P4 will scale above 10GHz

Raystonn

Distinguished
Apr 12, 2001
2,273
0
19,780
"Over its life, it (the Pentium 4) will scale above 10GHz"

http://quickenexcite.cnet.com/news/0-1003-200-5742941.html?tag=pt.intuit.newsfeed..ne5742941?tag=

-Raystonn

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

Guest

Guest
That is remarkable even my evilness is wanting to be good to amd guys cause i pitty them soo much.

SPUDMUFFIN

<font color=blue>Just some advice from your friendly neighborhood blue man </font color=blue> :smile:
 

jlbigguy

Distinguished
Dec 31, 2007
1,001
0
19,280
Interesting. But the P4 will be a dead product once Northwood comes out. Are you saying the P4 will hit 10GHZ this year?

<font color=blue>This is a Forum, not a playground. Treat it with Respect.</font color=blue>
 

Fltsimbuff

Distinguished
Mar 30, 2001
114
0
18,680
Foster is still considered to be a P4... albeit a modified (improved) version... kinda like AMD's Athlon Classic vs. T-Bird. I only hope that the FPU is a lot more powerful.

--Fltsimbuff
 

Raystonn

Distinguished
Apr 12, 2001
2,273
0
19,780
P4 a dead product? Whatever are you talking about? When we say "P4" we're talking about the design of the CPU, not any specific model of it. We'll be using the P4 design for years to come.

-Raystonn

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

girish

Distinguished
Dec 31, 2007
2,885
0
20,780
how many zeros does 10 GHz have?
realistically, P4 wont go till 10G, but P5 or P6 will, surely...

core upgrades are ok, but basically p4 does not have the potential to reach 10G, i guess it would retire at around 3.5 - 4 GHz and give way to P5s, or may be Itanium and its decendents if 64 bits do take over.

look at intel history - a new processor every 3 years.

<font color=blue>die-hard fans don't have heat-sinks!</font color=blue>
 

Raystonn

Distinguished
Apr 12, 2001
2,273
0
19,780
The P4 is designed to scale beyond 10GHz, and that's what it will do. If you haven't, please read the URL I supplied. Given the fact that we will see 2GHz in July, and given Moore's law, a 10GHz P4 should be available in about 3.5 - 4.0 years.

-Raystonn

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

leonov

Distinguished
Dec 31, 2007
66
0
18,630
10GHz May be a lot of zeroes but you will find that it is not too far away as far as performance scaling goes. The performance roughly doubles every 18 months or so (faster now but let's take this as a guide).

We are at 1.7GHz now double it to 3.4GHz in about 18 months and again to 6.8GHz 18 months after that. Basically in 3 years we will reach ~7GHz. New CPUs come along about every 5 years or so 10GHz is a real possibility. In fact we may have to force the next CPU generation (much like P4 was forced) to get more GHz.

L
 
G

Guest

Guest
If AMD said that the Athlon could go 10 ghz we would here nothing from yer monkey mouths now would we? Keep it down its hard to be evil with monkey chatter all the time=)

SPUDMUFFIN

<font color=red>Being Evil Is Good. Cause I Can Be A Prick And Get Away With It.</font color=red> :lol:
 

jlbigguy

Distinguished
Dec 31, 2007
1,001
0
19,280
The P4 is a dead product, just as the Athlon classic is a dead product. Will Northwood require a new socket? New motherboard? Where is the "room to grow" (AMDMeltDown) for the current P4 user?

If the Northwood is an improved P4 (perhaps what the P4 should have been if Intel didn't rush to market) it will be significantly different then today's P4, and incompatible (hardware wise) as well.

Todays P4 sounds like a dead product to me.

<font color=blue>This is a Forum, not a playground. Treat it with Respect.</font color=blue>
 
G

Guest

Guest
AMD will achieve the same speeds as Intel.It's anyone's guess as to who will be faster to come out with the 10GHZ...
Isn't the latest trend showing AMD leading and coming out with each new cpu faster? (Except for the P4 1.7GHZ now that is...)
All I know is that there seems to be a lot more interest in P4s now that Intel is changing over to DDR-support from Rambus. Rambus is a dead place, that'll teach Intel for going against the entire memory market-or not...Seems to me they've made this mistake in the past as well.
 

SammyBoy

Distinguished
Mar 16, 2001
689
0
18,980
Okay, here's the thing. Sure, the P4 (Northwood, as the current incarnation of the P4 was DOA) might be able to scale up to 10GHz, but you have to remember, that current processing techniques don't allow for a die that small to be created. In order to do that, EUV has to be used, and while it was recently announced that the first prototype is built, it will be at least a few years before it is converted and used for fab plants. Yes, I understand that Moore's Law says this and Moore's Law says that, but you have to understand that it is really an average, and there will be times that it will go faster or slower than doubling every 18 months. In fact, Moore's Law probably would have been refuted if not for upstart AMD coming into the scene, providing much needed competition to Intel. Therefore, the P4 that hits 10GHz will most likely be as identical to the Northwood as a boy and a girl. Hey, they're the same species and can work together, but look, different parts! But seriously, it would, in my mind, be corporate suicide to try to stretch out a core for 5 years, like the Northwood would have to be in order to reach 10GHz. I mean, the P3 lasted, what, 2, maybe 3 years? To take the P4 out to 2006, when the makers of EUV expect the first 10GHz chip to be available, is kinda far-fetched. Plus, there might be some breakthrough in CPU logic by then that requires a reworking of CPU architecure.
The only reason why I can see Intel holding onto the P4 as long as that is that they wish to try to force people, mostly businesses, to build a machine that uses a dual mobo that has both a 32 and 64-bit processor. I mean, would you, as a consumer, really want to completely drop all your 32-bit apps, which work just fine, just so that you can run the one or two 64-bit app that you just got? I mean, it's another marketing move from Intel, if that is the case, and the P4 may have been designed specifically to run with a 64-bit processor no problem. Maybe that is why they ridicule AMD for making the Hammer, which has both 32 and 64-bit compatibility, they see it as a waste of money and a way to lose profit. So yeah, P4 may reach 10GHz, and that might be a plan, but it could be a conspiricy to squeeze a little more cash out of the consumers, or it just might be that they really expect the 32-bit apps to die off quickly after the first intro of 64-bit apps. But I really don't think that I would dump my 32-bit apps. I mean, I still use games and programs that were made to run with a Pentium 100... and I'm gonna run them on a 1.2 T-bird. Works for me, and I'm sure a lot of other people would not want to give up their old apps that they've grown to love and know.

Oh yeah, here's a link to EUV news:
<A HREF="http:// http://www.tomshardware.com/technews/technews-20010412.html " target="_new">http:// http://www.tomshardware.com/technews/technews-20010412.html </A>

-SammyBoy

Without Evil, there can be no Good. Therefore, without an Intel, there can be no AMD.
 

Grizely1

Splendid
Dec 31, 2007
7,810
0
30,780
We're not talking talking about the same P4 as is out right now. There will be core upgrades, etc. Not a very big deal if you ask me.

---------
I am the first and only one with a 16MB GeForce2 GTS graphics card! :smile:
 

girish

Distinguished
Dec 31, 2007
2,885
0
20,780
double speed in 18 months... true till now...

but how long? do you always pull the progress graph in a straight line? there has to saturation somewhere.
1.7GHz now
3.4GHz in about 18 months
again to 6.8GHz 18 months after that.
3 years we will reach ~7GHz
and in 4 and half years - 15 GHz
and in 6 years - 30 GHz
and in 9 years - 120 GHz

be realistic guys... silicon cannot work this fast
there are other alternatives like GaAs (Gallium Arsenide) that might go to that speed, but as current silicon tech stands i dont think it wil go much beyond 5~10 GHz, that too in such a short period.

there will be saturation and after some point Moore's Law will not hold true anymore

and we dont know just as yet where this point is...


<font color=blue>die-hard fans don't have heat-sinks!</font color=blue>
 
G

Guest

Guest
You guys are forgetting that the P6 core includes the Pentium Pro, Pentium II, Celeron, Celeron A, Pentium III and Celeron II's. Again all those cpu's use the same basic P6 core. That's alot of mileage from a core design. So when Intel says the P4 will scale up to 10 GHz during its lifetime they are really implying that the P7 CORE will reach those speeds when infact the actual cpu features could be very different from what the P4 is today. Hope this made sense to you guys.. if not.. oh well, i tried.
 
G

Guest

Guest
Sounds very plausible. The Pentium Pro archatecure (albeit in various packages, first with SSE bolted on, some with on-die cache and some without!) has lasted from 150MHz to 1000MHz, a factor of 6.67. So if we consider the first 'serious' P4 as a 2.0GHz Northwood, the same factor produces over 10GHz. Although things are obviously more complicated than this, it's still an indicator.

The 10GHz chips will require a new type of etching, however, and how long this takes to perfect is another question. Either than or we'll be stuck with huge dies and massive cooling, a prospect I do not relish!


~ The First Formally Rehabilitated AMD Lemming ~<P ID="edit"><FONT SIZE=-1><EM>Edited by Verteron on 04/27/01 07:05 PM.</EM></FONT></P>
 
G

Guest

Guest
<<and in 9 years - 120 GHz
be realistic guys... silicon cannot work this fast>>

Back during the Pentium 100 days Im sure people like you said the same thing about hitting 1.2GHz (hands claspped over there ears shouting "Impossible! Impossible!" over and over.)
It's a good thing 9 years is long ways away.. it'll give you time to forget you made such a foolish comment.
 

Grizely1

Splendid
Dec 31, 2007
7,810
0
30,780
Actually I agree. The physical limitations of a silicon chip weren't even being grazed upon with the Pentium 100. 120GHz, is, well, <b>A LOT FREAKIN' CLOSER.</b>

---------
I am the first and only one with a 16MB GeForce2 GTS graphics card! :smile:
 
G

Guest

Guest
Ahahaha the monkey men are mad they are chatting left and right like monkey's do on the discovery channel. Geeze calm down you AMD gimps need to know when youve been beaten and beaten bad. Aahahaha see you monkey's later.

SPUDMUFFIN

<font color=red>Being Evil Is Good. Cause I Can Be A Prick And Get Away With It.</font color=red> :lol:
 
G

Guest

Guest
Food is good. I really really like beans though "get in my belly".

SPUDMUFFIN

<font color=red>Being Evil Is Good. Cause I Can Be A Prick And Get Away With It.</font color=red> :lol:
 

SammyBoy

Distinguished
Mar 16, 2001
689
0
18,980
<<and in 9 years - 120 GHz
be realistic guys... silicon cannot work this fast>>

Back during the Pentium 100 days Im sure people like you said the same thing about hitting 1.2GHz (hands claspped over there ears shouting "Impossible! Impossible!" over and over.)
It's a good thing 9 years is long ways away.. it'll give you time to forget you made such a foolish comment.
Sadly, you have forgotten to take your own advice and read the link that I supplied with the comments on EUV, and you choose to flame those who make a realistic statement based on theory and fact. Embedded within all those links and articles are pharses that state, and I am paraphrasing, that silicon can only support etchings as small as 40 atoms across. Now while that is extremely small, I don't think that the 7th gen core will ever get that small (or an AMD silicon chip, for all you damn AMD haters), due to the fact that by that size, a flaw that might exist in a perfectly good CPU today would magnified millions of times, and probably cause some kind of instant CPU death. The quality control costs would be so extreme and the yeilds so low that the price of a silicon chip with 40 atom wide etching would be astronomical. Therefore, it is safe to assume that once EUV, or the other etching process thats in the pipe right now, begins to reach a stage of more waste than product, a different material has to be found. In fact, it is rumored that some experts think that the only thing better than silicon chips would be biochips, but hey, that's getting to the extreme part of technology. And my other comment, on how the P6 core lasted from the Pentium Pro to Pentium 3, is that there have been relativly few advancements in CPU logic, and now some of the really old ones that were dismissed years ago are coming back by force. I point you to this link and implore you to read it before you make another attack, as all it does is undermine your credibility.
<A HREF="http:// http://www.tomshardware.com/technews/technews-20010422.html " target="_new">http:// http://www.tomshardware.com/technews/technews-20010422.html </A>

In closing, I'd just like to say to I enjoy having debates and discussing issues with others, as long as there is an attempt at finding sources to back up the "facts" and a refrain from the "Intel Rules, AMD sucks" and vice versa BS that I see way too much on these boards. I understand that everyone has their chip of choice, but to belittle people because of that is flat out wrong and immature, and again, only undermines your own credibility, as well as irritating others who wish to learn from these boards. I hope that you people know who you are, because it seems to me that everyone else does, and it is in your best interest to stop the brand bashing and focus on the issue at hand. This goes for both the Intel and AMD "lemmings"

-SammyBoy

Without Evil, there can be no Good. Therefore, without an Intel, there can be no AMD.
 
G

Guest

Guest
I remember reading recently that Intel announced that they discovered a major breakthrough in CPU manufacturing process technology which, if memory serves me correctly, would allow them to keep up with Moore's law for at least the next ten years. Im sorry, but I dont have a link to back this up right now.. but i will try to find it. When I do you'll see that reaching 100+ GHz in 10 yrs is not as unrealistic as you may think.