Several years ago Intel discovered surprisingly enough that its NetBurst architecture was not very good for the mobile space.  As wonderful as the idea of battery powered space heaters was, Intel quickly discovered that to build the perfect mobile platform you had to start from scratch and design a CPU that was built for the mobile space.  By doing so Intel could make tradeoffs that it wouldn't normally make, performance for power reduction, many of which we diagrammed in our first Centrino articles.

Intel also discovered the power of the platform; by bundling a good CPU with a good chipset and wireless controller, three independent Intel products were transformed into a marketing powerhouse.  The Centrino brand simplified notebook purchasing and quickly became a mark associated with a notebook you wanted to buy.

It took AMD a bit longer to get on the bandwagon, putting marketing first and worrying about architecture last.  We had heard rumors of a mobile-specific AMD microarchitecture, but nothing ever surfaced until now.  AMD's design team out of Massachusetts worked on the project, and today we're finally able to tell you about it.  The processor is called Griffin, and the platform is called Puma, both are codenames; AMD will undoubtedly come up with a phenomenal name for the final product (sorry we couldn't resist).

When Intel started development on the first Centrino processor, Banias, it had to go back to the P6 for a starting point.  The Pentium 4's NetBurst architecture was hardly suitable and the design team was intimately familiar with the P6 core at the time.  The end product hardly resembled a P6 and if you look at what the architecture evolved into today, you would be hard pressed to say it was similar at all to a Pentium III. 

AMD didn't make the misstep of a Pentium 4, it made a solid evolutionary step to K8 from K7.  Griffin's execution core and underlying architecture is based on the current generation 65nm K8 design, not Barcelona/Phenom.  You can take everything you are looking forward to being in Phenom and throw it out the window, as AMD is starting from the same K8 core that launched in 2003.

By no means is it a bad starting point, but thankfully AMD did toss in some enhancements.  Griffin gets a new North Bridge, a new memory controller, a power optimized Hyper Transport 3 interface and a 1MB L2 cache per core.  Griffin will still be built on a 65nm process as AMD will have, at best, only begun its 45nm production by the time Griffin debuts. 

Right off the bat you see a disparity between AMD's approach and Intel's approach; while the K8 is arguably a better starting point for a mobile-specific architecture than the P6, the K8 was heavily designed for servers and scaled down.  But as we've seen, the K8 is quite power efficient, with 35W TDPs easy achievable for dual core versions, so the race isn't over before it has started.

Griffin will go into production at the end of this year, and AMD is targeting availability in the first half to middle of 2008.  Given the launch timeframe, much like Phenom, AMD won't be competing with today's Core 2 processors but rather tomorrow's Penryn based notebooks.  Penryn does have some mobile-specific power improvements that even Griffin does not, but the opposite is also true as you will soon see.  AMD quoted a maximum TDP of 35W for dual core Griffin CPUs.  AMD hopes that notebooks based on Griffin can offer beyond 5 hours of battery life, but do keep in mind that battery life will vary greatly based on OEM implementation.

Truly Independent Power Planes
Comments Locked

36 Comments

View All Comments

  • yyrkoon - Friday, May 18, 2007 - link

    I can not help but think 'WHY' a laptop *NEEDS* more than one core on the CPU. Anyone claiming that their laptop is comparrible to the performance of a good desktop, is only kidding themselves. From my point of view, a laptop is a tool, used to do whatever you can not do on the desktop, for whatever reason ( travelling, and away, from your desktop, etc ). Yes, I understand that multiple core CPU have been availble in laptops for some time now, but that does not really answer my question. The point I am getting at here, is that if the laptop CPU could be made aroudn a single core, there should be plenty of room for other enhancements, and a potential for a lower TDP.

    quote:

    Now, instead of executing writes as soon as they show up, writes are stored in a buffer and once the buffer reaches a preset threshold the controller bursts the writes sequentially. What this avoids is the costly read/write switch penalty, helping improve bandwidth efficiency and reduce latency."


    The statement above, to me resembles something along the lines of the difference between USB v2.0, 400Mbit firewire , and syncronous/asyncronous read/write capabilities, there *HAS* to be a performance hit here . . . Who knows, maybe I am wrong ?
  • Justin Case - Sunday, May 20, 2007 - link

    A second core does make a very significant difference in terms of system responsiveness, even if you rarely (or never) max out both CPUs. I started using dual-CPU systems about 8 years ago and never looked back.

    More and more people are using a laptop as their only PC, so that theory that laptops are just typewriters isn't true anymore. They used to be just typewriters because that was all they _could_ do.

    With proper power management, a dual-core CPU can consume as little as a single-core one, for the same amount of work, so the only issue becomes the price of the chip itself. And since dual-core CPUs are so cheap, these days, there's really no reason not to have one.
  • Eris23007 - Friday, May 18, 2007 - link

    quote:

    I can not help but think 'WHY' a laptop *NEEDS* more than one core on the CPU.


    My company issued me a laptop, complete with anti-virus, software firewall, and hard disk encryption preinstalled that I can't turn off. My system runs two antivirus scans a week on the whole hard drive (at about 4 hours each). Given that an antivirus scan requires considerable hard disk access, that means that in addition to the processor load for running the virus-scanning algorithms, there's also substantial load for decryption.

    Unfortunately this laptop is a single-core Pentium M - no Core 2 Duo for me. My friend who hired on a little after I did got the Core 2 Duo version. The difference is astounding. My laptop (on XP, of course) has significant usability constraints - even the UI just isn't very responsive because of all the crap running in the background.

    I would *LOVE* to have an extra core around, dedicated to handling this bullcrap. It would make a tremendous difference in my ability to actually "get work done" when my system would otherwise be busy running all this junk.

    I'm aware that the extra core wouldn't improve the hard disk situation - but it would DEFINITELY help a WHOLE lot.
  • ADDAvenger - Friday, May 18, 2007 - link

    There's more people than you'd think who have a laptop as their only computer, and I am one of them. Yeah a laptop doesn't NEED dualcore, but believe me, it does make more of a difference than you want to give it credit for. I'm in college and take notes on this thing all day, a desktop just wouldn't work for me, and I just don't have the money to have both right now.
  • Wolfpup - Friday, May 18, 2007 - link

    Totally agree. I can't wait for quad core laptops. And yes, I intend to use my laptop as my only computer.
  • sprockkets - Friday, May 18, 2007 - link

    because it was demonstrated that adding another core did not make any difference in battery life while increasing performance.
  • JackPack - Friday, May 18, 2007 - link

    quote:

    The point I am getting at here, is that if the laptop CPU could be made aroudn a single core, there should be plenty of room for other enhancements, and a potential for a lower TDP.


    The new Intel Merom M0 stepping processors already supports something similar. With single threaded apps, one core goes down to C3 while the other speeds up to a turbo frequency (bin + 1) while staying below or at the specified TDP.

  • goku - Friday, May 18, 2007 - link

    I agree with this as well. People really expect laptops to be something that can do everything and it really is unreasonable. That said, AMD and Intel are making their computers more powerful so that people can start using laptops exclusively as that is the market trend, to go away from desktop machines.

    I'm wondering if AMD supports completely turning off a core and having it run in single core operation as I'd think running in single core mode with a slightly higher clock speed (if at all) would be more power efficient than having two slower cores running simultaneously. No need to have two cores running when you're just typing in MS word. It'd also be nice if intel and AMD would support far slower clocks speeds as IIRC the minimum clock speed increased from 600mhz on the Dothan and Banias machines to 800mhz on newer machines.

    If I could get my laptop to downclock to 100mhz when I'm typing in Word and reduce the voltage considerably, I bet you'd see a much higher jump in battery life. But with vista being so bloated today, the computer would probably never downclock to a speed as low as that since XP and vista seemingly are always busy.
  • CrystalBay - Friday, May 18, 2007 - link

    Kinda sounds like Barcelona, is not the panacea people were hoping for...
  • Hulk - Friday, May 18, 2007 - link

    Anand pretty clearly stated that Griffen is not based on Barcelona but on K8.

    I'm not really getting how this strategy will help AMD in the mobile market though. The current K8 core is getting pasted by C2D today clock-for-clock. And from what I just read I don't think the improvements made to create Griffen will enable to catch up to C2D IPC-wise.

    Add to that the fact that by the Griffen release time Penryn will be widely available at 45nm. More power efficient than current mobile C2D offerings, faster, and with better IPC.

    Like I said, I don't get it?

Log in

Don't have an account? Sign up now