Prescott- AT calls it a flop??

[lookin4dlz]

Okay, just for fun I googled & found what the temps for some mid-range Athlon's are:Text - 58C to 76C idle & 64C to 86C at load. Looks like the Prescott may run slightly cooler or at worst case the same temps. So, my system temps are going to be a lot higher than I'm currently experiencing but nothing that those with AMD chips haven't overcome.

 

[txxxx]

Originally posted by: lookin4dlz
Okay, just for fun I googled & found what the temps for some mid-range Athlon's are:Text - 58C to 76C idle & 64C to 86C at load. Looks like the Prescott may run slightly cooler or at worst case the same temps. So, my system temps are going to be a lot higher than I'm currently experiencing but nothing that those with AMD chips haven't overcome.

AMD chips dont get THIS hot under load , when was the last time you looked at an AMD solution?

 

[SilverBack]

Well just got a Beta BIOS for my IC7 Max3, it supports Prescott, so I'm already I just need the chip! LOL

 

[Acanthus]

Originally posted by: THUGSROOK

Originally posted by: MadRat
People labelling it a flop forget the huge increase in the cache system. I'll give up 15W for an extra 512K of L2 cache AND out-of-the-box 4GHz potential any day of the week...

air cooling cant afford to give up 15W.
4ghz may be too hot to run for some ppl.
besides, 4ghz (OCed) results are yet to come ~ thats what we are waiting for ATM.
its just that so far things dont look so good, especially for OCers.

I remember when Pentiums were flame throwers because they needed a heatsink

And dont forget those flame throwing athlons!

Heatsinks will evolve like they always have.

 

[chsh1ca]

Originally posted by: lookin4dlz
Okay, just for fun I googled & found what the temps for some mid-range Athlon's are:Text - 58C to 76C idle & 64C to 86C at load. Looks like the Prescott may run slightly cooler or at worst case the same temps. So, my system temps are going to be a lot higher than I'm currently experiencing but nothing that those with AMD chips haven't overcome.

Ummm, Dude, I'd have to say that that article is full of crap. I had a T-Bird 1.2GHz and there's no way it EVER hit 82C with an HSF, and all it had was stock cooling. Even now my 1.667GHz Athlon XP 2000+ is lucky to break 50C with stock cooling.

I think part of the problem comes from:

From: Their test setup page:

The test system was placed in the middle of a temperature chamber. As this test is to show efficiency of CPU coolers and not of case designs, the cpu cooler was tested in the open. A constant, slight air flow kept the temperature in the chamber evenly distributed. Ambient temperature was kept at exactly 35°C. This is about the temperature you may find inside a computer case.

Last time I checked my ambient case temp was lucky to be above 20C, and all I have is a side intake fan and PSU exhaust. This is not a spiffy special case or anything. It seems to me they are using a flawed methodology in general. In the test setup, they applied the temp. probes to the heat sink itself. The heat sink is DESIGNED to get hot, because it takes heat from the processor core and bleeds it away. Am I way off here?

The numbers for the T-Bird 1.3/1.33s, possibly the hottest processors from AMD ever (correct that if I'm wrong, Bartons may run hotter) are 71 and 73W, the P4Es START at 89W...

 

[Fraggster]

Originally posted by: Freejack2
What a piece of crap this processor is, I think I'll stick to my Northy at 2.8ghz or look into an Athlon 64.
This hunk of junk doesn't overclock, runs hot as hell, and slow as crap. Maybe if this thing could do a guaranteed 3.6ghz overclock it might be worthwhile but Hardocp and Toms hardware said it was a piece of crap overclocker.

hardocp is sh!t, but i do think intel kinda screwed itself here

 

[DAPUNISHER]

Originally posted by: pastorjay

But as of right now I would buy a A64 3000+ if I needed to upgrade.

-Por

You may be right. But for now, I think we ought to wait and see what happens whenwe get the retail chips. I have a 3000+ and am totally impressed with it. But if this Prescott scales as high as what is imagined, we might be liking it VERY much. Bottom line is, we will not know until we get them...

PJ

That's the millon dollar question.

 

[Nyati13]

Originally posted by: lookin4dlz
Okay, just for fun I googled & found what the temps for some mid-range Athlon's are:Text - 58C to 76C idle & 64C to 86C at load. Looks like the Prescott may run slightly cooler or at worst case the same temps. So, my system temps are going to be a lot higher than I'm currently experiencing but nothing that those with AMD chips haven't overcome.

My A64 3200 idles at 32C (gotta love CoolnQuiet ), and maxes out at 47C (with a relatively high 30C MB temp, because I have sound insulated my case)

Jeremy

 

[Nyati13]

Originally posted by: lookin4dlz
Okay, just for fun I googled & found what the temps for some mid-range Athlon's are:Text - 58C to 76C idle & 64C to 86C at load. Looks like the Prescott may run slightly cooler or at worst case the same temps. So, my system temps are going to be a lot higher than I'm currently experiencing but nothing that those with AMD chips haven't overcome.

My A64 3200 idles at 32C (gotta love CoolnQuiet ), and maxes out at 47C (with a relatively high 30C MB temp, because I have sound insulated my case)

Jeremy

 

[lookin4dlz]

Originally posted by: txxxx
AMD chips dont get THIS hot under load , when was the last time you looked at an AMD solution?

Ummm.... you didn't read my post "just for fun I googled & found what the temps are" Obviously, if I have to google I'm not running an AMD system & don't know what the temps are...

Anyway, you've missed the point which is that other chips have run as hot and were able to be cooled.

 

[CraigRT]

Originally posted by: THUGSROOK

Originally posted by: JustStarting
...Who's buying??

not me

not me ... I don't buy anything with the name Intel on it.
I like my hard earned money

 

[DAPUNISHER]

Originally posted by: CraigRT

Originally posted by: THUGSROOK

Originally posted by: JustStarting
...Who's buying??

not me

not me ... I don't buy anything with the name Intel on it.
I like my hard earned money

Then don't buy an A64 either Craig

 

[iguanaman]

Anyway, you've missed the point which is that other chips have run as hot and were able to be cooled.

The point is that this is the first die shrink that uses MORE power/mhz than it's larger ancestor. And it's not the extra cache...Intel has screwed up the 90nm process and they haven't been able to fix it even after a years delay. I doubt any chip has used the watts per square cm that a Prescott@ 4ghz eats up. Intel made a BIG mistake not going with SOI technology like AMD and IBM.

 

[Snufalufagus]

I'm selling my 2.4C to a friend for a good price. I'm going to upgrade my CPU only. For the price which CPU would be the best for overclocking?
2.8C
2.8E(waiting for retail review)
3.0C
3.0E(waiting for retail review)
Its only going to cost me $60 to $80 for a new CPU after I sell my 2.4C. Will it be worth the extra cost for me to get a new CPU? I want to overclock higher without having to push my system so hard.

 

[MadRat]

Originally posted by: iguanaman

Anyway, you've missed the point which is that other chips have run as hot and were able to be cooled.

The point is that this is the first die shrink that uses MORE power/mhz than it's larger ancestor. And it's not the extra cache...Intel has screwed up the 90nm process and they haven't been able to fix it even after a years delay. I doubt any chip has used the watts per square cm that a Prescott@ 4ghz eats up. Intel made a BIG mistake not going with SOI technology like AMD and IBM.

When you consider that this SAME BASIC CORE LAYOUT will also be used in the XEON then I don't think its a flop at all. Xeon is moving towards an on-die 4MB L3 cache, added to the already impressive 1MB L2. The Prescott also has a second execution unit and an extra deactivated D-cache stuffed into it, which means Xeon could have basically dual-cores activated - and add to it the fact that they already have SMT...

Every die has been hotter compared to the last when the transistor count increase relative to the surface area is taken into account. Just pump 2v into a TBred and see how hot it gets! (Easily will compete with a TBird at the same speed if not get hotter due to higher leakage by the smaller process.) The voltage didn't exactly take a nose dive like it usually does on a die shrink and thats the main difference. Intel pushed the new core onto an older existing package, which by default makes putting power to the core inefficient already. More pins will allow better powerflow and this will change with the Socket T package's better electrical handling. When the voltage drop occurs, the socket change solves their powerflow problems, and Intel refines their low-k process, then the temperatures will decrease accordingly. Until then they still have one heck of a desktop CPU.

 

[Wingznut]

Originally posted by: iguanaman

Anyway, you've missed the point which is that other chips have run as hot and were able to be cooled.

The point is that this is the first die shrink that uses MORE power/mhz than it's larger ancestor. And it's not the extra cache...Intel has screwed up the 90nm process and they haven't been able to fix it even after a years delay. I doubt any chip has used the watts per square cm that a Prescott@ 4ghz eats up. Intel made a BIG mistake not going with SOI technology like AMD and IBM.

"... years delay"? Since day one of the PSC announcement, Intel has said "H2 '03", with a target of Christmas. If you want to call it a 6 week delay, that would be fair. But a year is just wrong.

 

[Keysplayr]

What was it that AMD introduced that was actually performing slower than it predecessor? Was it the first barton? or the first with new die shrink? I forget. But I know it was a big dissappointment to AMD fans. Then AMD I think refined there new process and scaled up the speeds and only then was it able to overclock well.

The same thing happened when Intel introduced the P4. Holy Cow what a let down! Then they to refined and scaled. Look at the bartons and northwoods now, they are powerhouses and overclock amazingly. I say Prescott should be given a little time to mature with refinements. The thing thats really kewl about this prescott is that it's comparably as fast as an equally clocked Northy. At least we didnt lose ground there. Anybody have any idea what the "secret" that intel will divulge on the 14th about the prescott? Very curious. I hope its something special like 64 bit capability or Quadrathreading or something like that.

My 2 nickels.

Keys

 

[MadRat]

The first P4 had advantages over the P!!!, just not in general purpose computing. To say it was slower is slander. The state of Illinois couldn't even prove the P4-based Celeron was slower than comparable clock speed P!!!c-based Celerons.

 

[Acanthus]

Originally posted by: Acanthus

Originally posted by: videoclone
I still dont think it will stand up to WindowsXP64 with an Athlon64 + 64bit games it will get smoked ... even now its being smoked in 32bit games.

In 2005...

Ill be the 1st to correct myself.

Windows XP-64 is now available in a 365 day trial beta version.
If you can get drivers for all your hardware, you can go 64-bit now.

 

[joe2004]

I think everybody that are claiming that Prescott is bad are missreading. Effortless overclock to 3.75 GHz on default voltage. That is not a joke. We are looking at the 4 GHz beast with a proper cooling. And beyond. You cannot do that with Northwood unless you use phase change.
This thing has potential and with a basic watercooling setup right now you can get it to 4 GHz. That is enough to make it attractive for overclockers.

 

[PetNorth]

Originally posted by: joe2004
I think everybody that are claiming that Prescott is bad are missreading. Effortless overclock to 3.75 GHz on default voltage.

Thats a 14% OC for 3,2... Is that great?

 

[Soulkeeper]

someone on another forum said they got it to 4.2ghz on air with 1.44v
they even posted a cpuid pic but i lost the link

i'll see if i can find it again

it's hard to beat a 1.2ghz fsb

 

[Duvie]

Originally posted by: joe2004
I think everybody that are claiming that Prescott is bad are missreading. Effortless overclock to 3.75 GHz on default voltage. That is not a joke. We are looking at the 4 GHz beast with a proper cooling. And beyond. You cannot do that with Northwood unless you use phase change.
This thing has potential and with a basic watercooling setup right now you can get it to 4 GHz. That is enough to make it attractive for overclockers.

Exactly Joe!!

now ppl just get an asus mobo and the temps will look fine if not even better then mine on my abit mobo.....I can't imagine any of them working on Abit mobos due to the higher then average reported temps....

 

[reever]

Last time I checked my ambient case temp was lucky to be above 20C

Last time I checked my house has heat, and anybody elses house most likely also has, which would bring the ambient temps well above 20C, not to mention the many heat generating parts on a motherboard

 

[Soulkeeper]

alright everyone Here is 4.2ghz incase you havn't seen this already

i'm impressed with that but i can't be sure if he did it on air or not cause i ended up having to search google for it and it came up listed on some german site
also i can't guarantee that he actually did it, but i don't see any reason why it isn't true

also there is someone else claiming to have their 3.2 prescott at about the same speed and voltage on air Over at vr-zone's oc database

lookin good
i wouldn't be surprised if this changed a few minds about what's possible