Performance Stagnation

[GavHall81]

Hey guys....just a quick question.

I bought a 2.4GHz P4 back in May 2003. At the time, I recall 3.0GHz being the fastest you could get.

That was almost 2 years ago. By the standards I was used to, I would have expected to see 6 or 7.0GHz chips by now.

What has happened?

Cheers
Gav

 

[RichPLS]

Well, looks like it will be about 2 more years before dual core speeds get past what they are now.

<pre><font color=red>°¤o,¸¸¸,o¤°`°¤o \\// o¤°`°¤o,¸¸¸,o¤°
And the sign says "You got to have a membership card to get inside" Huh
So I got me a pen and paper And I made up my own little sign</pre><p></font color=red>

 

[mrmonsoon]

The issue is Intel has had increasing problems with voltage leakage on their tranistors as the are made smaller and asked to operate faster-ie they make too much heat( Amd is reaching this issue too, but more slowly, as their processors run slower.

This can be most clearly seen by intel stopping @3.8Gig and moving to dual core processing(Amd is also moving to dual core processors).

It will take new materials and designs to get the spped going up again and that costs money....

 

[slvr_phoenix]

Actually, what I think happened was that the speed bubble burst. (You know, like the internet bubble.) Things just got way out of hand between Intel and AMD in their race to beat each other. It was getting psycho. Then the speed bubble burst. They started slowing down and now are almost crawling ... that is, crawling at increasing their speeds at the rate that they <i>used</i> to increase before the race started. Or in other words, they've finally settled back down to <i>normal</i>.

<pre> <font color=purple>I express to you a hex value 84 with my ten binary 'digits'. </font color=purple></pre><p>@ 185K -> 200,000 miles or bust!

 

[P4Man]

This topic comes up frequently.. so a while ago, I wasted some time compiling some data from sandpile, and came up with this graph:
<A HREF="http://users.pandora.be/compas/freq.gif" target="_new"> Clock frequency over time </A>

I used Intel cpu's only for obvious reasons. The yellow dotted line is an exponential regression (which becomes linear, since its a logarithmical vertical scale)

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[Cybercraig]

What has happened?

Very little, to be honest.


Abit IS7 - 3.0C @ 3.6ghz - Mushkin PC4000 (2 X 512) - Sapphire 9800Pro - TT 420 watt Pure Power
Samsung 120gb ATA-100 - Maxtor 40gb ATA - 100
Sony DRU-510A - Yellowtail Merlot

 

[Spitfire_x86]

Nice Graph!

------------
<font color=orange><b><A HREF="http://www.mozilla.org/products/firefox" target="_new">Rediscover the web</A></b></font color=orange>

 

[slvr_phoenix]

Cool chart. It looks to me that if you take the pre-2000 rate and compare it to the post-2003 rate, you're looking at about the same rate. :O

<pre> <font color=purple>I express to you a hex value 84 with my ten binary 'digits'. </font color=purple></pre><p>@ 185K -> 200,000 miles or bust!

 

[P4Man]

>Cool chart. It looks to me that if you take the pre-2000 rate
> and compare it to the post-2003 rate, you're looking at
>about the same rate. :O

Its looks to me, that the "inflated" frequencies of the hyper pipelined netburst achitecture are the only real anomaly in this graph, and this effect is temporarely as it is hitting a thermal brick wall.

If someone has release data on earlier cpu's (from 286 and up), I'd be happy to include those. Maybe I should also try with AMD chips to eliminate the netburst effect, but I spent so much time reading your anti INQ posts, it will have to wait

= The views stated herein are my personal views, and not necessarily the views of my wife. =