• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Can a degraded cpu run hotter than usual?

Ace

Ace

Ace

Associate
Joined
18 Oct 2002
Posts
288
I built a pc for a friend a little over a year ago, a i7-4770K @ 4.3ghz with H100i. It used to max out at 85c when encoding (less when gaming) for the year he has had it and suddenly in the last couple of weeks it has jumped from 85c max to hitting 100c and throttling.

I got him to remove the heatsink and reapply some mx4 which made no difference, I think it's probably the pump that is causing the issue although the pump is reporting an rpm in the corsair software so appears to be running.

Is there any chance that the cpu could have degraded and that's why the temps have shot up? I have heard of cpu's degrading with too much voltage (which this one didn't have) causing them to die or becoming unstable but never the temps shooting up?
 
In my experience CPUs either work 100% or they are completely dead. Regarding the degrading you'd probably just find that over many years (i.e. CPU is >10 years old) you find you lose stability at the higher overclocks and have to wind them down a bit to regain stability.

I would go with there is a problem with the cooling.
I'd say the thermal paste or poor contact between heat sink and CPU, but since you've checked that I would check the cooling. I have never used any form of water cooling but could there be an air lock, low fluid (leaking) or failing pump?

Unless you have put some absolutely ******* crazy voltage through the CPU then it won't have degraded :).
 
It would need more voltage for the same clocks if its degrading

Pumps and flow rate could probably degrade idk
 
There's always the TIM between the chip and the IHS.

That could degrade but I doubt it's that, most probably the cooler is failing.
 
As suggested above, for temps to go up that much suggests a cooling issue. If it's been running for a year then depending on the environment there could be a lot of dust/grime/fluff etc stuck in the radiator and on the fans and case grills. Try removing what bits you can and give them a good clean out with compressed air to make sure the airflow isn't being restricted.
 
If he's not been messing with any BIOS settings (voltages?) I'd look towards possible fault with the H100 too.

I don't know why people bother with AIO coolers, they aren't much better than high end air (if at all) and have so many additional things that can go wrong, pump failure, fluid issue, fan failure.. at least with air you've only got possible fan failure to worry about (and they're easily replaced).
 
Last edited:
McGraw said:
Is it possible top stick the stock cooler on and see what temps he gets?
Click to expand...
That would be ok for idle temps, I certainly wouldn't recommend running any encoding type program at stock on a haswell with the Intel cooler, let alone any form of oc.
 
setter said:
That would be ok for idle temps, I certainly wouldn't recommend running any encoding type program at stock on a haswell with the Intel cooler, let alone any form of oc.
Click to expand...

Surely the cooler isn't fit for purpose then or maybe intel is happier than us that these chips run hot.

Mine touched the 90s while encoding at stock/stock...surely you shouldn't have to upgrade your cooling at stock?
 
I used a stock cooler for one day when I moved to DC, most I done was browse the web. At idle the temps were in the 50's, brand new stock cooler so no bodged paste job on my part. Shudder to think what any heavy load would have been like. Fitted my k2 the next day, temps in the mid 20's, mid 50's at load when gaming. Just the nature of how hot theese chips are unless you scalp the lid off and repaste, or as I prefer to call it finish the job Intel forgot to do.:D
 
Ace said:
I built a pc for a friend a little over a year ago, a i7-4770K @ 4.3ghz with H100i. It used to max out at 85c when encoding (less when gaming) for the year he has had it and suddenly in the last couple of weeks it has jumped from 85c max to hitting 100c and throttling.

I got him to remove the heatsink and reapply some mx4 which made no difference, I think it's probably the pump that is causing the issue although the pump is reporting an rpm in the corsair software so appears to be running.

Is there any chance that the cpu could have degraded and that's why the temps have shot up? I have heard of cpu's degrading with too much voltage (which this one didn't have) causing them to die or becoming unstable but never the temps shooting up?
Click to expand...


No, semiconductor degradation will not cause a chip to operate at a high temperature as it has no bearing on it. However as you know heat can be a factor in degradation in the first place. Load, or malicious software may be causing increased temps, but at a hardware level it's not possible.

Degradation is measured at an end user level by the amount of current required for a particular frequency. If you need more for the same frequency as you were using prior, then this is a sign that the transistor degradation has suffered a push most likely through heat or voltage.


Laptop CPU are however more prone to this slew rate and more noticeable due to the lack of internal heat sinks, general operation cycles take slightly longer through what is refereed to as steady state leakage, causing the chip to run at an elevated temperature. It's common in old mobile processor. I've yet to see it on anything recent, least of all a Haswell desktop CPU.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom