P4EE @ 4.1G vapochilled action!

Quote:

---

Intel is toying with the idea of releasing an Extreme Edition version of every high-end Pentium 4 (e.g. Prescott 3.40GHz Extreme Edition), however nothing is set in stone yet. We have already passed along the information that an Extreme Edition processor would truly be worthy of the name if Intel would unlock the processors, allowing overclockers to freely push their processors. In order to combat remarking, we also passed along the suggestion that only lower multipliers be made available.

Both of these suggestions were provided by AnandTech readers and were very well received by Intel, it may take some time but we may be able to get the chip-giant to budge on this one.

---

intel ee > athlon 64???

Jason, you and I both know that if you only take into account system scores, your 2.4GHz isnt as powerful. ;)

This chip shouldn't be compared to an A64. It should be compared to an FX. Noone in their right mind compares a $900 chip to a $400 one.

It just so happens the only highly-OC'd FX on the ORB is Viggens, and atm its system scores are untouchable. I mean, cmon. 140FPS Car High. Insane.

I dont think the P4EE is its prime though, as 1:1 should really boost the Car High. As of now it seems that AMD > Intel @ Car Chase, and Intel likewise claims lobby.

What needs to happen: JC getting a better card. ;)

http://www.aceshardware.com/read.jsp?id=60000270

Caugh, Caugh p4 what??????

Stock Athlon64 scores 20k points in 2001 mark and intel needs near 4ghz to hit that with a stock 9800 pro.

Quote:

---

Originally posted by Dr.Demonic69
http://www.aceshardware.com/read.jsp?id=60000270

Caugh, Caugh p4 what??????

Stock Athlon64 scores 20k points in 2001 mark and intel needs near 4ghz to hit that with a stock 9800 pro.

---

I refer you to my post in the other thread.

Regards

Andy

ALT-F13

harasho brat dai esho choot choot

lol just some russian. russians own overclocking =D

CrashOv3r1De
he, prikolno, tut okazivaetsa ne tolko mi takie :)

sorry everyone. Just said hi.

Quote:

---

Originally posted by CrashOv3r1De
ALT-F13

harasho brat dai esho choot choot

lol just some russian. russians own overclocking =D

---

But's he's Ukranian ? It's like me saying you sounded like a Canadian ! :D

LOL

Regards

Andy

there are multiple sse2 registers in teh cpu on teh ath 64/fx, but only SOME, i think around 60% of them or something, can fit within the 32bit addressing, or something like that. SO the rest can ONLY be used under 64bit mode operating, this is certain, or there is no logical reason for the same motherboard, gfx card, cpu, ram becoming that much faster under 64 bit mode.

I'm also sure i read that teh 1meg p4 was canned, and i really can't see them making a 1 meg version, its possible but i feel very unlikely as they won't have a plant set up for that right now, which would mean changing fabs over to that, then to the prescott. the 2meg EE chip didn't need to be made differently, just rebadged xeons already being made.

is the prescott 64bit extensions certain, last time i really read on it(got bored and was a while ago now) it was all rumours a la "the inquirer" so i ignored them. It would seem strange as, they've taken flak for not having a amd equaling 64bit capable chip yet they said nothign till recently, assuming they did. Also, they've been touting the prescott and its 13 new instructions(sse3) as teh only real feature of the prescott, afaik. SSe3, is a big joke, sse2 was useless for a long long long time, but is actually semi useful when apps are writen for it, sse3 won't be, full stop.

also, afaik by the time the prescott is out, which if at 3.4Ghz, afaik will be slower than the p4 3.2EE, will be their top end chip of release, by that point we should certainly have the 3400+ out, new stepping maybe, maybe not, as in good new stepping. But by feb or so we might be looking at a 3600+ and the 200mhz faster fx version in socket 939. As i see it, prescott, till it moves to 0.09 process, jsut won't be able to compete, and will have awful heat issues as i said before, same heat or more than the EE, but way less die area to cool from.

intel always hated ocers, the fact that intel might release a cpu especially dedicated to ocers now only shows their fear to loose customers to amd.

i wonder what a fx with some L3 cache would perform like.

drunkenmaster, prescott IS 90nm already...

isnt it funny how intel says 90nm process works great but delays the prescott at the same time :D

and prescotts run a lot warmer than nws made with 130nm at the same clockspeed, and from what ive heard intel has big problems to get prescott chips running at high clockspeeds because the chip has a few hot spots that already reach a critical temp at low clockspeeds.

Still can't help myself wondering what will the price of the 939 pin be.

AMD was known for cheap products but the $750 A64 FX-51 just don't fit into this category.

However the A64 3200+ was a very good bang for the buck and still is. It was great when the P4 3.0C was around $400 but now its @ 270$ so that may be turning some people off from A64 and gdeciding to go intel.

ish, but very soon, with this 3000+_, and when the 3400+ comes out we will have some very cheap, i'm 99% sure none speedbinned(not for a bit) chips that will simple kill p4's with simple cooling. I mean most of the market is for aircooling, even most of the overclocking/enthusiast market is aircoole,d huge amjority. With crap ram low timings, not best gpu clock terrible ram clocks i hit 23k finally in the nf3, all stock , silent aircooling. I don't think thats possible with a P4 rig, i really don't. Not this simple anyway. temps, stability wise, i think i'm set to go significantly higher still on stock air cooling, so as those t-break thread from a month or two ago suggested, 25k stock cooled ath 64 rigs will be pretty common.

AS ath 64 prices tumble with newer higher speed versions coming out, they will again but the cheaper option.

don't get me wrong, actually you won't, i obviously love amd, they've broughnt me cheap clocking for a while now. But fan boy, no, just honest opinion, the prescott is smack in teh middle of major problems, thats obvious, and for the benifit we are gonna see for teh clock for clock speed against a current P4, its not worth it, a simple p4 northwood moved to 0.09 would have been the best way to move to it, then perfect the 0.09 for a while, then add more cache to it.

also, i was pretty convinced intel were gonna, as amd, release new chip on 0.13 and move it to 0.09 asap, feb/march kinda time. was taht the original plan a while back, or, meh, i can't remember.

All the FX-51 is was a very, very good guess by AMD that Intel would, like complete tools, paper launch a nigh-nonexistant chip only to steal their thunder.

Seems AMD guessed right.

Quote:

---

Originally posted by MickeyMouse
MY comment was sarcasim :0

ALT> BTW if i havent said so awosme JOB in a short time with that chip.. :) I thought I was P4 safe at 26.8k thta doesn't look to good any more :)

---

Yer but your 'sarcastic' comments and screenshots are flooded all over the forum dood ;) Also as someone else mention, your comparing the wrong cpus...compare the FX51 to the 3.2 EE pls :banana:


ALT-F13 : Very nice overclocking and benchmarking here guys. As someone said in the other thread, the AMD guys are getting scared again...but this isnt a cpu contest thread is it :thumbsup:

JC, How come you think not many cpus will hit retail? I have one on preorder now and the shop locally here says about 2 weeks so I guess Ill just cross my fingers and wait and see :)

Dr.Demonic69 lol you preach harder than my chaplain on a Sunday morning at church :p:
hehehehehehe

Also I dont know about in the USA, but we wont be seeing any of the new AMD chips below $350 for another 6 months downunder here in Australia....:(


ps
I noticed your using 13x319fsb for this 26K run......VERY nice run, but not what the avg person will be doing no? You have an ES chip....you have unlocked multis.....can u pls run the chip like an avg user would so we can see the *REAL* performance pls? eg; default multi and highest fsb etc :)

Or are all EE unlocked? :confused:

Isn't it kind of early to tell? I mean until OPP or macci gets one of these and is allowed to have their ways with it, who knows ;)

drunkenmaster, the 13 new instructions is not the only new feature of the processor. SSE3, along with longer pipelines, an improved branch predictor, increased trace, L1, and L2 caches, imoproved architecture, improved HT, 64 bit extensions, and more registers all contribute to why this processor runs so hot. It will be at least 10-20% faster than the current P4 at the same clock speed.

Quote:

---

Caugh, Caugh p4 what??????

Stock Athlon64 scores 20k points in 2001 mark and intel needs near 4ghz to hit that with a stock 9800 pro.

---

3.2C (non-EE) at stock speed with R9800Pro hits 19564 in Mark2001.

Quote:

---

Originally posted by macci
3.2C (non-EE) at stock speed with R9800Pro hits 19564 in Mark2001.

---

You must have had a really good run m8, this is what I get 3200 MHz 1:1 P4C800 R9800pro stock

at 4 GHz it goes to 20500

You must be using some poor settings there mate ;)
19.6k stock

At 4GHz its 21573

Quote:

---

Originally posted by macci
[B]You must be using some poor settings there mate ;)

---

Well i test it like any joe average would.. WinXP, DX9b and the latest cats. memory is at 2-5-2-2 TURBO PM enabled. There's definately no faster BIOS settings in the BIOS :)

The Athlon system is just as "untweaked" anyway and gets 20.100 at stock speed and 21.400 at 2700 MHz at 200 FSB. Then the radeon is such a bottleneck that the score doesnt rise much anymore

BTW, 250 FSB 1:1 2-5-2-2 doesnt really count, there is no memory for sale that does that :( I run 250 5/4 fasterst timings

if teh rpescott is 10-20% faster than the p4 northwood clock for clock, then, with the 3.2EE being about 10% faster, yet teh prescott has been said by intel to not be as fast as the p4 3.2 EE. WEll, if you can explain how intel say that it will be slower than the EE, yet you say it will be 20% faster than the northwood. to me those numbers don't add up at all.

JC & Macci

JC you really have small score... Mine was 19300 with 3.2C, P4P800 and default clocked raddy...

CrashOv3r1De
Lol.

zakelwe
well, it not that:) I do not consider Ukraine and Russia as separate countries... And i'm russian myself though living in Ukraine. So it's not like USA/Canada :)

Hallowed
Yup... If me or JC will get a card like 560/410 probably we'll reach the top...

Quote:

---

Originally posted by drunkenmaster
if teh rpescott is 10-20% faster than the p4 northwood clock for clock, then, with the 3.2EE being about 10% faster, yet teh prescott has been said by intel to not be as fast as the p4 3.2 EE. WEll, if you can explain how intel say that it will be slower than the EE, yet you say it will be 20% faster than the northwood. to me those numbers don't add up at all.

---

It depends on the application. Some don't benefeit at all (0%) from the extra cache, and some benefeit by a huge amount. In CPU intensive tasks the Prescott will be the faster and more efficient processor due to its architecture. In some places the EE may have a slight edge but I doubt it will be by much, and only in places where the large cache comes in to place.

I have seen certain apps where the 3.2 EE is tied with the normal 3.2.

Quote:

---

Hallowed
Yup... If me or JC will get a card like 560/410 probably we'll reach the top...

---

if your EE maxes at 4.1Ghz and almost 1.8V then i would say no. so not happening:) 27.5k would be your max. but even if you will, you are going to be taken down by an FX chip, i can guarantee you that m8:) 2700mhz FX, 550/400 would break 28k easy and that's with too low Car Low run. Nice job there guys nonetheless.