Just built my Conroe System - crazy temps!

[evilr00t]

Version 2.15's been out for a while...
http://cpu.rightmark.org/download/rmclock_215_bin.exe

 

[suzukii]

<RMclock v2.15>

AaaHaaaa!!!! Got it, installed it, Tanx.

 

[levicki]

red alert red alert.........
did you said you put a drop of arctic silver on the center of cpu, and then mount heatsink?? what about step 2, spread the arctic silver around the whole surface.
the entire surface of the chip = equal all the shiny shiny part. the silicon part??? use your credit card and spread a thin uniform surface.

if you have only one drop, your heatsink will spread only a small amount of total silicon surface. contact between the heatsink and cpu will be very bad.

Why there are so many smartass noobz here?!?

Haven't your mother told you to be quiet if you don't have anything smart to say?!?

Check the step #9 of those instructions:
http://www.arcticsilver.com/arctic_silver_instructions.htm

Only a small amount of Arctic Silver is needed

P4- About the size of an uncooked grain of short-grain white rice or 1/2 of a BB.

RECHECK to make sure no foreign contaminants are present on either the bottom of the heatsink or the top of the CPU core. Mount the heatsink on the CPU per the heatsink's instructions. Be sure to lower the heatsink straight down onto the CPU.

Once the heatsink is properly mounted, grasp the heatsink and very gently twist it slightly clockwise and counterclockwise one time each if possible. (Just one or two degrees or so.)

Our testing has shown that this method minimizes the possibility of air bubbles and voids in the thermal interface between the heat spreader and the heatsink. Since the vast majority of the heat from the core travels directly through the heat spreader, it is more important to have a good interface directly above the actual CPU core than it is to have the heat spreader covered with compound from corner to corner.

 

[levicki]

Version 2.15's been out for a while...
http://cpu.rightmark.org/download/rmclock_215_bin.exe

This thing doesn't work properly.

It shows 25% load on core 0 here while Task Manager shows between 0-3% on both cores and it also shows 56°C which is plain wrong.

 

[evilr00t]

56C is 15 degrees off, which suspiciously sounds like they messed up their calculations for the core temperature (ie. Merom's Tcasemax is 100C, which is 15 degrees higher). However, the CPU use is probably correct. The OS may not be using it (thus the low utilization in the process manager), but the CPU is busy managing hardware interrupts (or something like that). See the pic I sent you: 14% CPU use according to the CPU, and 1% CPU use according to the OS.

 

[Mondoman]

levicki and the others who say you haven't installed the hsf properly are correct.
The Intel LGA775 hsf is notoriously easy to not-quite-fully install. In order to properly install the 4 pins all the way through the holes in the MB, you need to *remove the MB from the case.* My guess is that every time you have removed and reseated the hsf, it's been with the MB still in the case. Take the MB, CPU, hsf, RAM out of the case, reinstall the hsf and make sure the pins are all the way through the holes.
Then, do a "bench test" with the MB outside the case on a non-conductive surface, installing the video card and connecting the power cables. Your CPU temps in the BIOS should be in the 30s or lower.

 

[levicki]

56C is 15 degrees off, which suspiciously sounds like they messed up their calculations for the core temperature (ie. Merom's Tcasemax is 100C, which is 15 degrees higher)

That must be the case since the Zalman is completely cold to the touch.

However, the CPU use is probably correct.

No way.

Checked with Process Explorer too, and it shows the same thing as Windows Task Manager.

Moreover, if I had anything taking 25%+ of one core all the time I would know it and the temperature for that core would be higher. As you can see from both the CoreTemp and Everest that is not the case -- both cores show the same temperature so RMClock is broken.

Anyway, get SysInternals Process Explorer and see for yourself.

 

[levicki]

you need to *remove the MB from the case.*

I wouldn't advise that.

First, mainboard is mounted on standoffs so there is enough room to push the pins completely in.

Second and more important, when the mainboard is fixed inside the case there is less chance to damage it by pushing the pins through then if you do it in free air.

Again I suggest removing the HSF, cleaning both HSF and CPU heat spreader with pure alcohol and lint free cloth then applying the AS5 as per instructions above and pushing the pins two by two diagonally as shown in the isntructions that came with the CPU and the mainboard. Remember, you must reset the notches (by turning all four clockwise as far as they'll go) before remounting HSF, and you must push them until you hear each one click in place.

You have to do it because running out of spec can damage the CPU.

 

[evilr00t]

:roll:
RMClock uses CPU-specific performance counters to get CPU utilization.
The other utilities ask Windows.

 

[octop8]

That's wrong. The pins LOCK when you push them in, and you hear a loud "pop".
You just push the pins in. To take it out, you turn the pins counterclockwise, and pull up.

Examine the push-pin assembly yourself; turned completely CCW, the pins have nothing holding them from going back up. Turned completely CW, the pins have a ridge preventing movement from a pushed-in position to a pushed-out position.

Who the hell are you to tell me I am wrong? I INSTALLED an E6600 myself on my own system and that is what I did. Push the pins in and use a screwdriver to turn it clockwise roughly 1/4-1/2 a turn. I am sharing my experience trying to help here with my FIRST-HAND experience and there is no need for me to take any sh*t antagonism from you. I did exactly the opposite of what you are saying and it works perfectly for me. I pushed the fasteners in and used a screwdriver to turn the indent perpendicular to the HSF (and not parallel as you stated). If you examine the fasteners closely, you will see a hoove that will lock the fastener when you turn the indent perpendicular to the HSF. When it is parallel, this little hoove does not lock the fastener - hanging off the side as most cases do, the HSF may not have a full contact.

I read the instructions that come with the CPU and I know the instructions say it just pops in and leave it parallel. But guess what? There is a reason that people criticise the LGA775 heatsink installation - it just doesn't work well, and I have found that you need to turn it clockwise with a screwdriver to fasten it in place. My E6600 works perfectly - the first time I just popped it in, it didn't contact properly and I was getting 50+C idle. Then I examined the fasteners very closely. Have YOU installed a Conroe chip?

 

[13thmonkey]

In case you or anyone needed the link from ASUS for the P5B mobo BIOS update released 09/08/2006:
Verison 0509
ftp://dlsvr03.asus.com/pub/ASUS/mb/socket775/P5B/P5B-0509.zip

Thanks Suzukii, I couldn't find the FTP site to link to it earlier. Not having an Asus board (nearly) I'm not familar with their site.

 

[Kronicide]

Alright, I'm going to install the Intel HSF one more time before recieving the Zalman I just ordered.

Now, from what I got so far, I should keep the mobo in the case, because I don't know how the notches would go through on a solid surface,
clean the HS and CPU surface with alcohol, then put a dab of AS5 in the middle (which is what I've been doing)
and place the HSF on making sure each click in place...

I know I've been snapping it into place each time... I would break the Mobo if I pushed any harder.

But I am still going to clean the heatsink and cpu surface, so maybe that will do something.

 

[sluzbenik]

Kronciide - I had the exact same issue. See this post.I remounted a half-dozen times, tried the various ways to apply Arctic Silver (single drop in the middle and thin spread) and it did not work. I also had your exact same temps. I idled only at 51-52c.

I wrote to Intel and complained, they responded with a bunch of links to their white papers. After I got the Arctic 7 I wrote back and said my temps have dropped significantly with a third party cooler - they have not responded.

I second the above poster's recommendation to get an Arctic 7 Freezer Pro. This worked fine for me on the first mount. And you don't have to scrape off the thermal grease because it is apparently high quality. I can idle at 37c if the room is cool enough.

After having remounted so many times already, you have probably stressed the plastic clips enough that getting a good mount is doubtful.

Just remember to turn the clips clockwise with a screwdriver before you mount. I would not follow the other poster's recommendation to turn the screws clockwise afterwards. The LGA 775 clips are made to click in and turn by themselves, if you try to mount with them in the unlocked position and then lock you will put a lot of stress on the clips and without the pressure from above to get them in place I don't see how you can get a good mount.

Mount diagnonally as well, top-left then bottom-right, top-right to bottom-left. This puts less pressure on one part of the motherboard and helps in getting a good mount.

 

[Kronicide]

Thanks for the replies, I'm still going to re-mount the HSF, but I am planning on getting a better result from the Zalman HSF I ordered.

 

[evilr00t]

Who the hell are you to tell me I am wrong? I INSTALLED an E6600 myself on my own system and that is what I did. Push the pins in and use a screwdriver to turn it clockwise roughly 1/4-1/2 a turn. I am sharing my experience trying to help here with my FIRST-HAND experience and there is no need for me to take any sh*t antagonism from you. I did exactly the opposite of what you are saying and it works perfectly for me. I pushed the fasteners in and used a screwdriver to turn the indent perpendicular to the HSF (and not parallel as you stated). If you examine the fasteners closely, you will see a hoove that will lock the fastener when you turn the indent perpendicular to the HSF. When it is parallel, this little hoove does not lock the fastener - hanging off the side as most cases do, the HSF may not have a full contact.

I read the instructions that come with the CPU and I know the instructions say it just pops in and leave it parallel. But guess what? There is a reason that people criticise the LGA775 heatsink installation - it just doesn't work well, and I have found that you need to turn it clockwise with a screwdriver to fasten it in place. My E6600 works perfectly - the first time I just popped it in, it didn't contact properly and I was getting 50+C idle. Then I examined the fasteners very closely. Have YOU installed a Conroe chip?

Maybe because I installed my LGA775 cooler on my Celeron D?

Let me begin by using stuff you wrote:

I am sharing my experience trying to help here with my FIRST-HAND experience and there is no need for me to take any sh*t antagonism from you.

I did exactly the opposite of what you are saying and it works perfectly for me.

I think we're agreeing on the same things worded differently:

If you examine the fasteners closely, you will see a hoove that will lock the fastener when you turn the indent perpendicular to the HSF. When it is parallel, this little hoove does not lock the fastener - hanging off the side as most cases do, the HSF may not have a full contact.

is the same as

Examine the push-pin assembly yourself; turned completely CCW, the pins have nothing holding them from going back up. Turned completely CW, the pins have a ridge preventing movement from a pushed-in position to a pushed-out position.

There is a reason that people criticise the LGA775 heatsink installation - it just doesn't work well, and I have found that you need to turn it clockwise with a screwdriver to fasten it in place.

Turning it clockwise after pushing it in? That's redundant (since it won't turn at all if properly set beforehand - ALREADY turned fully CW), but if that floats your boat, go for it.

Turned completely CW, the pins have a ridge preventing movement from a pushed-in position to a pushed-out position.

That's right... I forgot to state that when the pin assembly is completely turned CW, the ridge does NOT prevent movement from a pushed-out position to a pushed in position. That's why:

The pins LOCK when you push them in, and you hear a loud "pop".
You just push the pins in.

It seems like you push the pins in without the pins locked, then turn it so that the lock engages. When the pins are locked, you cannot pull up on them and have them disengage from the board.

What I did:
1. Turned pins completely CW
2. Pushed them in

What you did:
1. Turned pins completely CCW
2. Pushed them in
3. Turned pins completely CW :roll:

Have YOU installed a Conroe chip?

Obviously, I don't need to.

 

[13thmonkey]

How are you getting on and can you confirm that the 2.24 Voltage for Vcore was a typo?

Also I'mm seeing a lot of P5B issues on the forums, is it possible to RMA it? it might save you a heap of pain later.

 

[Kronicide]

The Vcore I posted was wrong, it really runs at about 1.20 or so.

As for teh temps, they're at about 50-52 idle, and about 55-58 while playing a game such as CS:S.

I will call Intel, then ASUS on Monday to see what they say about the temps.

 

[levicki]

:roll:
RMClock uses CPU-specific performance counters to get CPU utilization.
The other utilities ask Windows.

Then it doesn't read them right because constant 25% load on one core would register as higher temperature of that core.

Who the hell are you to tell me I am wrong?

He is Mr. Right this time and you are arrogant dumb asshole. According to you, Intel intentionally put the wrong instructions with the CPU. So you are smarter than the guys who designed the LGA775 socket, right? Give me a break! You probably got it right by accident.

As for me, I owned Pentium 520, 630, 930 and now Core 2 Duo E6300 and I put many computers together from parts for other people, got paid for that and received no complaints so you are clearly the last person on this fucked up planet to teach me or anyone else how to deal with LGA775 socket properly.

 

[Artmic]

To answer all of your quetions, yes the CPU fan and case fans are spinning just fine, and I have the stock intel HSF. I used about a half of a grain of rice droplet for the AS.


Oh and just now, I turned on the PC and it started at 70c and now its working its way up to a steady 90c

Did you spread the AS5 on the cpu? or you left the drop on the die?
I put AS5 and spread the sucka on the die with my finger (rubber glove).
and i applied some to the SI-128 heatsink.

 

[Kronicide]

To answer all of your quetions, yes the CPU fan and case fans are spinning just fine, and I have the stock intel HSF. I used about a half of a grain of rice droplet for the AS.


Oh and just now, I turned on the PC and it started at 70c and now its working its way up to a steady 90c

Did you spread the AS5 on the cpu? or you left the drop on the die?
I put AS5 and spread the sucka on the die with my finger (rubber glove).
and i applied some to the SI-128 heatsink.


Yese actually, I spread the AS5 with a razor blade and the results have been maybe a couple degrees C.

Now, when I use PcProbe II, it shows idle temps at like 50. But when I use Core Temp (for conroes), it shows about 39-40.

I think that Core Temp may be almost correct to my real temps, but hopefully its off by a bit

 

[levicki]

Now, when I use PcProbe II, it shows idle temps at like 50. But when I use Core Temp (for conroes), it shows about 39-40.

I think that Core Temp may be almost correct to my real temps, but hopefully its off by a bit

PCProbe is junk. Core Temp seems to be correct. If you are down to 40 you are ok then.

 

[ma2ha3]

Instruction are written by homo sapien. Homo sapien have different views and different ways of doing the same thing. Homo sapien make mistake too.

I have no idea, how would spreading the AS5 very thinly can result in air bubbles.

By pushing the AS5 down and twisting, you cannot see the gaps and air bubbles, you never know it is good.

The AS5 only use to fill the micrometer gap of both heatsink and heatspreader, gap that human eye cannot see. it should be as thin as possible and evenly spread for best result.

Heatspreader spread the heat around, so with no paste at the edge, would reduce and not improve the heat dissipation.

putting the paste in the middle of cpu and spread using heatsink force, sound like instruction for noobs. Kind of remind me of those china apple operator paste job. real pro spread it thin and even.

 

[Kronicide]

Now, when I use PcProbe II, it shows idle temps at like 50. But when I use Core Temp (for conroes), it shows about 39-40.

I think that Core Temp may be almost correct to my real temps, but hopefully its off by a bit

PCProbe is junk. Core Temp seems to be correct. If you are down to 40 you are ok then.

Ok, thanks. Now that my CPU is finally somewhat stable, do you think any damage was done?

Or should I RMA my Mobo or something?

 

[Granite3]

Have you called Asus yet?

Your MB temps are about 9-10 degrees above mine. My board (same as yours) is posting 31-37 depending on the load. Averages 33 or so.

Asus has bench techs available via phone, but most are not up to speed Conroe wise yet, if my 3 calls were any indication.

After the bios update, both core and MB temps were down by more than 10 degrees each.

One other question, I am assuming you loaded all the drivers from the cd?

 

[octop8]

He is Mr. Right this time and you are arrogant dumb *******. According to you, Intel intentionally put the wrong instructions with the CPU. So you are smarter than the guys who designed the LGA775 socket, right? Give me a break! You probably got it right by accident.

As for me, I owned Pentium 520, 630, 930 and now Core 2 Duo E6300 and I put many computers together from parts for other people, got paid for that and received no complaints so you are clearly the last person on this ****** up planet to teach me or anyone else how to deal with LGA775 socket properly.

Where did I say that Intel intentionally put wrong instructions? "Got it right by accident"? You are so presumptious - don't make so many assumptions, you know nothing about me. I probably snapped and used too strong language because I was upset by his antagonistic tone. Yes, I read the instructions but I installed this way having tried the official way many times and had trouble with it so I am sharing my experience here trying to help to the poster. I have no intention whatsoever of teaching you, the high and mighty professional Intel master, how to do anything. I have mainly built AMD systems in the past and this is indeed my first Intel compared to you who gets "paid for that and received no complaints". But I don't want to hijack this thread, good luck to Kronicide, I hope your system works out.