Final Words

While we're still comparing to Socket-939 and only using RD480, it does seem very unlikely that AMD would be able to make up this much of a deficit with Socket-AM2 and RD580. With Conroe's performance advantage averaging over 20% it looks like Intel's confidence has been well placed.

Also keep in mind that we are over six months away from the actual launch of Conroe, performance can go up from where it is today. We also only looked at the 2.66GHz part, the Extreme Edition version of Conroe will most likely be clocked around 3.0GHz which will extend the performance advantage even further.

AMD still does have some time to surprise us with AM2, but from what we've seen today, they are going to have to do a lot of work to close this gap. We saw performance today in the two areas that we were most concerned about with Conroe: gaming and media encoding, and in both Intel greatly exceeded our expectations. Also remember that Conroe should be lower power than the AMD offering we compared it to, although we weren't able to measure power consumption at the wall in our brief time with the systems.

Going into IDF we expected to see a good showing from Conroe, but leaving IDF, well, now we just can't wait to have it.

More from the show as we get it...

Media Encoding Performance
Comments Locked

220 Comments

View All Comments

  • IHYLN - Tuesday, March 7, 2006 - link

    I stopped reading when I saw that the FX-60 rig couldn't even detect the FX-60. Like what a previous poster brought up something is definitely fishy. FEAR at that resolution is GPU limited so I can't understand how the cpu ALONE could make such a huge difference. This simply doesn't make sense. Obviously Anand doesn't know a CPU from a stick in the mud. I wouldn't even have published this nonsense after seeing that.
  • brownba - Tuesday, March 7, 2006 - link

    could it be it wasn't recognized since it was overclocked?
  • IHYLN - Tuesday, March 7, 2006 - link

    since when do you overclock a CPU and the motherboard doesn't recognize what processor class it is afterwards? Right...
  • brownba - Tuesday, March 7, 2006 - link

    all the screenshot says is 'model unknown'...
    i'm pretty sure that's what my athlon xp says with a small overclock,
    since it's not running at an official model clock anymore, it doesn't know what to report.
  • IHYLN - Tuesday, March 7, 2006 - link

    I've had 2 s939 boards, a K8N NEO2 Platinum and a DFI UT Lanparty SLI-DR. Both ran my FX-55 which is overclocked to 2.8ghz and both recognize what CPU class it is regardless if it runs at stock or overclocked. I'm not buying that excuse and neither is anyone else.
  • JarredWalton - Tuesday, March 7, 2006 - link

    I've got several boards (including the DFI Lanparty RDX200) that report "AMD Athlon Model Unknown" on any overclocked CPU. I mean, go ahead with the conspiracies if you really want to, but what point would it really serve to blatantly lie (either by us, Intel, both, etc.)? If we do it, we lose credibility. If Intel does it, we'll have the truth before launch anyway.

    Occam's Razor: Intel is damn happy with their performance, and they want to brag about it. I would if I were in their position. We're 4-6 months from launch, so there's not a whole lot AMD or anyone can do to immediately counter.

    As I said earlier, who seriously thinks that Intel set out to match or come close to Athlon 64/X2 performance when they started architecting Core? [Intel: "You know, AMD is currently faster than us. Let's throw a lot of money at the problem and see if we can close the gap a bit...."]
  • clnee55 - Tuesday, March 7, 2006 - link

    AMD fan boys would think so.
  • Questar - Tuesday, March 7, 2006 - link

    You don't need to buy any excude. You disqaulified yourself from having any meaningful input when you said FEAR is GPU limited at 12x10 on a x1900 crossfire setup.
  • IHYLN - Tuesday, March 7, 2006 - link

    my mistake, I missed the crossfire note.
  • photoguy99 - Tuesday, March 7, 2006 - link

    These results (if validated) are simply amazing on so many levels - forget the fact that this processor is not released for months - It will take AMD *years* to catch up.

    - The sheer amount of percentage increase has rarely been this large over the previous records with the introduction of any CPU (new architecture or not). In fact how many times has there been a jump this large in the last 10 years?

    - They did it at a lower clock speed. Devastating because AMD just doesn't have the headroom to out scale them on frequency.

    - They did it at lower power consumption. Again, very difficult to counter.

    - AM2 will probably not even provide a 10% boost. The tabloids are reporting initial samples were actually *slower* than socket 939.

    - Even if AMD has a new architecture out in 6 months to a year, it will be hard gain back that much performance per clock.

    - This spanking only required a 2.66Mhz CPU! And my god the Conroe EE will run at 3Ghz? Good luck competing with that.

    - Price to manufacture has always been in Intel's favor because they are ahead on the process curve. So not only will they lead in performance, they can cut the profit legs totally out from under AMD because Conroe is cheap to manufacture on 65nm, and it will only get worse with 45nm.

    Not only am I not an Intel fan boi, I'm writing this on an FX-60 clocked at 3Ghz which I paid a lot of money for thinking it would be one of the fastest systems around for a year. Looks like I won't even be close.

Log in

Don't have an account? Sign up now