I've been traveling for the month of September. It started with a trip to Santa Clara, then San Fran, then Denver, then San Fran again and I just closed out the month today with a 24 hour trip to Toronto (I'll touch on why in a moment).

IDF was interesting; it was smaller than previous shows, and definitely more dead than I remember it being last year, that being said there were some highlights.

1) Nehalem. Even AMD will admit (quietly) that the Nehalem demo was impressive. Penryn hasn't even launched yet and Nehalem is already up and running OSes. I don't doubt that clocks are low and the chip is buggy, but we're about 12 months away from launch too. AMD is skeptical that Intel can pull off an integrated memory controller that quickly, if Nehalem does indeed launch in mid to end of 2008 then AMD will be in some trouble for a while.

2) Silverthorne. I love this chip; well, the idea of it at least. If you've read my frustrations with processors in CE devices, I do believe that AMD and Intel hold the answer to improving performance and interoperability in the consumer electronics space with their new low power, low cost x86 cores that we'll begin seeing soon. Silverthorne is just the beginning, the real excitement is...

3) Moorestown. The Intel mock-iPhone demo is exactly where I see Moorestown being used. It'll be a couple of years at least but I can stand to wait. I think the next 3 - 5 years will see some serious innovation in the mobile market, especially with regards to ultra mobile devices like the iPhone. The iPhone is quite impressive today, and in 12 - 18 months we'll see some real competition to it, as well as continual updates from Apple itself. Life should be good. Moorestown will be particularly interesting at 32nm, the initial 45nm offering may be a little too large/power hungry for the type of exotic applications I'm envisioning.

Those were my highlights from IDF, now on to Toronto. I was there visiting AMD to talk about you know what. There's not much I can say, which is unfortunate. I really wonder how many times Intel will have to allow the press to benchmark an unreleased processor under supervision before AMD will let us do the same.

I have a feeling Phenom may end up being a little more competitive than expected, especially because Penryn won't hit the mainstream until next year. Once dual-core Penryn reaches the market in early '08 then AMD will be in for a rough ride. I seriously doubt we'll be seeing an upset on the level of what Intel did with the Core 2 launch however; that processor and microarchitecture look like they'll remain on top for the near future.

I talked to AMD about triple core, which will be coming in the beginning of 2008 as quad-cores with one core disabled. It doesn't look like triple core will ever be its own design, so the product will last as long as it can. Either quad-core will get cheap enough to make triple core obsolete or triple core demand will be high enough to warrant it sticking around, but I figure it'll end up going away. Until then it should offer an interesting affordable alternative to dual-core, unfortunately we'll have to wait until next year to see it.
Comments Locked

4 Comments

View All Comments

  • Viditor - Monday, October 1, 2007 - link

    Firstly Anand, I agree that it's absolutely great to see your blog happening again, it's been one of my favourite parts of AT!

    On Barcelona, I wonder if you are planning either an update or another article on benchmarking it. One of my concerns is based on Kris Kubicki's http://www.dailytech.com/article.aspx?newsid=8804">blog...
    In it, he stated that the stepping that was reviewed was the B1 stepping and not the BA stepping listed in the http://www.amd.com/us-en/assets/content_type/white...">errata sheet.
    He also says that the BA stepping (the actual shipping stepping) is 5%+ faster across the board because of an errata fix.

    While I realise that this was completely out of your control, could you let us know if you plan on updating the review?
    Also, could you let us know which steppings you are reviewing in your write-ups?

    Many continued thanks for being our "voice" in hardware...
  • Regs - Monday, October 8, 2007 - link

    I would love to know too. I would also love to hear your best guess, Anand, why AMD is being so hush hush over this launch.

    My best guess is that their goal was to release at faster clock speeds, yet they couldn't. Are they waiting for the processors to ramp up? It almost seems like their Marketing team is crying over the fact that their only retail Barc right now is 1.9 Ghz.
  • AmberClad - Monday, October 1, 2007 - link

    I'm a little curious about those supervised benchmarks.

    Does an Intel or AMD company rep literally sit next to you and watch you like a hawk? Or is it more a case of you simply doing the benchmarks on their company premises (i.e., you're not permitted to take the unreleased chip back home to the AT labs)?

    Is it to keep the media from taking close-up pictures of the technology / finding out and then publishing details that might be useful to the competition? To prevent reviewers from running programs or using features that the company knows are not yet stable? etc.

    Anyways, thanks for providing us with behind the scenes details on IDF.
  • slashbinslashbash - Thursday, September 27, 2007 - link

    I'm so glad that you're blogging again. I went through withdrawal symptoms multiple times. I just noticed on the sidebar and I almost couldn't believe my eyes. Don't ever abandon us again pleeeeease? :D

Log in

Don't have an account? Sign up now