"Pentium 4" Dead; dual-core not equipped with HT

[Mephistopheles]

According to <A HREF="http://www.theinquirer.net/?article=19187" target="_new">an article</A> from the inquirer, the Pentium 4 name is dead.

They're going to change its name by the time the dual 2.8, 3.0 and 3.2Ghz processors are introduced. These are said to be netburst-based smithfields, but they will not have HT enabled. I think this is a little strange... as far as I know, HT doesn't quite add a lot of heat!

I wonder if this is not a netburst core? I think it would be great if the next main desktop processor NOT called P4 were something smarter than netburst. 3Ghz is a little too much for a dothan core, however. (if it's not, and it is indeed a dual dothan @ 3Ghz, well, then Intel has a true trump card left... but that is VERY unlikely, even considering that they still have 9-12 months to develop these dual-core processors...)

Besides, the X20, X30 and X40 processors will probably be less crappy than the 6xx series. The 6xx series will probably have EM64T support, but a crappy one at best... and there will most likely be no reason to get a 6xx. Maybe x?? will be more respectable...

 

[priyajeet]

Who needs virtual processors when we have two physical ones (i mean 2 cores). But if they implement HT in those, we get 4 processors overall. There werent many apps that took full advantage of HT when it was launched. Will there be any taking advantage of Dual cores by 2005 ? Heres a line from Anandtech

<b><i>"We expected Intel to launch Hyper Threading with killer applications and benchmarks that would truly show its necessity on the desktop, but we were rather surprised to see that the best we got two years ago were some scripts that simulated isolated situations. Our fears are that 2005 will hold a repeat of Intel's HT launch on the desktop; while no one is arguing that dual core won't have a future, we're wondering if it may come a bit too soon to actually do anything."</b></i>

:tongue: <A HREF="http://www.geocities.com/priyajeet/fing.jpg" target="_new"><i><font color=red>Very funny, Scotty.</font color=red><font color=blue> Now beam down my clothes.</font color=blue></i></A> :tongue:

 

[Mephistopheles]

Ah yes, I read that too and I found that to be a very appropriate concern.

But still, I find it interesting that Intel would ditch HT, which is a highly hyped term still nowadays, when going to the next generation of processors - the X?? whatevers... I wonder what reason they had to do that? I remember many people saying that intel dual core solutions would have HT... how come they suddenly don't? Plus, remember that Yonah (dual-core dothan) has already taped out some time ago...

Maybe they'll be smart and not go the netburst way...

 

[Mephistopheles]

Consider this...
Typical TDP for a 90nm P4 @ 3.2Ghz: <b>103W!</b>
(from <A HREF="http://users.erols.com/chare/elec.htm" target="_new">this link</A>)

Now would they really try to put two of those cores on one die? I mean, 200+W processors? No, they have stated that they won't ship 200W processors! They said it themselves.

Unless "smithfield" would actually be a dual-northwood shrink... or something else, yonah-like... too strange. X?? whatever processors will still be 90nm.... so what the hell?

<i>Edit: TDP for a 3.2Ghz running on LGA775 appears to be 84W, which is a more reasonable figure. However, that's still a good way to make a 150+W processor, so my argument still stands.</i><P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 10/20/04 03:28 PM.</EM></FONT></P>

 

[pitsi]

Maybe what I am about to say is really stupid but, are we really sure that they are talking about two cores @ 3GHz? Or maybe is it possible that they are talking about two cores @ 1.5GHz each, which makes a total of 3GHz?

 

[Zoron]

With the way marketing is, it wouldn't surprise me. However, we all know that 2 processors don't double your performance... so it would be very dubious to claim that 2 cores running at 1.5GHz = 3GHz performance.

<font color=red> If you design software that is fool-proof, only a fool will want to use it. </font color=red>

 

[Mephistopheles]

No, I doubt it. That would make them <i>completely and utterly</i> underperforming by that time. Just imagine two 1.5Ghz P4 cores against two 2.0Ghz A64 cores.

The 3.4 and 3.6Ghz P4 cores that are available NOW would kick a 2x1.5Ghz anytime. Doesn't make any sense.

Can't be. They might be stubborn, but they cannot possibly be that dumb!

 

[endyen]

About the HT and heat- when scotties first came out, they would throttle almost instantaniously if HT was enabled, but would be fine with HT turned off, so HT does create a bit of heat. This would be far more critical if X2.
As far as what it would take to outdo a dual core @ 1.5ghz, The 2.8c would put it to shame, even in HT enabled SMP progs.

 

[Spitfire_x86]

Are they going to ditch Pentium brandname?

------------
<A HREF="http://www.foood.net" target="_new">FOOOD's Icons</A>

<A HREF="http://geocities.com/spitfire_x86/myrig.html" target="_new">My Rig</A> & <A HREF="http://geocities.com/spitfire_x86/benchmark.html" target="_new">3DMark score</A>

 

[Mephistopheles]

Yes, exactly, they are. (it's actually been too long, I think)....

Maybe they'll come up with a more interesting name... Who knows?

 

[raretech]

Octanium. I better go out and trademark that really fast...

<i>Nemo me impune lacesset</i>

 

[endyen]

No good. O comes before P. What you mean is Quicktanium.

 

[Mephistopheles]

Weeell, if they don't match AMD by then, you might as well call it... valium.

 

[endyen]

I'll take some of that.

 

[Spitfire_x86]

"Quicktanium" sounds somewhat like "QuickTime". Apple may sue Intel.

------------
<A HREF="http://www.foood.net" target="_new">FOOOD's Icons</A>

<A HREF="http://geocities.com/spitfire_x86/myrig.html" target="_new">My Rig</A> & <A HREF="http://geocities.com/spitfire_x86/benchmark.html" target="_new">3DMark score</A>

 

[RichPLS]

PlexTaniumXX for the dual core
QuadTaniom?? mainstream
QuaintTaniom ss
<pre> u </pre><p>

_____________________________________________
<font color=red> 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 </font color=red>

 

[RichPLS]

flu shot...

 

[RichPLS]

PlexTaniumXX for the dual core

QuadTaniom?? mainstream
<pre> ` u </pre><p>QuaintTaniom s X for low end

_____________________________________________
<font color=red> 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 </font color=red>

 

[raretech]

The XX at the end of Plextanium inspired me...

Sextanium, comes in X for single core, XX for dual core, XXX for triple core(which will be nicknamed "hard core").

I look forward to calling dell and placing an order with the nice lady on the other end for a hard core sextanium.

<i>Nemo me impune lacesset</i>

 

[Spitfire_x86]

Plextanium is similar to Plextor

------------
<A HREF="http://www.foood.net" target="_new">FOOOD's Icons</A>

<A HREF="http://geocities.com/spitfire_x86/myrig.html" target="_new">My Rig</A> & <A HREF="http://geocities.com/spitfire_x86/benchmark.html" target="_new">3DMark score</A>

 

[endyen]

They will probably call it the Intelium. That shows as much foresight as thier chips have lately.

 

[Xeon]

If they change the Pentium name that means Intel has no need of X86 any longer either. The two have been attached at the hips for quite some time I think spring sometime of 1993.

Be nice to see IA64 come over to the desktop if this is true.

Xeon

<font color=red>Post created with being a dickhead in mind.</font color=red>
<font color=white>For all emotional and slanderous statements contact THG for all law suits.</font color=white>

 

[Mephistopheles]

The only possible way they could get acceptance for IA64 would be making an engineering trick. Like implanting a small IA64 execution core - and they're not too big, if you put a reasonable amount of cache... But I don't see that happening... Many would still deem the IA64 core as useless and buy the 32-bit extended to 64-bit core alone.

Who knows, maybe IA64 will take a giant leap in performance with Montecito... some people are expecting that... But even so, it would be hard on Itanium. They might as well rename Itanium as well. Oh, and there's also the very unusual socket compatibility issue... Which makes things that much unpredictable. Can you imagine a gargantuan ~2.2-2.5Ghz Montecito with 24MB of cache (1.7 <i>billion</i> transistors) on LGA775 with dual DDR2-800 by 2006? Although... I think it will take longer... But it almost makes no sense!! Special OSs would be required... hardware is very different...

I wonder how that setup would play Doom 3-based games?...

 

[Spitfire_x86]

Mircrosoft won't make IA64 OS again. IA64 WinXP has been available for 3 years, but how many apps got ported to IA64?

IA64 is dead. Intel's official declarataion is only matter of time.

------------
<A HREF="http://www.foood.net" target="_new">FOOOD's Icons</A>

<A HREF="http://geocities.com/spitfire_x86/myrig.html" target="_new">My Rig</A> & <A HREF="http://geocities.com/spitfire_x86/benchmark.html" target="_new">3DMark score</A>

 

[Mephistopheles]

And going to dothan: over 50% of all mobile parts sold by Intel are now dothan-based. Intel has lowered prices agressively (-35% or so) and launched the 2.1Ghz Dothan, which <i>still</i> only consumes 21W TDP! Gotta love these processors. Why don't we see them on the desktop?

Remember, x86-secret actually managed to OC one of the initial Dothan processors to 2.4 easily!! And the processor's maximum temperature did not exceed 30C! You don't even <i>need</i> cool'n'quiet with those processors! Come on, Intel... what are you waiting for? Netburst is dead in the water. Yonah, a dual core dothan actually running on 65nm, has already taped out a while ago, and we have not heard about anything netburst-based and dual-core having taped out. Isn't it conceivable that a yonah on 65nm could run at speeds of 3Ghz?... Maybe that's what x?? is actually about... yonah still doesn't have hyperthreading, Intel <b>is</b> pushing 65nm still for 2005.....

Why not? It would justify a new name. Dual-core netburst wouldn't deserve another name. If by end 2005 they can come up with a 3ghz yonah for desktops, that would be one hell of a change. A dual-core 3Ghz PM-like processor would kick some serious *ss.

<P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 10/21/04 01:17 PM.</EM></FONT></P>