The end of Pentium 4 and netburst

P4Man

Distinguished
Feb 6, 2004
2,305
0
19,780
Seems like the P4 "netburst" architecture is walking on its last legs:
<A HREF="http://www.theinquirer.net/?article=14257" target="_new">http://www.theinquirer.net/?article=14257</A>

Sounds to me like Prescotts problems are more than just a temporarily glitch, and Intel is ditching the P4 architecture in favour of a Dothan-esque cpu. Seems like the sensible thing to do too, too bad its going to take so long. The writing was on the wall years ago when Banias launched.

I fear all those expecting great things from Prescott/Tejas may be in for a disspointment, and the real intel cpu to look forward to is only coming in 2006.

I also wonder how intel is going to market lower clocking (but faster) desktop cpu's ... after changing their minds on extending x86, maybe they will change their minds on PR rating as well ?

= The views stated herein are my personal views, and not necessarily the views of my wife. =<P ID="edit"><FONT SIZE=-1><EM>Edited by P4Man on 02/20/04 04:21 AM.</EM></FONT></P>
 

Kanavit

Distinguished
Jan 6, 2004
390
0
18,780
Dothan is a good core. The banias(Centrino) 1.6ghz is equal to a P4 3.3ghz and Barton 10x220 running SETI@home. Should be an awesome gaming chip.

-------
<i><b>512MB-P4 2.8B-RADEON 9500PRO</i></b>
<A HREF="http://arc.aquamark3.com/arc/arc_view.php?run=1148239139" target="_new">29,473 Aquamarks</A>#1 in class!
 

P4Man

Distinguished
Feb 6, 2004
2,305
0
19,780
I’m all for using Banias/Dothan as a platform for future chips. The performance per watt and per mm² is way ahead of any netburst core, or even K7/8 core. It just makes sense.

But Banias is nowhere near a 3.3 GHz P4 or comparable Athlon overall. 2.4 Ghz might be more like it.

Anyway, the only caveat is that Banias might be difficult to scale in clockspeed.. After all, its based on a 9 year old design (Pentium Pro) and so far, it seems Dothan is having a hard time outclocking Banias by a wide amount.

I don’t have any inside information, so my guess is as good as anyone’s, but while Pentium-M is a very decent performer today, it will be less than stellar in more than 2 years if they can’t crank up the clock.

Or.. as I speculated elsewhere, maybe intel will abandon the clock race, and go wide instead. Think about how well a multicore Dothan could do. The core is so small, it shouldn’t be hard implementing dual core on 90nm and quad core on 65 even for mainstream chips.

Hyper threading might just have been an attempt to get software developers to start writing threaded apps, so that future multi core chips will profit, even at low clockspeeds and limited performance per core.. I think we are beginning to see the end of an episode, and the future will not be so much more about “faster” (clock) but “more” (cache, cores, etc).



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

FUGGER

Distinguished
Dec 31, 2007
2,490
0
19,780
Next you will be telling us Intel will stop making processors because AMD is just too good.

<A HREF="http://ourworld.compuserve.com/homepages/ArcadeWin/DONKEY.HTM" target="_new">AMD lays off 4200 people in the US and 1600 overseas</A>

<b>"You haven't proven anything that once 64-bit support comes out, it will perform even better." -EDEN</b>
 

P4Man

Distinguished
Feb 6, 2004
2,305
0
19,780
>Next you will be telling us Intel will stop making
>processors because AMD is just too good.

Hmm, yeah that is basically what I said.. I guess

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

Coop

Distinguished
Nov 27, 2003
217
0
18,680
Finally, Fugger gots the message, by some intel lovers it takes more time...

P.S: Fugger do you get you`r Intel CPU`s for free from intel ?


Toms Hardware Site is a joke !
 

Mephistopheles

Distinguished
Feb 10, 2003
2,444
0
19,780
Yes... Even if Dothan doesn't scale much past 2Ghz anyway, it's supposed to have +15% performance due to architectural changes. If true, this processor will be one killer chip when it gets to 2.0Ghz (15x133) on 533Mhz FSB by the end of the year.

<i><font color=red>You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete</font color=red> - Buckminster Fuller </i>
 

Kanavit

Distinguished
Jan 6, 2004
390
0
18,780
I wanna see a Centrino at 3ghz.

-------
Intel Pentium 4 2800 MHz
ATI RADEON 9700
Microsoft Windows XP
<A HREF="http://arc.aquamark3.com/arc/arc_view.php?run=1148239139" target="_new">29,473 Aquamarks</A>#1 in class!
 

DeEvolution

Distinguished
Jan 22, 2004
113
0
18,680
I will believe it when I see it. In the mean time I will simply remind that The Inquirer (.net) is no more a testament to hard-hitting journalism than The Inquirer (.com). The word <i>sensationalism</i> springs to mind. :lol:

Besides, Intel's silence about the P4 could very simply be representative of their reluctance to embrace x86-64.


"Sad is the elephant upon the ice who went to put on his wooly coat only to realize that he left it in his other trunk." - DeEvolution
 

P4Man

Distinguished
Feb 6, 2004
2,305
0
19,780
> In the mean time I will simply remind that The Inquirer (.
>net) is no more a testament to hard-hitting journalism than
>The Inquirer (.com).

They are both the same site. Well, anyway its a different sort of journalism, with lots of rumours and speculations, its not a press-release printing site like ZD or something, but you have to credit them for having extremely good resources. Most, if not all of their major rumours have come true, wether it be on prescHott, intel 64 bit, Sun/opteron, the delays (prescott, dothan, A64,..).. I can't think of one example where they clearly dropped the ball really, especially Charlie (aka Groo). Only reporter I often doubt is this Fuad "Fud" Abazovic.

>Besides, Intel's silence about the P4 could very simply be
>representative of their reluctance to embrace x86-64.

Nah, its their long term silence; no more promises of 10 GHz chips, plenty of Centrino raving, talks about needing a paradigm shift, talks about heat/power being THE main concern for the future, talks about pushing Dothan into the desktop market, etc, etc.. Time will tell, but I firmly believe netburst doesnt have a future beyond ~5 GHz. And it will struggle to get there. After that, it will be multicore, large cache, relatively low clocked, dothanesque chips that rule (for both Intel and AMD).

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

FUGGER

Distinguished
Dec 31, 2007
2,490
0
19,780
Someone forgot to mention that this IDF was marketed heavy towards mobile. I did not go because I am not into mobile. Yes we all know the IA32e was being announced. No reason for me to go and listen to that in person...

Nothing is free, processors are purchsed through developer channel at a discount.

I do not work for Intel.

<b>"You haven't proven anything that once 64-bit support comes out, it will perform even better." -EDEN</b>