Final Words

It took dual-core chips falling below $200 to start increasing their prevalence in the market, and today only one of our standard CPU tests won't see a performance increase from a dual core chip. I believe we're at the beginning of that same transition for quad-core CPUs. Many of our tests show a benefit from having four cores over two but in the next two years that should change significantly. The advent of GPU computing and the impending release of Larrabee will both bring about more focus on multi-threaded development. In the coming years a new group of applications that can run on both GPUs and multi-core CPUs will cement the transistion from applications that struggle to stress more than two cores to applications that scale to a virutally infinite number of cores.

The sheer affordability of quad-core processors today is impressive; $180 - $190 will buy you a Core 2 Quad Q8400 (2.66GHz/4MB L2), a Core 2 Duo E8500 (3.16GHz/6MB L2) or a Phenom II X4 940 (3.0GHz). Whether you go dual or quad is really a personal choice depending on the types of apps you run. If you look at our SYSMark 2007 results you’ll see that the E8500 is a better choice overall. Personally I’d opt for the quad core but that’s because when I’m most performance constrained it’s in applications that scale well to four cores, but if you don't do any 3D rendering or video encoding (or heavy multitasking between two multithreaded apps) then a fast dual-core may make the most sense for you today. If you're buying for a system that you plan on keeping for 3 - 5 years however, I suspect that quad-core is the way to go.

Between the Q8400 and the Phenom II X4 940, at stock clock speeds, the 940 is the way to go unless you're very concerned about power consumption or happen to be running applications that are very well optimized for Intel's Core architecture. Update #2: Intel has just confirmed that the Core 2 Quad Q8400 does support Intel's VT-x from the start, so the update below is incorrect. The Q8300, E5400, E5300, E7500 and E7400 will also end up transitioning to versions with VT-x support as well but only the Q8400 supports it from launch. Update: As many readers have pointed out, the Q8400 does not support Intel's VT for hardware accelerated virtualization. Honestly it's silly that Intel is attempting to use VT as a profit driver at this point. Not supporting VT on any quad-core CPU just doesn't make sense. The Phenom does support AMD's hardware virtualization AMD-V, and thus gives it a tremendous leg up if you care about the feature.

If you plan on doing some light overclocking, the Q8400 has more inherent potential. Start bumping up core voltages and the Phenom II X4 940 regains strength as it's able to increase the clock speed advantage once more. Throw overclocking into the mix and the comparison isn't quite as clean cut, both AMD and Intel trade blows in their advantages. I'd say AMD would probably have more wins in our applications but at the expense of much greater power consumption.

It's good to see that there's competition here, but Intel's profit margin advantage on the Q8400 is ridiculous. AMD has to sell something Nehalem sized for under $200 to remain relevant today. I'm far less concerned about who pulls ahead while overclocked and far more concerned about AMD's health at the end of all of this. Maybe the right way of looking at this isn't by talking about a 6% performance advantage, but instead talking about whether or not you want there to be a real competitor to Intel in the future. Maybe the Phenom II X4 940 should get the win here just to ensure we have an AMD to talk about in a couple of years...

Overclocking with a 10% Increase in Core Voltage
Comments Locked

60 Comments

View All Comments

  • Scali - Friday, May 8, 2009 - link

    I was already considering XP mode in my earlier post.
    I don't think that will be a big deal. Most people won't need it. Besides, you can still stick to XP or 32-bit Vista (not sure if 16-bit apps work in 32-bit Windows 7, else even that would be an option) if you have some applications that won't work in Win7. That way you don't need hardware virtualization either.

    I don't think hypervisors are going to be a big deal for the majority of people, since they simply don't need any kind of virtualization to begin with. It may be interesting for servers and such, but not for your average office machine, of which there are FAR more in the world.
    And for home users it won't be an issue anyway, because they generally won't even run a version of Windows 7 that supports XP Mode in the first place.
  • leexgx - Friday, May 8, 2009 - link

    but with intel you have to Realy check what CPU you are getting to see if it does support VT as thay do not state fully if it has VT or not all the time (not from the Q8x00 model any way) where as you get sempron it has no AMD-v get an norm AMD64 it has AMD-v that simple

    should never be turnd off realy its only been turnd off in the first place all of there cpus support VT (q8x00 i think is the exception as none of them have the VT option)

  • Scali - Friday, May 8, 2009 - link

    I'd like to comment on the closing statement of the article.
    I don't like the idea of supporting a company that has failed to deliver the goods for a few years now. It seems like AMD has dug a hole so deep that it's impossible to get out of it now. Buying AMD products isn't going to bring competition back. People ARE buying AMD products because they have such low prices (recently AMD has taken back some marketshare from Intel), but because of the low prices AMD just doesnt' bring in enough money. This also means they won't have a good budget for developing more competitive products in the future.

    In my opinion it seems AMD has already lost. They made two mistakes:
    1) They completely underestimated Intel and didn't do enough to prepare for Intel's Core2 back when AMD was still performance leader with the K7/K8 and was on top of its game.
    2) Their acquisition of ATi was ill-informed. They paid way too much for that company, and they seem to have purchased it at the worst possible time.

    This has weakened AMD so badly that it seems inevitable that they go bankrupt... The question is just: when.
  • just4U - Saturday, May 9, 2009 - link

    I disagree with you on this. How was AMD able to "do enough" when they were the performance leader? It's not like they were making boat loads of money. Intel was. Didn't seem to matter that the P4 sucked in comparison they just kept on keeping on.

    That was due in part, to interesting business deals that Intel made to keep Amd down (or so the lawsuits say) but the juries still out on all that (just thought I'd note it)

    I don't really see AMD making mistakes at all right now. They've got a nice lineup of video cards, and Cpu's that are in price brackets where the bulk of hardware is being sold. It will be interesting to see if they can build some momentum from it and hopefully it's not to little to late!
  • Scali - Sunday, May 10, 2009 - link

    Well, if you're performance leader, you have to milk that position.
    For starters, AMD didn't do enough to market their processors. As you say, P4 sales just kept going. Probably for a large part because most people didn't even know about the Athlons anyway, or had no idea about how fast they were.

    Then you also have to make sure you bring in the money by selling the right products at the right prices. And then you have to invest your profits into better production facilities and future products.

    After the success of K8, the K10 is a pretty mediocre CPU. It's little more than two K8s joined together, with some L3-cache, and some minor architectural changes. Now either AMD just didn't put enough effort into designing a newer architecture, because they thought K8 was going to keep Intel off for years... or they just made the wrong decisions, screwed up (K9?), and then had to redo the design, wasting a lot of money, and coming up with a product that was much delayed, and far less spectacular than they would have hoped.
    Either way, AMD should have done better with the K8's successor. They didn't, and that's why they are where they are right now.
  • lef - Friday, May 8, 2009 - link

    i can overclock my 940 just by setting the multiplier at 17x to 3400 with stock voltage (asus m3a79-t) at 2 seconds and you could only do 16x 3200?. what's the matter with you guys? are you incompetent or you are just misleading people here?
  • Anand Lal Shimpi - Friday, May 8, 2009 - link

    Our Phenom II X4 940 was not stable enough to run through all of our tests at 3.4GHz without additional voltage. Remember that when overclocking your mileage may vary. We saw the same 3.2GHz limit without increasing the stock voltage on our other 940 when we first reviewed the chips back in January.

    It is possible that later productions are more overclockable and as we all know the Phenom II X4 955 does much better at stock voltages than any of the AM2+ parts we'd seen up to that point. It's really a question of whether or not AMD is focused on producing more 940s or will production eventually shift over to 945s entirely. If it's the latter, I don't expect stock voltage overclocking to improve much more on the 940. If it's the former, then the 940 should eventually be as good as the newer AM3 parts for overclocking.

    Take care,
    Anand
  • sluk - Friday, May 8, 2009 - link

    Developer of VirtualBox (Sun Microsystems) claimed that hardware accelerated virtualization in fact perform slower than software virtualization. I wonder if there is any 3rd party benchmark test can confirm this or if it is only true for Virtualbox. If this is the case, why you need to care about VT support in CPU, you can just run your XP virtual machine with any type of CPU in Windows 7.
  • stmok - Sunday, May 10, 2009 - link

    It isn't so straightforward.

    The first generation of VT support in CPUs really benefits the software developer. It makes it easier for them to implement a robust virtualization environment. It isn't primarily about performance. (The gain is very little).

    The 2nd generation with features like nested paging are. But that is only in Phenom I/II and Core i5/7 generations.

    It leads me to believe Microsoft was scrambling to address the WinXP compatibility issues with custom business software in Windows 7.

    Software using hardware virtualization is far easier to develop. (They have a deadline to meet, as well as customer expectations).
  • DeepThought86 - Thursday, May 7, 2009 - link

    Ummmm, how can we assume that wafer costs are the main costs associated with CPU production? Both AMD and Intel have already invested billions in fab construction, fab equipment, researcher salaries, marketing budgets, admin staff etc etc.

    To say that after all that what's killing AMD is the costs of the wafers is plain silly. Sure, it's a higher variable cost. But what proportion are variable costs compared to fixed costs? If AMD stopped selling Semprons and X2s and shifted their production to 100% Phenom II 2.5 GHzs, would their costs go up significantly compared to the losses they've been making?

Log in

Don't have an account? Sign up now