blah blah..wee bit faster..", my Adobe image and video applications use filters that will take everything you throw at it. So, you might consider, it's not about some useless 'average' of performance improvement like the article author alludes to, but rather performance improve the applications you care about.
    • Performance?

      It's not additional performance if it's not being used. With current software, the extra processors are all but twiddling their thumbs. He's not questioning the need for more performance. He's questioning the availability of it right now.

      As for the nostalgia bit, I'd say you're feeling a bit touchy today.
      Dr. John
  • My chip is bigger than yours...

    This game of rather silly one-upmanship is getting old fast. We can hardly use 4 cores on a consistant basis, and two cores is really plenty if you didnt skimp when you bought the processor (running an AMD X2 3800, no issues for mild to heavy gaming, converting, net browsing). This whole thing is getting childish, Intel has AMD beat for raw speed, and have for a long time, but this is rediculous.
    Critical Mass
  • On the usage of cores

    Of course the first usage of that many cores on a desktop would be CPU intensive applications. I can point to many fields that can benefit, from video to CAD/CAM, software development.... but as you stated, not that many applications are aware of multi threading possibiities, especially when you begin to count past 2 simultaneous threads.

    The second point would be that not that many users are actually running applications with such intense CPU usage. So just switch to the avergae guy average desktop... or at least to what it shoudl look like

    you first of all have the OS with all its fancy GUI effects, you have an aniti virus, anti phishware, anto spyware suite. you have the software firewall, the local search engine with its indexing processes. you have the explorer that calls loads of extension to produce thumbnails. then you have IM software, file sharing software...

    So you have tens of applications running already without having launched a single application. some of those require CPU to respond to user interactions, while other try to work in the background. You have to deal with goint from background dominant apps to foreground dominant apps without having any perceptual impact on the system responsiveness.

    I was talking single user there; but you can easilly log different people at the same time, we almost enver did that in the past, but as OSes have eveolved it is getting more and more convenient. You will have your kids wanting to play a game while you still have all your apps loaded and perhaps running; and believe me the kinds won't want their gama lagging.

    I can see plenty of use for many core, as i can see plenty of use for memory past 4Gb... etc we're not done yet
    • your right, and my amd atholon 3400, w' xp, 2gigs or ram, rocks

      i can stream music with winamp, while playing half life online at the same time, with no issues.

      i can watch a dvd movie (minimize it to 1/2 screen) and work on an offic document at the same time, while have a few instances of firefox open too.

      and why do i need even 4 core's, or even a faster chip.

      this one-upmanship is getting old. (and will really not give me enough performance gains to really know the difference)

      all my apps pop open in a split second anyways.
  • Lousy article

    Sorry but nowhere in the article did you mention these are server chips. And yes there is a use for seevers with more logical cpu's. Databases can do a good job of scaling, as do some other backend applications.

    Talking to clients in the server world the reason they are not loading servers to full capacity is not the problem of scaling, it is reliability. They are not loading the systems further because the operating system and applicatrions vendors have not made systems that a reliable enough. So right now they weugh the cost of mors servers and power/cooling for them, vesus the impact of a crash on a higher number of uses.
    • good point, few remarks

      Databases are mostly I/O and multicore is not the main application here. As pointed out by the article, virtualization is a major selling point. Other good candidates for parallelization are image and video processing, software development, physical modeling, research, and slowly video games as well. etc.

      The point is that desktop systems, nowadays can't benefit from more than two cores, if at all. But once you do use software that benefit from more cores, this scales pretty well. So once you software benefits from 4 cores, it will benefit from 1024 cores. We are in a transition phase, but multicore systems do have a place and will likely scale pretty well for many more years to come.
      • Perhaps the main point

        Is that we're getting cores and no new software that take advantage of the architecture. Hence the What For question. If the argument is more reliability as someone else noted then that's a message that's not being conveyed.

        Maybe I'll do a sequel from the software side.
        Larry Dignan