SI
SI
discoversearch

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 : Intel Corporation (INTC)
INTC 37.81-4.3%Dec 12 9:30 AM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: wanna_bmw who wrote (155914)1/17/2002 1:51:53 AM
From: Joe NYC  Read Replies (2) of 186894
 
wbmw,

If this were true, then scaling would be superlinear, and you just got finished saying that it's not. model numbers have been increasing by 100 for every 67MHz of frequency. That does *not* reflect true performance.

What I am saying (about the model numbers) is that when 2200+ is released, it will not be 10% faster than 2000+. It will be faster by approximately the same percentage as 2.2 GHz Northwood is vs. 2 GHz.

This is what makes me skeptical to marketing graphs that show Hammer getting >4x improvement over a 1GHz Athlon

This is my theory how this is done: Let's go back to a post about 1 GHz Athlon classic vs. 1.4 GHz XP.

Suppose you have a benchmark, and you get a score for 1 GHz Athlon and say 1.4 GHz XP. You also need the scores of 950, 800, 750 MHz Athlon to come up with a scaling graph. Now extend this graph up to see what speed of the classic Athlon processor it would take to match 1.4 GHz Athlon XP.

Suppose from the extrapolated performance numbers, you would get that it would take 1.6 GHz Athlon classic to equal 1.4 GHz Athlon XP.

Now, knowing that you need 60% increase in clock speed of Athlon classic to match 1.4 GHz Athlon XP, is it unreasonable to say that XP is 60% faster than the classic? In terms of marketing-speak, I don't think it is the worst transgression I have ever heard.

Joe
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext