Thursday, May 18, 2006

I still hate Lucy

It is not clear from my latest post that I still despise Lucy.

Lucy is still headed for bankruptcy, that is a company that doesn't care about the customer. That doesn't fly once the customers find out that they are unappreciated.

And I am not more optimistic about AMD because Lucy, the bottom of the barrel, decided to smudge AMD's shinning reputation by finally giving in to the pressure and associating with it.

As a matter of fact, I consider far more important than the Dell news, today's opinion of the Illustrious Rahul Sood from Voodoo PC who said "In the meantime I can't talk too much about Conroe other than to say I am very impressed with the performance - it's $%*#ing fast man, seriously" because it made me change my opinion on Core2: First, since I still don't understand how Intel could reduce so dramatically power consumption on Woodcrest (see my previous posts where I detail why the features didn't seem to me sufficient to really explain Core2 claimed power efficiency, summary below), "The Inquirer" news about the subject were received with skepticism by me; but now R. Sood adding to the Core2 chorus really weakened my convictions.

I fear that tomorrow we will lose much of that 15% we gained today, just as what happened when AMD reported fantastic earnings: Everybody was waiting for a good earnings report as an exit point for this roller coaster, perhaps now a lot of people will use these news as an exit point too. I am not sure, though, these news are very disrruptive, the boost of market confidence is very significant. I think that the Turion X2 and the power efficient processors are compensation for my improved understanding of the real threat that Core2 is, thus Dell may be seen as a net positive.

Bunnysuit IconIt is important that you know that I didn't think this deal possible, because I think that Dell and Intel have a obscure secret about their relationship that prevents the former to be free and deal with AMD as anyone else; thus I was very wrong, and I sort of lost the bet, so I deserve a "bunnysuit" for this post.

Again, I ask you, members of my honorable audience to not forget to leave a name or callsign in the comments, a simple "I am "whoever", blablabla" would do perfectly.

Summary of why Core2 power inefficient: They are larger, even at 65nm, than Pentium Ms, they have more than twice the transistors, which switch faster, and to top all that off, they have much more execution units that can only be used marginally, with less power efficiency. Intel claims that they did micromanagement of power in the new arquitecture, but after looking at the material readily available, I came to the conclusion that those features can not be enough; so I am really curious to know why or how Core2s could be as power efficient as touted.


Anonymous said...

i believe Woodcrest's TDP announced by Intel is merely a smoke screen to confuse regular Joes.

Intel's definition for TDP is actually the power dissipation under normal circumstances.
stated on the site, intel is "relying on the fact that most applications barely use the CPU, assuming that it will remain idle most of the time." as a result, intel's TDP is NOT the actual TDP in reality.
taking Prescott for example, its TDP advertised is 89W, when under load is 111w.

on the other hand, AMD's definition for TDP is the "maximum current the CPU can draw", using default voltage, and under the worst case temperature conditions. this means the advertised TDP is significantly higher than TDP in reality.