How hot is your prescott?

R

ryaneverett

Limp Gawd
Joined
Aug 14, 2001
Messages
452
How hot is your prescott? Mine (3.2E) seems to be running 64C idle on stock cooling. That seems f**king wrong.....
 
Seems to me like your system was an oven in it's previous life.
 
Try reseating the HSF, and use thermal paste like artic silver or OCZ ultra 5, no cheap shit, and no intel stock cooling pads.
 
Ya that pretty hot. I have a 3.0E runs 44c load. I also am using a zalman 7000. I'd check to see if your Heatsink is seated correctly.
 
40c idel and 50c under load. And that's correcting for the Asus's reading low. It says 32c idle and 46C under load.
 
ABIT Temps.

3200 @ 3600.

46*C Idle. 58*C Load.

Not sure what it is now though, as either my motherboard or RAM went tits up today and decided to stop working.
 
SimGuy said:
ABIT Temps.

3200 @ 3600.

46*C Idle. 58*C Load.

Not sure what it is now though, as either my motherboard or RAM went tits up today and decided to stop working.
Click to expand...

Is that corrected? Abit's usually read about 10C high.
 
Dillusion said:
Try reseating the HSF, and use thermal paste like artic silver or OCZ ultra 5, no cheap shit, and no intel stock cooling pads.
Click to expand...

I tried a Thermatake Spark 7, and even with the fan top speed it never went below 58C. I think the temp gauge is busted.
 
I'll let you guys know what my temps are like on my water cooling pretty soon......in about a week and a half when I get my 3.4E installed and crank it up to 4.25GHz in a few seconds. :cool:
 
I hate to be the voice of reason, but I would not hold a candle to any of those temperature marks that are read from the motherboard. If you really want to know how hot they are you should buy the right tools for that job. About the only thing the motherboard temperature gauges are good for is shutting down during unusual increases like fan failure. Other than that you can’t honestly use that information to compare with other people using many different brands and rev numbers.

Find the information on your CPU and check to make sure that your system is giving the correct amount of voltage. The Prescott is notorious for running at 60 with the OEM fan. Most of these Intel fan and Heat sinks are designed to speed up the fan when the CPU runs at high temperatures. So you should expect these processors to stay at the correct temperatures and use less power while idle rather than taking the extra time to cool down to 40 unlike most 3rd party heat sinks and fans that keep running at full speeds consistently. Hopefully that made a little since here?

If your system is not crashing then don’t stress about it. ;)
 
My 3.2E runs very hot as well, Idle sometimes as high as 60c, Prime 95 @64 to 65c. No crashes. The only time I have any problems is when I go above 220FSB, almost imediatley after I load windows it throtles down Prime reports errors so on and so on. I ordered a ThermalRight 120 hope it helps, or I will be looking for a Northwood. :
 
my 2.8E with an aero jet 4 and arctic silver 5 runs 66c idle and around 74c or so load
 
380rudy said:
My 3.2E runs very hot as well, Idle sometimes as high as 60c, Prime 95 @64 to 65c. No crashes. The only time I have any problems is when I go above 220FSB, almost imediatley after I load windows it throtles down Prime reports errors so on and so on. I ordered a ThermalRight 120 hope it helps, or I will be looking for a Northwood. :
Click to expand...

Could be more then temps holding you back if you're getting those kind of immediate errors. It's doubtfull as to whether the cooler will help at all or not, but the relatively small increase from idle to load really does show how the fan isn't working full tilt when nothing is under stress....much like the stock fan on the X800's...which require aftermarket software to have them run 100% all the time. :rolleyes:
 
Tivon said:
I hate to be the voice of reason, but I would not hold a candle to any of those temperature marks that are read from the motherboard. If you really want to know how hot they are you should buy the right tools for that job. About the only thing the motherboard temperature gauges are good for is shutting down during unusual increases like fan failure. Other than that you can’t honestly use that information to compare with other people using many different brands and rev numbers.
Click to expand...
Absolutely correct, but I do find the relative differences to be of value, much like the Vcc readings you get out of the BIOS; also notoriously inaccurate, in terms of absolute value....but - very accurate, I believe, in terms of relative values.

Explain; the +12 reading on my MoBo in the BIOS / MBM reads about 11.5 volts. My trusty Fluke 77 (decent volt-ohm meter) reads about 12.06-ish under load. Guess which one I believe...?? I view the temps the same way; you probably "really" don't know how accurate the absolute value is, but when you changing settings (FSB, Vcore, etc.) and suddenly you see your idle temps go up by a few degrees....you have a good clue as to what's changing, and by how much.

I also agree that since P4's "clock throttle" when they get hot, the chances of laying waste to one is pretty remote. Keep your eye on the north bridge, though; they run hot as hell in these i865 / i875 MoBo's and really do benefit from improved cooling.

Regards - B.B.S.
 
cornelious0_0 said:
Could be more then temps holding you back if you're getting those kind of immediate errors. It's doubtfull as to whether the cooler will help at all or not, but the relatively small increase from idle to load really does show how the fan isn't working full tilt when nothing is under stress....much like the stock fan on the X800's...which require aftermarket software to have them run 100% all the time. :rolleyes:
Click to expand...
What is it holding me back?
 
you probably "really" don't know how accurate the absolute value is, but when you changing settings (FSB, Vcore, etc.) and suddenly you see your idle temps go up by a few degrees....you have a good clue as to what's changing, and by how much.
Click to expand...

This is about all i use the mobo readings for. They are usefull as a point of reference, but it's easiest to use your mobo sensors to mainly just see how the temps are being effected by your action...not as an actual reliable source of information.

The easiest and most accurate way I have found to get an accurate reading of your CPU temperature is to pick up some kind of thermal monitoring device that has a probe that doesnt have that insulation around the wires so that you can physically put the probe under the cpu, and attach it beneath the cpu...which I consider far more accurate then relying on any motherboard sensor.

What is it holding me back?
Click to expand...

That isn't something I can straight away just know the answer too without actually sitting down with someone (or on MSN) or being at the system in question to see what's going on. What I'm getting at though, is unless you're trying to clock the cpu WAY to high all at once, you won't usually run into heat related issues/lockups/errors within the first few minutes or so of running Prime95 or whatever else. It is still a possibility, don't get me wrong.....I'm far from telling you NOT to get new cooling. I'm actually at the point with my X800XT-PE where at 1.65v with the stock cooler the GPU gets to about 75C under load...up a full 40C from idle. With those kind of crazy (but thankfully temporary) temps I'm able to pull off benchmarks at up to 636MHz...but even 640MHz causes heat related lockups.....as the card just isnt able to maintain that extra 4MHz on the OC under those extreme conditions. So, it DOES happen that heat can set it quite quickly with an overclock, and depending on your particular CPU/video card you may run into those issues sooner then others....because if the cpu may not want to overclock as much to begin with, you're going to be "pushing the envelope" that much sooner with the clocks, and heat may start to play a role before you'd think it normally would.
 
cornelious0_0 said:
This is about all i use the mobo readings for. They are usefull as a point of reference, but it's easiest to use your mobo sensors to mainly just see how the temps are being effected by your action...not as an actual reliable source of information.

The easiest and most accurate way I have found to get an accurate reading of your CPU temperature is to pick up some kind of thermal monitoring device that has a probe that doesnt have that insulation around the wires so that you can physically put the probe under the cpu, and attach it beneath the cpu...which I consider far more accurate then relying on any motherboard sensor.



That isn't something I can straight away just know the answer too without actually sitting down with someone (or on MSN) or being at the system in question to see what's going on. What I'm getting at though, is unless you're trying to clock the cpu WAY to high all at once, you won't usually run into heat related issues/lockups/errors within the first few minutes or so of running Prime95 or whatever else. It is still a possibility, don't get me wrong.....I'm far from telling you NOT to get new cooling. I'm actually at the point with my X800XT-PE where at 1.65v with the stock cooler the GPU gets to about 75C under load...up a full 40C from idle. With those kind of crazy (but thankfully temporary) temps I'm able to pull off benchmarks at up to 636MHz...but even 640MHz causes heat related lockups.....as the card just isnt able to maintain that extra 4MHz on the OC under those extreme conditions. So, it DOES happen that heat can set it quite quickly with an overclock, and depending on your particular CPU/video card you may run into those issues sooner then others....because if the cpu may not want to overclock as much to begin with, you're going to be "pushing the envelope" that much sooner with the clocks, and heat may start to play a role before you'd think it normally would.
Click to expand...
Don't get me wrong I am happy with the way my system works with the exeception of the heat. It runs very hot without overclocking, idle is around 60c and running Prime 95 64c and that is at 200FSB, when I do overclock it is not much hotter maybe 1c more. I never ran the FSB up all at once it took me about 3 weeks to get to 220, running Prime 95 for 24Hrs each time to test stability. I am however dissapointed that this is where I hit a brick wall, I was hopping for close to 4.0Ghz. I don't thnink it is the heat preventing this I think it might be something else I just don't know what. I have quality Memory Corsair TwinX 1024-3200, My power supply should be enough Antec Tru Power 430, My VideoCard is an ATI AIW 9800XT I have a WD Cavier 250GB HDD SATA, and only one optical drive at this time. If anything my week link is the PSU. I havent had any lockups or BSD's just errors in Prime 95 above 220FSB. If you have any sugestions I would certainly appreciate the advice.
 
380rudy said:
If anything my week link is the PSU.
Click to expand...

Which is just what I was thinking.

Even though software monitoring of PSU rails isn't the most accurate way to do things.....what is your 12v rail sitting at? You could also give it a look in your BIOS. To eliminate the memory as a possible culprit you'd simply have to set it to 3:2 or 5:4 and start to scale the CPU...so it's doubtfull that your memory has anything to do with it.
 
12 V is at 11.96 according to my Digi Doc, I dont know what you mean by setting the memory
 
35-38c Idle and 50c load on retail cooler.

I <3 winter oh and its a 3.0E
 
380rudy said:
12 V is at 11.96 according to my Digi Doc, I dont know what you mean by setting the memory
Click to expand...

The 12v is nothing to worry about in that case (most likely anyways) and when I said "setting the memory" I meant going into your BIOS and changing the DRAM ratio to 5:4 or 3:2 to clock the memory down, eliminating it as a possible problem when overclocking. Depending on your board, 5:4 might appear as DDR320 and 3:2 may be in the form of DDR266.....but it'll be there, one way or another. At 200fsb, "setting" your memory to 5:4 will clock it down to 160Mhz, or 320Mhz DDR.....as (200x4) / 5 = 160. The same goes for 3:2, except (200x2) / 3 = 133...or 266MHz DDR.
 
I just bought Prescott 3.4 with HYPER 6 from COOLERMASTER and i got some awesome results, my IDLE is 34C and FUll 2 hours Load and testing with PRIME95 I have 51C . This cooler is Awesome.

I tried running XP-120 with this processor but the Temp was high.
I recommend running Hyper 6 with these hotties!!! :cool:
 
I put my new 3.4 in this weekend. It could do 6 hours of prime 95 at 3.7 GHz with a voltage bump to 1.425 in the bios (around 1.36 -1.4 under load). But the temps were nuts. I hit 76 degrees c under load at 3.7 GHz. Over 50 at idle. At 3.4 and stock volts it was still doing 70 c under load. This was with the stock cooler, and case temps of 28 c. Even though this is an Abit board, those temps are just crazy. I ordered some watercooling for it, since I have to watercool my x800 (since i fried the fan on it.)

Still, it was amazing that it was stable at 3.7 at such a high temp.

The adjustable multi is cool. 14x and 17x kinda suck though. I would like a 15x and 16x multiplier better.

The stock fan is a piece of crap. Highest speed I could get it to go was 2500 rpm.
 
Jonsey said:
I put my new 3.4 in this weekend. It could do 6 hours of prime 95 at 3.7 GHz with a voltage bump to 1.425 in the bios (around 1.36 -1.4 under load). But the temps were nuts. I hit 76 degrees c under load at 3.7 GHz. Over 50 at idle. At 3.4 and stock volts it was still doing 70 c under load. This was with the stock cooler, and case temps of 28 c. Even though this is an Abit board, those temps are just crazy. I ordered some watercooling for it, since I have to watercool my x800 (since i fried the fan on it.)

Still, it was amazing that it was stable at 3.7 at such a high temp.

The adjustable multi is cool. 14x and 17x kinda suck though. I would like a 15x and 16x multiplier better.

The stock fan is a piece of crap. Highest speed I could get it to go was 2500 rpm.
Click to expand...

Be sure to let us know how it goes.....3.7GHz is a decent clock....but not for a 3.4E, as 300Mhz isn't really that amazing for a P4. Hopefully the h2O takes care of your temps as much as you need it to and you clear 4GHz.
 
cornelious0_0 said:
Be sure to let us know how it goes.....3.7GHz is a decent clock....but not for a 3.4E, as 300Mhz isn't really that amazing for a P4. Hopefully the h2O takes care of your temps as much as you need it to and you clear 4GHz.
Click to expand...

Well, I can't imagine I'd get 300 MHz more just from watercooling. I wouldn't mind though. I'm hoping to drop the load temps 30 c. I really don't want to be above 70c. What's a safe voltage to take a prescott too?
 
Hey guys what is a relaible way to check the temps?
I used to just browse my bios, but lately been using a program called hdd_thermometer_v1.2
for my HD anyone know some pretty decent temp programs for monitering?

^___^ i also made some recent mods to my computer
added a Zalman CNPS7000A and a Vantech 120mm dtealth fan for intake and two lil brusless fans for outake. Also added some round cables for better air flow.

http://s4.yousendit.com/d.aspx?id=67F36C161CDD75AA787E7CF717DAE453

Here is a picture of my rig =)
 
36c idle, 48c load w/ Prescott 3.2e using stock hsf and generic goo (don't waste my money on AS). Three fans total in computer (hsf, power supply fan and rear case fan).
 
hagbard said:
(don't waste my money on AS).
Click to expand...

Wow, you're freely open to your own opinions, but don't get me started on how ridiculously ignorant that sounded. I'm sure you've got your reasons for not buying AS, but calling it a plain waste of money kinda puts you off in your own little corner here. ;)
 
hagbard said:
36c idle, 48c load w/ Prescott 3.2e using stock hsf and generic goo (don't waste my money on AS). Three fans total in computer (hsf, power supply fan and rear case fan).
Click to expand...


Those numbers are pretty awesome, I can't remember any review that had the stock HSF getting near that. You are lucky sir!
 
cornelious0_0 said:
Wow, you're freely open to your own opinions, but don't get me started on how ridiculously ignorant that sounded. I'm sure you've got your reasons for not buying AS, but calling it a plain waste of money kinda puts you off in your own little corner here. ;)
Click to expand...

Here is how I came to that conclusion. I built two identical machines, one used AS the other Vantec goo. I then measured the temps using MBM, the bios and Asus Prob. The Vantec stuff outperformed the AS in the first test. I then cleaned off the cpus, and swtiched the goo on each (following manufacturers instructions to the letter). Exact same results, the Vantec stuff outdid the AS by about 3c. The best I can say about AS is that it isn't that much worse than the Vantec stuff, maybe Vantec just makes a superior product, and they're both good. But I don't think so. To me, people pushing AS are victims of the bandwagon effect, or work for AS or receive some kickback from them ;)
 
ryaneverett said:
Those numbers are pretty awesome, I can't remember any review that had the stock HSF getting near that. You are lucky sir!
Click to expand...

I had a look at my Asus Probe history file, I actually maxed out at 51c a few days ago. What helps is having a 120mm output fan right next to the cpu fan (its an Antec Sonta case). Its also not being overclocked, though I did slightly overclock it when I built and I don't recall going much beyond the mid-50s (if that). Again, I think people make way too much of hsf, cpu goo, etc. Unless you're seriously into overclocking, I wouldn't waste the money (I'd just put that money into a faster cpu from the start). Mind you, I have a AMD 2500+ system o/ced to 3500+, but it didn't require much tinkering (I did buy an Vantec Aeroflow for it, which dropped the temp by a tiny amount over the stock hsf).
 
On my Gigabyte MB I can change the Mem from 2x to 1.65x, and did. I installed my new Thermalright XP-120, and am currentley @ 225FSB, Folding, running Prime, and typing this Post NO Problems at all. My CPU temp is at 42c, that is down 20c. Thanks for the advice I couldn't be any happier. I guess I need better memory, any sugestions?

Thanks again
cornelious0_0 said:
The 12v is nothing to worry about in that case (most likely anyways) and when I said "setting the memory" I meant going into your BIOS and changing the DRAM ratio to 5:4 or 3:2 to clock the memory down, eliminating it as a possible problem when overclocking. Depending on your board, 5:4 might appear as DDR320 and 3:2 may be in the form of DDR266.....but it'll be there, one way or another. At 200fsb, "setting" your memory to 5:4 will clock it down to 160Mhz, or 320Mhz DDR.....as (200x4) / 5 = 160. The same goes for 3:2, except (200x2) / 3 = 133...or 266MHz DDR.
Click to expand...
 
380rudy said:
On my Gigabyte MB I can change the Mem from 2x to 1.65x, and did. I installed my new Thermalright XP-120, and am currentley @ 225FSB, Folding, running Prime, and typing this Post NO Problems at all. My CPU temp is at 42c, that is down 20c. Thanks for the advice I couldn't be any happier. I guess I need better memory, any sugestions?

Thanks again
Click to expand...

How much money have you got to spend on some new ram.....and np on before.....it's what I'm here for.
 
I am currently running Corsair. 1GB Twinx1024-3200C2PT, I would love to get to 4Ghz. Right now I've made 3.6. I don't know what you mean by PC Rating.
cornelious0_0 said:
Alright, now sorry if I missed it, but what ram are you running right now, and what sort of "PC" rating and/or FSB are you looking for?
Click to expand...
 
380rudy said:
I am currently running Corsair. 1GB Twinx1024-3200C2PT, I would love to get to 4Ghz. Right now I've made 3.6. I don't know what you mean by PC Rating.
Click to expand...

As in PC4000 or PC3700 or whatever else. You could just go for some PC4000 that's gauranteed to run at 250fsb no matter what so you wouldnt have to worry about the memory clocks holding you back.
 
You must log in or register to reply here.
Back
Top