scan wont admit theres a problem with my q6600

[Blitzen]

I just cant see the argument here?

I would be annoyed at those temperatures aswell............but its within the Intel Specified Limits.
Look at it like this for an example:
Scan accept your RMA.
They return it to Intel as a Quality Issue.
Intel say no as its within their limits.
Scan are left with a CPU that cant be resold and they have replaced yours.


Because of this, there is no recourse that the etailer is obliged to take. If anything is made good it will be a good will gesture and nothing more.

On a side note, i ran a Q6600 in an AW9D-MAX...the temperatures were what you are reporting as i dont think the bios flash recalibraed the temperature sensors.
I bought a different board and get the results i do now.

Hopw can you be sure its the CPU anyway? Does the HS/FAn feel hot to the touch? If not, you are barking up the wrong tree anyway.

[SiM]

From my understanding of this thread the temps are within spec at IDLE. If under load the CPU would shoot out of specs if cooled by the retail cooler. Hence, it is faulty imo.

Does the HS/FAn feel hot to the touch?

Do this. If it has a heatpipe, touch that. If it doesn't feel hot then the thermometer on the cpu is probably faulty (which means the whole cpu is faulty). I am assuming you have checked for concavity?

[Dreaming]

I just cant see the argument here?

I would be annoyed at those temperatures aswell............but its within the Intel Specified Limits.
Look at it like this for an example:
Scan accept your RMA.
They return it to Intel as a Quality Issue.
Intel say no as its within their limits.
Scan are left with a CPU that cant be resold and they have replaced yours.


Because of this, there is no recourse that the etailer is obliged to take. If anything is made good it will be a good will gesture and nothing more.

It depends how you measure it. On a warm day, with the stock cooling, it's not going to take much to puss it from idling at 60*c now up to past the 71*c threshold. Neither the OP nor Scan if I'm reading correctly have testing with the stock cooling, but I presume the specifications are only applicable when using the stock cooling...?

I'd also want to try it in another motherboard or double check with an external thermometer though, seems the OP is going to do that.

[moogle]

polly why not set your bios vcore at 1.35 and put the stock hsf on and stress it. If it goes past 71 then ask for a new one. Sure it can run at lower volts but the Intel spec page for G0 quads state 1.35 - 1.5v. My old AMD 4200 X2 was hitting 110c under load and all it took was 1 screenshot of the temperatures for AMD to take it in for rma.

[SiM]

polly why not set your bios vcore at 1.35 and put the stock hsf on and stress it. If it goes past 71 then ask for a new one. Sure it can run at lower volts but the Intel spec page for G0 quads state 1.35 - 1.5v. My old AMD 4200 X2 was hitting 110c under load and all it took was 1 screenshot of the temperatures for AMD to take it in for rma.

wow thats good customer service! they sent you a replacement 4600 x2 too, didn't they? SiM goes off to photoshop his way to a processor upgrade

[moogle]

wow thats good customer service! they sent you a replacement 4600 x2 too, didn't they? SiM goes off to photoshop his way to a processor upgrade

yes they did
Only took 1 week to get sent to the Netherlands, be tested and new one sent back
I'm pretty sure they'll test the claims SiM

[Saracen]

My opinion ..... you are all wasting your time comparing ideas of what thermal performance is expected and looking at that 71c figure for comparison.

To be valid for comparison, any thermal testing has to be done in accordance with strict metrology guidelines. Intel do have a (121 page ) document defining these guidelines, but you need to be looking at controlled ambient air temperature, case to ambient cooling, and so on, as well as measuring the temperature at a very specific location. I doubt anyone has the facilities to do what's needed at home, and I doubt Scan do either, unless they have a very well set up lab.

So, any comparisons with Intel-quoted temps are going to be approximate at best.

In my view .... the point about temps has been made. We now need to let Scan look at it. But at the end of the day, Blitzen is right - as long as the CPU is operating within spec, then regardless of how disappointing actual performance may be, there's not much to be done about it, and Scan aren't obliged to do anything. If they guaranteed a given level of overclocking or thermal performance it would be different. But they don't (that I'm aware of).

Best bet - wait for Scan to deal with this.

[Mossy]

i will make the technical department aware of this thread for Monday Saracen we do not guarantee any level of overclocking.

[polly69]

it was 67c within minutes last night just loading up programs ive tried it in 3 boards and the temps are the same i have the asus P5K and there is no problem with the sensors as its reading the E4400 at 34c with the akasa revo cooling i couldnt use this to cool the quad as it was not good enough even though scan recommened it. I had to drop the voltage to 1.2 as it would have gone over 71c today is very warm and muggy the ambient temp would have pushed it over spec no problem at all i had to take it out.

Ive not overclocked it even though scans site seem to suggest that you can and should i had intended to do that thats why i have such cooling and the case and for one of scans tech to suggest that maybe i dont know how to mount a fan thats unbeliveble. even with scans test 14c under the 71c spec is way to close for comfort it only takes a warm day or a small media case and that is 71c in use and hence faulty.

Im glad everyone agrees that there seems to be a problem with this cpu you only have to read all the forums and threads to see that normal temps are half of this with good cooling.

As a answer to the quetion yes the HSF feels warm and i cant see any concave on the surface of the cpu.

[Noxvayl]

From everything I have read in this thread polly69 has done the following:
1. Tested the CPU in multiple boards to make sure its not a sensor issue
2. Used different heatsinks to make sure the heatsink isn't the problem
3. Stressed the CPU for a few minutes only to get his high temps(I'm sure intel's spec temps are for hours of stress testing not only a few minutes)
4. Lowered the voltage in order to lower temps(shouldn't be necessary!)

All that done at stock frequency and with environments that are perfectly satisfactory for his other 2 desktop PC's which are running at lower than average temps.

There is no doubt in my mind that the CPU has been properly tested with multiple systems to safely deduce that the CPU is in fact the problem and nothing else.

What I would do to get really ridiculous reeding's is to put the voltage up to 1.35v as its meant to be, stick the Intel HSF stock cooler on and then proceed to run Orthos(or whichever test you feel is best) for 5-10 hours and be sure to do a recording of the temps for the entire time. From what polly69 has said I think this would actually kill the CPU or force it to shut down from overheating but then no-one believes you so go ahead and break the CPU to prove to them that the CPU is the problem because they don't seem to want to listen otherwise.

The only reason the temps have not exceeded 71C from my understanding is because polly69 has stopped the test when the temps got that high and because he has been using after market coolers which are MUCH better than the Intel HSF.

[moogle]

What I would do to get really ridiculous reeding's is to put the voltage up to 1.35v as its meant to be, stick the Intel HSF stock cooler on and then proceed to run Orthos(or whichever test you feel is best) for 5-10 hours and be sure to do a recording of the temps for the entire time. From what polly69 has said I think this would actually kill the CPU or force it to shut down from overheating but then no-one believes you so go ahead and break the CPU to prove to them that the CPU is the problem because they don't seem to want to listen otherwise.

My thoughts exactly.

[Saracen]

Saracen we do not guarantee any level of overclocking.

I thought that was the case, but thanks for confirming it.

.... Im glad everyone agrees that there seems to be a problem with this cpu you only have to read all the forums and threads to see that normal temps are half of this with good cooling.

Everyone doesn't agree that there's a problem. For a start, it depends what you mean by "problem".

Is the CPU meeting your expectations? Clearly not, so it's problem in that sense.

Are those expectations of probably performance reasonable? They seem so.

Are processors guaranteed to meet those expectations? Nope - only to meet Intel's spec.

If a CPU is within that Intel spec, but outside your expectations, is it a problem? Only in that your expectations exceed what you can guarantee to get when you buy the chip.

If the CPU performs as you describe, is it well below what most get get? Yup.

Are you very unlucky? Looks like it, on this occasion.

Did you get what you paid for? If it performs within spec, yes, you did.

What I would do to get really ridiculous reeding's is to put the voltage up to 1.35v as its meant to be, stick the Intel HSF stock cooler on and then proceed to run Orthos(or whichever test you feel is best) for 5-10 hours and be sure to do a recording of the temps for the entire time. From what polly69 has said I think this would actually kill the CPU or force it to shut down from overheating but then no-one believes you so go ahead and break the CPU to prove to them that the CPU is the problem because they don't seem to want to listen otherwise.

I would strongly advise against that.

If you do that and send a cooked chip back, any supplier may conclude that you cooked it deliberately. Perhaps left the heatsink off. Or just pointed a heater at the PC to ensure in ran outside temp limits. The problem is, if you do that there is then no way for anyone (such as Intel) to test the chip to see if it's running outside spec whilst still inside the ambient profile.

And that's especially true if it happens after being given that advice on here.

Paul has said Scan technical will look at this thread on Monday. Until then, and we know what they conclude, anything else is just looking for problems. Nothing much can be done until Scan have looked at this, which won't happen 'til Monday..

[Dreaming]

Best of luck to polly + scan obviously scan is just the retailier and it's Intel who decides what's ok but I think all can agree it's less than satisfactory to have a q6600 idling in the 60s I'd be gutted personally and would kick up a stink with intel. (fwiw I got screwed myself with the x1950 pro vrm overheating, sapphire refused to acknowledge the problem despite many people testing it and proving the fault, they said it was because you need 800w+ PSUs only since it was a high end graphics card. Funny that they rectified the fault in the second revision but I never got my money back - £120 down the drain)

[Noxvayl]

I would strongly advise against that.

If you do that and send a cooked chip back, any supplier may conclude that you cooked it deliberately. Perhaps left the heatsink off. Or just pointed a heater at the PC to ensure in ran outside temp limits. The problem is, if you do that there is then no way for anyone (such as Intel) to test the chip to see if it's running outside spec whilst still inside the ambient profile.

And that's especially true if it happens after being given that advice on here.

Paul has said Scan technical will look at this thread on Monday. Until then, and we know what they conclude, anything else is just looking for problems. Nothing much can be done until Scan have looked at this, which won't happen 'til Monday..

Its not like he is running the CPU at higher clocks and over volting it. He would only be running it in the manner any average customer would with the exception of testing it with a stress test(which is perfectly acceptable). I also specifically requested recorded monitoring for proof of the fact that the CPU was run under normal conditions if anything should happen to it so he shouldn't have any issues when sending it back given that he is a certified computer technician.

Polly69 has the proof for his findings and the qualifications to back it up so there shouldn't be any problems.

[Steve B]

Polly69 has the proof for his findings and the qualifications to back it up so there shouldn't be any problems.

Shouldn't be any problems what exactly?

I stopped caring about this thread after this:

...and the best of all is im a fully qualified pc/network tech who know more than they do.

[Wiczy]

May I draw your attention to the quote above your avatar

Better to be pissed off than pissed on

Or does that only apply to you?