Prescott Time

Scout

Distinguished
Dec 31, 2007
1,068
0
19,280
Fry’s had a 2.4 Prescott on sale and I couldn’t resist… I know… 533 FSB, no Hyperthreading, and it generates a TON of heat, but what the heck… I’m fascinated by new stuff at the right price! I immediately popped the FSB up to 167 to push the chip to 3 GHz. on the stock cooler and standard 1.4 volts. At that speed, the chip temperature is in the mid 40’s at idle and goes to the mid 60’s under 100% load. Hotter than I would like, but it doesn’t seem to show any initial instability at those temps.

Now the impressive part; I run a lot of SETI (3rd place on the THG Team) and I use SETI to compare machines. My fastest Athlon XP rigs can average about 400 Mflops/sec on a work unit. The new Prescott just crunched a unit at an average of 615 Mflops/Sec! Just to be sure I’m running the same work unit on my XP3200+ and sure enough… it’s running around 400 Mflops... My old Northwood 2.2 was a SETI weakling! But I assume this impressive Prescott performance has something to do with SETI fitting in that nice 1 Meg cache…


Scout
700 Mflops in SETI!
 

Crashman

Polypheme
Former Staff
Try setting it at 200MHz bus to get 3.6GHz!

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>
 

P4Man

Distinguished
Feb 6, 2004
2,305
0
19,780
mid 60's should be fine, if you can achieve it with a noise level that doesnt bother you. And if its achieved with an ambient/case temperature you won't exceed in the summer. Prescott starts throtteling at around 80°C (see <A HREF="http://www.digit-life.com/articles2/p4-throttling/index.html" target="_new"> digit life article </A>).

I do have to wonder why you picked up a Prescott instead of a northwood though. Unless you crush Seti for a living ;) NW would probably give you better performance for most things, and be cooler/more silent to boot.

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

Scout

Distinguished
Dec 31, 2007
1,068
0
19,280
I always have problems with Summertime heat with my rigs. I tend to clock them down a little bit to compensate for the higher room temps. I've read the Prescott throttling datasheet and it's an interesting concept. It pretty much makes it sure I'll never cook this thing. It throttles at one temp and shuts itself down if it gets to a certain point.

It's no more noisy than my Athlon boxes in terms of fan noise. Course' that's not to say it's very quiet. In my upstairs room I use a very quiet setup, but down here in the garage, noise isn't a problem...

Why buy it? Cause it was $102 and the Northwood 2.4's are more like $150! Plus, I already have a Northwood and it's a nice chip but not terribly impressive. I know the 2.4's would likely make it to 3 GHz. as well, so you're right, right now the Northwood would be the choice for most people. My next box is going to be an A64, but the price has got to come down some on those for me to buy one.

Scout
700 Mflops in SETI!
 

Scout

Distinguished
Dec 31, 2007
1,068
0
19,280
I'll get to that after I make sure this thing runs well at 3 GHz. for a while. I read one article (Lost Circuits I think) where they showed putting the voltage to 1.5 had a dramatic increase in the CPU temp on these Prescotts. And I'm going to need to start pushing the voltage if I go for more speed I think.

Right now I'm running UD on it all last night and the CPU temp is 64 this morning. Today is supposed to be a nice sunny day, so I'll see what the temp does this afternoon. To get to 3.6 I'll certainly have to do something better for cooling. Hummm... water perhaps...

Scout
700 Mflops in SETI!
 

P4Man

Distinguished
Feb 6, 2004
2,305
0
19,780
>Why would Northwood be quieter? Doesnt it use the same
>stock cooler as the Prescott?

I think not, but if it does, it will definately be hotter :) Hence my phrase "cooler/more quiet". You could probably let it run as cool as a northwood using a turboprop powered fan, or you could make it as silent as NW if you don't mind >70°C temps. Can't do both with the same HSF though.

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

addiarmadar

Distinguished
May 26, 2003
2,558
0
20,780
God that is an expensive heating element!!!

Mobile Barton 2500+ @ 2420mhz 11x220 1.7v
Asus A7N8X Dlx 440 FSB
1gb Geil GD pc3500 Dual Channel (2-3-3-6)
Segata 80gb SATA 8.5ms seek
ATI Radeon 9800 Pro(420/720)
 

Scout

Distinguished
Dec 31, 2007
1,068
0
19,280
It's an ECS (don't laugh) 848P-A based on the Intel 848 chipset (basically an 865 with single channel memory). The board and processor were a combo package. I have a 512 stick of Corsair PC2700 in it (Value Select CAS 2.5)

Scout
700 Mflops in SETI!
 

P4Man

Distinguished
Feb 6, 2004
2,305
0
19,780
>I've read the Prescott throttling datasheet and it's an
>interesting concept. It pretty much makes it sure I'll
>never cook this thing.

Probably not. it will rather be the <A HREF="http://www.hardocp.com/article.html?art=NTg3" target="_new"> motherboard that burns a hole through your case </A> or the PSU blowing up (sorry, don't have link, think it was on Sudhian).

:)

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

Scout

Distinguished
Dec 31, 2007
1,068
0
19,280
Yep, the Prescott definitely puts a lot of stress on things. I'm only using a 300W Power Supply too! It's Enermax though so pretty good quality.

It's a little warmer in here today so it's now cooking along at 70c... under 100% load with UD. Strange thing though is I have some Athlons that run awful hot as well and the heat sink on this Prescott doesn't feel as hot as those Athlon sinks to the touch. I know the Athlon XP has a diode as well now-a-days, but I think most boards take the temp off an Athlon below the socket with a separate sensor don't they? I'm wondering if this diode based Prescott temp is not that much different from my internal temps on my Athlons...

Scout
700 Mflops in SETI!
 

phial

Splendid
Oct 29, 2002
6,757
0
25,780
well thats because of the reduced die size and hence reduced contact with the heatsink


sounds like your not getting effective cooling then.. clean the cpu and reapply your cooling compound, make sure theres even contact and whatno

-------
<A HREF="http://www.albinoblacksheep.com/flash/you.html" target="_new">please dont click here! </A>
 

Scout

Distinguished
Dec 31, 2007
1,068
0
19,280
With the heat spreader on these Intel chips, the contact area is actually much larger than with an Athlon. The die itself is smaller, and perhaps that's part of the problem with these chips. The die is so small it can't get the heat into the integral heat spreader...

I used AS3 and did a careful fit check for good contact before I installed the heat sink.

Scout
700 Mflops in SETI!
 

phial

Splendid
Oct 29, 2002
6,757
0
25,780
oh yea what am i thinking about DUH HAHAHA


omg im such a newb

-------
<A HREF="http://www.albinoblacksheep.com/flash/you.html" target="_new">please dont click here! </A>
 

P4Man

Distinguished
Feb 6, 2004
2,305
0
19,780
>It's a little warmer in here today so it's now cooking
>along at 70c... under 100% load with UD. Strange thing
>though is I have some Athlons that run awful hot as well
>and the heat sink on this Prescott doesn't feel as hot as
>those Athlon sinks to the touch

Phial got it right. when you "measure" (feel) the heatsink temperature, you are "measuring" total heat production by the chip. With an identical power draw, and identical HSF, prescott's die would be considerably hotter because of the higher thermal density (W/mm²). The HSF on the other hand would be just as warm/cool. Heatspreader doesnt change this, its thermal resistance is at best as good as the HSF itselve.

[edit] think of it this way: how hot do you think a needle tip would have to be to warm up the HSF (and heatspreader) as much as your current cpu ? several hundreds of degrees for sure [/edit]

However, you are also right that temperature measurements ought to be taken with a serious grain of salt if your using bios measurements. Its accuracy is limited, and its only accurate using the same motherboard and then comparing relative heat of different cpou's/HSF's, etc. If you start comparing completely different MB's with completely different chips, those temps are at best a rough approximation. But don't fool yourselve, prescott is HOT, and a powersucking beast. Its power draw is substantially higher as an equivalant clocked northwood, plenty of tests show that. Obviously, the resulting die temps will be even higher due to the smaller diesize.

= 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 03/30/04 05:05 AM.</EM></FONT></P>
 

ChipDeath

Splendid
May 16, 2002
4,307
0
22,790
...hence the 'fire and brimstone' temps? :wink:

---
Epox 8RDA+ rev1.1 w/ Custom NB HS
XP1700+ @205x11 (~2.26Ghz), 1.575Vcore
2x256Mb Corsair PC3200LL 2-2-2-4
Sapphire 9800Pro 420/744