Derek and I both worked on the Pentium 4 6xx article that went up today, so you know it's just got that extra bit of love that you all come here for :)
Overall I was relatively disappointed with the Pentium 4 600 series (and definitely in the new Extreme Edition processor). A larger, higher latency L2 cache is not the way to go - especially when looking at microprocessors from a performance per transistor added perspective. I'll be talking a lot about performance per transistor as a metric in my Cell article, which got put on hold while I worked on the 600 series article, but I will resume my focus on that today.
Looking at the amount of additional die area that the extra 1MB of cache consumes and compare that to the less than 10% of the die that AMD's on-die memory controller occupies and it's quickly obvious which offers the greater performance benefit per transistor.
It's very clear that, internally, Intel is facing a struggle between fighting for performance and fighting for dollars. Because while each revision or clock bump to Prescott does nothing to outclass or outperform AMD, Intel continues to do extremely well business wise. Convincing management to look at the AMD threat more seriously is difficult when what you're doing is making money, a lot more than AMD is making. I don't get the impression that Intel is going to be any more attractive even by the end of this year, although it does look like they will be first to dual core desktops, which is a sign of taking the AMD threat a little more seriously.
Overall I was relatively disappointed with the Pentium 4 600 series (and definitely in the new Extreme Edition processor). A larger, higher latency L2 cache is not the way to go - especially when looking at microprocessors from a performance per transistor added perspective. I'll be talking a lot about performance per transistor as a metric in my Cell article, which got put on hold while I worked on the 600 series article, but I will resume my focus on that today.
Looking at the amount of additional die area that the extra 1MB of cache consumes and compare that to the less than 10% of the die that AMD's on-die memory controller occupies and it's quickly obvious which offers the greater performance benefit per transistor.
It's very clear that, internally, Intel is facing a struggle between fighting for performance and fighting for dollars. Because while each revision or clock bump to Prescott does nothing to outclass or outperform AMD, Intel continues to do extremely well business wise. Convincing management to look at the AMD threat more seriously is difficult when what you're doing is making money, a lot more than AMD is making. I don't get the impression that Intel is going to be any more attractive even by the end of this year, although it does look like they will be first to dual core desktops, which is a sign of taking the AMD threat a little more seriously.
Facebook
Twitter
RSS
22 Comments
View All Comments
Andy Bellenie - Tuesday, February 22, 2005 - link
#7 - Oh, I understand the technical reasons, I was questioning along the lines of "why can't they make something better??" They have such huge R&D resources I find it difficult to understand how significantly smaller chip companies can keep beating them.Rick - Tuesday, February 22, 2005 - link
I think Intels problems run a bit deeper then everyone thinks. The same can be said about AMD. The leakage power of the smaller manufactoring processes is much more a problem now then it was previously. Check that out:http://www.anandtech.com/cpuchipsets/showdoc.aspx?...
This problem will not go away with the introduction of dual core processors. A bigger problem for the multicore movement might be that there are not many user apps that benifit from multiple threads.
I am more interested in what the Cell processor will and will not be able to do. I also head that IBM and Sony are writting an operating system...for Cell perhaps. We'll have to wait and see, but it would be pretty cool if Cell made its way into the PC.
Mephisto - Tuesday, February 22, 2005 - link
#9 Jonny: "that is the whole system, not just the CPU"Well, that looks a lot better - thanks for the correction.
Jonny - Tuesday, February 22, 2005 - link
#8, that is the whole system, not just the CPU. So, the whole P4 system uses 233 watts. Doesn't really change the whole outlook, but just wanted to point it out.Mephisto - Tuesday, February 22, 2005 - link
So one Pentium 4 3.8GHz EE CPU under load uses 233watts (http://www.anandtech.com/cpuchipsets/showdoc.aspx?...One Mac mini under load uses 20watts (http://www.tomshardware.com/howto/20050216/apple-m...
So you can run 11 Mac minis on the power necessary to run the latest Pentium 4 CPU
Add the rest of the PC and you'll be nearing 500watts
or 25 Mac minis running Xgrid
Seems like distributed power efficient CPUs is the way to go. Of course, you can't fit the I/O gear you need to the mini (no optical) - but it's the principle I'm pointing out.
I can see Intel's next slogan:
"Pentium - doing our bit for global warming"
or, perhaps,
"Pentium - hotter than a Sun"
viditor - Tuesday, February 22, 2005 - link
Andy - "I still can't fathom why Intel is having such trouble getting more performance out of it's chips"You should read Johan's article here on AT. It will give you a better idea why Prescott has failed...
http://www.anandtech.com/cpuchipsets/showdoc.aspx?...
radeonguy - Tuesday, February 22, 2005 - link
#1 Even if Intel hurries up and releases a dual core cpu the way intel has been making proceesors latley its going to suckMichael2k - Tuesday, February 22, 2005 - link
#1: Maybe because, like Anand implies, they don't need to?bob - Monday, February 21, 2005 - link
"aging Anand" ??He isn't that old, is he?
Live - Monday, February 21, 2005 - link
"I don't get the impression that Intel is going to be any more attractive even by the end of this year"Unfortunately it looks like you are right aging Anand