The Socket-AM2 piece that Wes put up today was what I was hoping would go live a couple weeks back, but performance issues made that impossible. So what we're left with is a quick look at the new socket and some of the mechanical changes that you'll see with Socket-AM2 boards, but nothing on performance yet.

Why are there no performance numbers in that review? Unfortunately it seems that the CPU is quite early and the motherboards aren't ready for prime time yet, so performance of AM2 is significantly lower than Socket-939 today. I do expect that to change, hopefully by the end of this month, but that's why you won't be able to see any realistic previews of AM2 performance this early. It is a bit disappointing, since I expected the move to be fairly seamless and I honestly was hoping for production-level performance by now.

I've been itching to get my hands on Conroe as well but I've had no luck as of yet. I know at least Merom has been floating around out there but I've yet to hear of anyone (who's willing to tell me at least) that they've had some hands on time with Conroe outside of Intel. I'm hoping that'll change in another month or so.

The iMac Core Duo experience was interesting and somewhat anti-climactic. On the one hand, I understand that Apple is bringing the first Intel based Macs a little earlier than publicly expected, but on the other hand I was hoping for something a bit more pronounced. As I alluded to in the article, I'm expecting the truly revamped platforms later this year as you don't just cut power by 1/3 and not try and move to a smaller chassis. The MacBook Pro is tempting, but Merom isn't too far away now :)

I've been actively working on this USB 2.0 power draw issue with notebooks for a while now, it's not as clean cut as it's been presented thus far and I'm trying to gather the data to show the complete story now. The problem with testing battery life on notebooks is that it takes around 3 - 5 hours to completely discharge one and then another 2.5 hours to bring it back to 100% before I can run another test. It's a lot more difficult to test theories when you have to wait 7 hours before you can run comparison data to prove/disprove your point. I guess this is the only situation where I can really complain about notebook battery life getting too long :)


View All Comments

  • MrKaz - Friday, February 10, 2006 - link

    You are all exited about AM2 but besides DDR2 and Virtualization there will be no substantial performance changes.

    The prime objective of socket AM2 was DDR2, because DDR2 is getting cheaper than DDR, so AMD couldn’t get behind Intel on the cost of the all platform because of the memory.

    Also I would be surprised if AMD releases processors for socket AM2 faster than the ones already existing for socket939, because AMD with that would give little reason for people to go after the “old” socket939 processors and boards. This is similar what it have happened with Sempron socket 754 32bit and Sempron socket A.
  • IntelUser2000 - Wednesday, February 15, 2006 - link

    Rumor around is AM2 is gonna be out June 6th, which is summer, not like spring as initially thought. Good luck getting the properly working test samples. Reply
  • Furen - Monday, February 6, 2006 - link

    It felt like the article was incomplete or part of a greater whole since it went way too in-depth to be a simple pin-out and heatsink mount overview... Hope you guys can get your problems solved as soon as possible so we can all see just how well AM2 performs. Reply
  • Quiksel - Monday, February 6, 2006 - link

    Do testbeds not come with an extra battery, even on request? That'd save a ton o time, would it not?

    To the Apple front, easily the most disheartening news for the Intel transition has got to be Adobe's lack of commitment to bring to the table some real Universal applications for the creative bunch (last I read, they are not re-releasing CS2, but say they will only release Universal builds at the next release)... I don't know for sure what percentage of Apple users are creative professionals, but this would keep many people from upgrading those G4's and G5's to the new Intel Core Duo products. Is not Apple dismayed by this?
  • Araemo - Wednesday, February 8, 2006 - link

    How many testbeds come with an external battery charger so you can charge a battery while you're running a test?

    Two batteries that can only be charged in the laptop doesn't solve the issue at all. VERY few laptops anymore have external battery chargers. :(
  • michael2k - Tuesday, February 7, 2006 - link

    Apple gets paid whether people buy a G5 or an Core Duo.

    And Apple/Adobe won't double their sales by releasing Universal Binaries this year instead of next year. If a professional needs their compute power now, they buy a G5; The price of a computer can easily be paid for in a month of work. The profit isn't there if Adobe sells a $49 upgrade disc with only UB support, and no one will pay $499 for an upgrade disc with only UB support, so it makes the most sense for Adobe to add UB support in a future release.

    Yes, it's disappointing to professionals who want to buy an Intel based PowerMac this year, but really if they are the target market of Adobe, they can afford to buy a quad G5 this year, and then a Core Quattro next year.
  • ProviaFan - Monday, February 6, 2006 - link

    The only problem with two batteries is that you don't know how much variation there is between two individual units, and testing that alone could account for all the time that you might save by having both of them in the first place. Also, you'd need a way to charge the other one while you test the first in the notebook. Oh well, I do wish that it were possible, and it may be, as my concerns are probably overrated. Reply

Log in

Don't have an account? Sign up now