Why Bother with Three Cores?

The table below shows the problem with four cores:

SYSMark 2007 Overall E-Learning Video Creation Productivity 3D
Intel Celeron 420 (1 core, 512KB, 1.6GHz) 55 52 55 54 58
Intel Celeron E1200 (2 cores, 512KB, 1.6GHz) 76 68 91 70 78
% Increase from 1 to 2 cores 38% 31% 65% 30% 34%
Intel Core 2 Duo E6750 (2 cores, 4MB, 2.66GHz) 138 147 141 120 145
Intel Core 2 Quad Q6700 (4 cores, 8MB, 2.66GHz) 150 145 177 121 163
% Increase from 2 to 4 cores 8.7% 0% 26% 1% 12%

When going from one to two cores, overall system performance increases a decent amount. SYSMark encapsulates a wide variety of applications and usage models and overall performance increases by close to 40%. Obviously areas like video encoding (represented in the Video Creation tests) see the biggest gain, but all aspects of performance increase tremendously. Making the argument for two cores these days isn't a difficult one, most desktop applications can at least take some advantage of two cores.

Looking at the move from two to four cores however reveals much worse scaling. In our 1-to-2 core comparison cache size didn't increase, so the theoretical scaling could actually be even higher but in the 2-to-4 core comparison the total L2 doubles since Intel's quad-core processors are simply two dual-core die on a single package. Despite the increase in cache size however, scaling is quite poor. Overall performance goes up 8.7% percent and the E-Learning/Productivity tests see no gains at all. Once again the biggest gains come from the Video Creation tests, followed by the 3D suite.

For the vast majority of systems, four cores just aren't necessary. There are some applications that do scale very well between 2 and 4 cores, but the overall landscape is much like what we saw with dual-core CPUs circa 2005, the time for quad just isn't now. Intel's CPU shipments also reflect that both the need and demand for quad-core CPUs just isn't very high:

Currently, less than 10% of Intel's consumer desktop CPU shipments are quad-core and that number won't grow much beyond 10% by the end of 2008. But just like the early days of dual-core, we'll see a steady ramp up continuing in the years ahead.

The point here isn't that quad-core processors aren't necessary, rather they aren't quite in their prime as far as demand goes. With such a small portion of the market purchasing quad-core CPUs, the ISVs aren't exactly jumping at the opportunity to make sure all applications scale well from 2-to-4 cores. Some inherently won't scale while others may with additional effort, which requires a large install base and once more we find ourselves in the midst of an overused analogy involving chickens and eggs.

For Intel, a slow adoption of quad-core CPUs isn't much of a problem. It's just as easy to make a Core 2 Quad as it is to make a Core 2 Duo, the former simply has two dual-core die on the package instead of one. For AMD however, things are a lot more complicated.

One often cited reason for Phenom's late arrival was its "native" quad-core design. Due to its on-die memory controller/north bridge, AMD could not simply take two Athlon X2 die and individually place them on the same package; all four cores would have to be behind the memory interface, meaning that all four cores would have to be on the same die.

Number of Cores Manufacturing Process Transistor Count Die Size
AMD Phenom X4 4 65nm 450M 285 mm^2
AMD Phenom X3 3 65nm 450M 285 mm^2
AMD Athlon X2 2 90nm 243M 219 mm^2
AMD Athlon X2 2 65nm 221M 118 mm^2
Intel Core 2 Quad 4 65nm 582M 286 mm^2
Intel Core 2 Duo 2 65nm 291M 143 mm^2
Intel Core 2 Quad 4 45nm 820M 214 mm^2
Intel Core 2 Duo 2 45nm 410M 107 mm^2

Looking at the die size column you can see an issue with AMD's current processor lineup. AMD likes building the 65nm Athlon X2s, they are nice and small at 118 mm^2 per die and it can make a lot of them on a single 65nm wafer. The Athlon X2 6400+ is still built on a 90nm process and its die, by comparison, is huge; AMD doesn't like making these chips very much (update: AMD has actually ceased production of 90nm X2s altogether).

Then we have Phenom. At 285 mm^2, Phenom is huge and AMD can't make that many per wafer, plus with such a large die the yield is lower than on a smaller chip. The triple-core Phenom X3 gives AMD something to do with those quad-core die that have a single defective core, rather than throwing the entire chip away it can now be repackaged and sold as a triple-core processors.

The other problem here is that there is no dual-core Phenom, so AMD must battle Intel's 45nm dual-core processors with its very old 65nm Athlon X2s. The Phenom X3 is designed to help alleviate the burden of those poor K8s by competing with Intel's dual-core in the sub-$200 space. It's a great marketing story too: you can get three cores from AMD for the price of two from Intel.

It's more likely than not that AMD's yields aren't bad enough to have too many quad-core Phenom processors with two defective cores, which is probably why we don't see a 285 mm^2 Phenom X2.

AMD has no plans to make a separate triple-core die, simply because it would require quite a bit of engineering resources and the need for triple-core CPUs diminishes over time as quad-core adoption increases. Right now AMD is focused on bringing its 45nm Phenom processors to market and those are occupying all of AMD's availability engineering resources. Should triple-core prove to be a worthwhile addition to the lineup, AMD could always work on designing a tri-core die but for now it will fulfill its role as a stopgap solution.

Index A New CPU & Intel's Biggest Problem
POST A COMMENT

44 Comments

View All Comments

  • bgd73 - Monday, April 28, 2008 - link

    I read a few pages from a 1360 page book about computer repairing, in the history section. It was big nm back then, big power going wild... it states 1mhz for 1mbyte transfer. No wonder I think my 2.8e from 2003 before all your multicore quibbling is still just as decent as modern times...with 2 cores not quite bragged about. They are simply organizing cpus more than ever and reducing the die size. Keeping performance of the first of dual cores is as far as it may go for years...until the mhz is increased. 2800 mhz as a width is as wide as it goes. Organizing it does bring performance, like defragging a drive. Furthermore, if software knows how to use it...other software running simultaneous is losing..just like an old "hack me cuz I am errored forever" single core. Single cores are done, clean up that room with at least 2 thread cpus, that is all I found to be with 2 or more threads..very strong stable,secure, won't blip to a light switching on in the same room on the same circuit. The rest is marketing, they have to say something don't they... Reply
  • gochichi - Saturday, April 26, 2008 - link

    These processors are all "good" but this performance mark is not the "holy grail", I'd like to see more performance over time, as I'm used to.

    I recently switched to Intel, and you know, I'm happy with their prodcuts. I think AMD needs to get moving, their product's weakness isn't good for the industry.

    Both Nvidia and Intel have no competition, their job is just to maximize their profits on old research and development rather than actually competing under pressure.

    Reply
  • hoelder - Friday, April 25, 2008 - link

    If AMD would create and express team on the Processor side, take the best die implementation they currently have, lock the doors and over and over 24/7 cast new dies until they have a mass producible Opteron 4 core or Phenom 4x with 4 GHz, then they are where the should be today. Because in the end it's the CPU clock. And with every smaller die add cores and cache, it is plausible. Intel can afford that of course, but has also a tail of people involved. With a smaller team you can create miracles and with good enthusiasm on the exec level, that works. Reply
  • haplo602 - Thursday, April 24, 2008 - link

    What about linux kernel compilation with j2/3/4/6/8 ??? I'd like to see that comparison ...

    Reply
  • MrMilli - Thursday, April 24, 2008 - link

    Your power usage chart is a bit deceiving.
    In the article you mention that Windows Media Encoder is actually hardcoded to only support powers of two number of cores. Still you use this for measuring the power on a Phenom X3. So basically the 3th core is just idling.
    I think that's the reason there is such a big gap between the X3 and X4.
    Reply
  • enjoycoke - Thursday, April 24, 2008 - link

    I think Intel won't be releasing their new platform until 3rd Quater because they have been having such a good run with their current platforms already and will be taking a bit of a breather against AMD and other rivals.
    They really need constant profits to keep their stock price in line and thats what matter most.

    Reply
  • Archibald - Thursday, April 24, 2008 - link

    It is appears, that if one ignores the 1-10% performance increase(s), the dual-core is plenty for a casual power user (i.e. non-gamer). After all, the multi-core Si-HW is here, but the SW arena is a chaotic battlefield:
    ....Justin Rattner, an Intel Senior Fellow, recently promoted to take over Intel R&D has been quoted as saying that the clock wars of the past two decades will be replaced with ?core wars? over the next few decades. ?Intel & Microsoft are working feverishly on developing ?Concurrent Programming Languages? to effectively take advantage of the concurrent processor architectures that represent the future of the industry. ?Multicore processors require concurrent software:?The Free Lunch Is Over? (for software developers)., for more see this: http://tinyurl.com/62986h">http://tinyurl.com/62986h.

    I tend to favor AMD's approach with 780/790 and Brisbane, although marketing of this combo might be a challenge, from an engineering point of view it may be a decent (quite usable) design.

    Comment: Is the UI design of this blog from the Stone Age?
    Reply
  • derek85 - Thursday, April 24, 2008 - link

    I think AMD just released the perfect CPU to go with their 780G platform for a HTPC:

    - Low cost
    - Lower power consumption
    - HT3 to boot graphics memory bandwidth for better performance
    - Multi-core horsepower for better encoding/decoding

    Phenom is much mightier than Athlon X2 when it comes to multimedia. Now there is just no more reason to choose a similarly priced K8 over this.
    Reply
  • ap90033 - Thursday, April 24, 2008 - link

    Wow "perfect"? Slower in gaming, check. Slower clock for clock than Intel check. Pehnom 9850 cost more than Q6600 check. lol Reply
  • derek85 - Thursday, April 24, 2008 - link

    If it's for a gaming PC I would agree ... but I think I said HTPC. Cheapset X3 is only $150, $50+ cheaper than a Q6600, and will do this job just fine with less heat and power consumption. Reply

Log in

Don't have an account? Sign up now