Wednesday, May 03, 2006

AMD answers Intel's guerilla benchmarketing


Tired of Intel's hit and run guerilla benchmarketing, AMD setup billboards all over the world in an effort to draw Intel out of the woods for a hand-to-hand duel. The billboards (see picture by BusinessWeek) display the electricity cost wasted by Intel servers. Currently, the amount is at over $1 billion dollars.

Previously, I estimated that Google can save $140 million a year on power by going AMD.



According to this report, DELL has joined the Green Grid, an AMD led effort for environmentally sustainable computing. Incidentally, a former DELL exec added 25,000 AMD shares to his portfolio.

Recently, Intel became an EMC reseller, competing against DELL. Previously, AMD exempted Michael Dell from testifying against Intel. Earlier, Michael Dell made the following comments on Intel's Conroe architecture.

5 Comments:

Anonymous Anonymous said...

Power consumption matters, sure. But AMD only seems to think it matters when Intel processors put out a lot of heat. Back with the Athlon XP - AMD's processors were pumping out tons more heat than the Northwood P4s of the time. Did AMD go around telling everyone that a low power consumption matters then?

9:29 AM, May 03, 2006  
Blogger Sharikou, Ph. D. said...

In the old days, Athlon produced more heat and had higher performance. Today, Opteron has much higher performance with less heat. The performance/watt situation changed.

10:27 AM, May 03, 2006  
Anonymous Anonymous said...

To each their own advantage!

Back then, CPU heat didn't matter (that much) otherwise Intel would surely have used it against AMD. Times have changed, table has turned, Intel gets burned (pun inteded).

As for the former DELL exec... didn't know you could get paid $300K (in options) for one years work as a director (if he sold @ $33/share).

Concidering current situations, AMD would have an advantage by refuse DELL as a custumer (if DELL wanted in).

11:17 AM, May 03, 2006  
Anonymous Anonymous said...

I've got several AXp processors on hand, an AXp 2500+ (and yes, it is overclocked, a 2.2ghz raw speed), two AXp 2400+, AXp 1800+, Athlon4 1ghz, and older T-Bird 850.

I also happen to have a 2.6ghz P4 on hand. Now, I've downclocked the P4 to 2.2ghz, 2.0ghz, 1.8ghz, 1.6ghz, 1.4ghz, and through some voltage tweaks on the motherboard in use (using a via PT880 chipset), down to 1.1ghz. To standardize on a testing methology I used Zalmans 92mm cu/al fan/heatsink. At all direct speeds the Northwood P4 reports a /higher/ heatsink temperature, and the internal tempature controls also reported being several degrees higher.

Now, I tested the AMD chips across Nforce, Nforce 2, Via KT133a, VIA KT266a, Via KT333, Via K400A, Via KT880, SiS... I actually have no idea what's in the laptop actually, and AMD 761 chipsets.

So, yes, my own tests are biased by comparing a single Intel processor on a single motherboard with some voltage tweaks against a multitude of AMD motherboards with different AMD processors. I'll freely admit that maybe the testing methology was not fair or unbiased.

But given that a single Intel chip put off more heat than a direct equivelent clocked AMD processor, I'm not exactly sure where you get "tons" more heat from.

Keep in mind that many AMD processors typically came with cheap low-end Coolermaster fans, it was probably more a perceived problem of heat output than a realistic problem of heat output.

Let me explain:

Intel already had an issue with the heat generated by the Intel processors and prescibed a very strict thermal envelope on what OEMS could actually deliver to the market. AMD's processors which ran at slower speeds did not need the stringent thermal restrictions applied to Intel processors. The result is that OEM's keen to make a buck shaved off as much cost as they could from making the computer. So, while an Intel Processor required at least a decent heatsink, OEMs were free to use generic small fans on the AMD units.

The result was a percieved heat difference between AMD and Intel computers. Consumers who just wandered up to a computer and put their hand over the exhaust were toasted by the inadequetly cooled AMD computers, while the Intel systems seemed to put out much less heat.

So, keeping in mind perceptions, would you mind giving specifics of what processors you used and on what platforms?

11:20 AM, May 03, 2006  
Blogger Sharikou, Ph. D. said...

There are measurements of Intel and AMD power consumption, just google to find it.

Here is an example.

12:07 PM, May 03, 2006  

Post a Comment

<< Home