Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: Ritz who wrote (247659)	2/12/2008 8:57:33 PM
From: wbmw	Respond to of 275872

Re: It really does not matter whether Phenom fails because of clockspeeds or IPC, the bottom line is that it is a miserable and total failure. You can't just chock it up to AMD's inferior 65nm process either. Brisbanes are up to 2.8GHz stock and best case 3.6-3.7 overclocked, while Phenoms are at 2.3 stock and 3.0-3.1 absolute max overclock. Basically the entire, years long, K10 development effort resulted in a 10-15% increase in IPC and a 15-20% decrease in clocks, resulting in equal or lower single thread performance. It isn't surprising to me that clock rates are lower for quad core parts, relative to duals. That just seems to be a reality of a monolithic chip design, since you are attempting to put 2x as many transistors - plus some rather significant crossbar logic - on the same die, without doubling the power envelope. It's no wonder that Intel stayed with the MCM approach until 45nm - at least to limit the growth to 2x and use the relatively simple FSB logic for routing between cores. However, I will agree that it's unacceptable to take this step backwards in single threaded performance. AMD must have figured that Intel would be limited by the same constraints, that their quad cores would be just as power hungry, and that consumers would be driven into greater threaded performance. Amazing, that with the benefits of an MCM approach, Intel was able to get the best of both worlds.... Re: Phenom is an utter and complete failure. It's a bit of an irony, actually. By itself, Phenom would have been an interesting upgrade for threaded users, because the threaded performance is quite good. But in the shadow of Intel, it does appear like an utter and complete failure. Just like Intel's first Pentium D appeared with respect to the first dual core Athlon. I remember the first Smithfield review to be quite good. But after AMD followed a week later with much faster parts, consumers very quickly switched attention to AMD. This time around, Intel led the way to quad core a year ago, and by the time AMD caught up, they were hopelessly slow. I suppose that is the means to a greater failure.