• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Intel CPU's have OC black box

D_o_S

Moderator
Joined
Aug 27, 2005
Messages
954 (0.13/day)
Processor AMD Opteron 144
Motherboard DFI Lanparty nF4 SLI-DR Expert
Cooling Watercooled - Alphacool Nexxxos XP, BIX2, Eheim HPPS
Memory 2x 512MB OCZ PC 3200EL Platinum Rev.2 (TCCD)
Video Card(s) 2x Gainward GeForce 6800 Ultra (430/1200)
Storage 4x WD Raptor 740GD in RAID 0
Display(s) Eizo FlexScan L768
Case CM Stacker
Audio Device(s) Audigy 2
Power Supply Silverstone Zeus ST65ZF
A few big retailers and etailers have confirmed that Intel actually can check if you overclocked your CPU. When you burn your CPU Intel asks its retailers and etailers to return the CPU back to its factory. Intel then reads data from a hidden part of the CPU, and instantly finds out if the CPU has been overclocked or overvolted.

Retailers and etailers said that they are not sure about Intel's methods of judging who gets the new CPU, but they said that Intel gets back to them if you overclocked too much, and simply refuses to RMA the part.

View at TechPowerUp Main Site
 
Interesting ,when they themselves promote the new C2D and new penryn chips for enthusiasts, ocers and gamers.
 
not a big surprise really, i have been saying this was coming for years, intel shouldn't have to replace parts that you burnt out overclocking

though if you do a good enough job burning it out, they won't be able to read it anyway now will they?
 
lol so if ya brun it.. you might as well break out the blow torch and go to town lol.. i'd like to see them recover that.. :laugh:
 
Muaahh ,
Intel say this so many times on P4 Times :laugh: they can nothing see :roll:
 
well than i guess when it starts to get unstable and the temp gets unbearable its time for a 5ghz suicide on air then?
 
I think this is fair enough, providing that the administration behind it doesn't cost more than they pay out for replacing CPUs anyway. If you broke it, you should pay for it.
 
it'd be nice if the golden rule worked.. i wonder how much this added to the price of the processor.. I'm not the type to rip people off at all.. but being price jacked for everyone elses juvenile behavior is also not very cool.
 
Interesting ,when they themselves promote the new C2D and new penryn chips for enthusiasts, ocers and gamers.

Show me where Intel promotes them for overclockers.
 
Good thing CPU's almost never die from overclocking.
 
i agree but lets keep the sarcastic comments to a minimum.
 
Honestly, this is fair enough. If I overclocked a CPU to the point of causing damage, then it is my own fault. If I install a CPU correctly, run it at stock speeds, and it still has problems, then I would expect Intel or the retailer to replace it. What I want to know is what happens when actual damage is done to the CPU... like pulling the heatsink out and having the CPU stick to the heatsink, leaving some pins in the socket??? I did have the CPU stick to the heatsink once, but I was lucky and none of the pins were left behind or bent. :) In a case like that, I feel it is partially my fault... but not entirely. The old lidless CPUs rarely had that problem because there was not enough surface area making contact with the heatsink. All the new CPUs have lids, and I have not figured out a way to easily remove the heatsink.
 
Honestly, this is fair enough. If I overclocked a CPU to the point of causing damage, then it is my own fault. If I install a CPU correctly, run it at stock speeds, and it still has problems, then I would expect Intel or the retailer to replace it. What I want to know is what happens when actual damage is done to the CPU... like pulling the heatsink out and having the CPU stick to the heatsink, leaving some pins in the socket??? I did have the CPU stick to the heatsink once, but I was lucky and none of the pins were left behind or bent. :) In a case like that, I feel it is partially my fault... but not entirely. The old lidless CPUs rarely had that problem because there was not enough surface area making contact with the heatsink. All the new CPUs have lids, and I have not figured out a way to easily remove the heatsink.
That's why you twist a heatsink, not pull.
 
i thought if you twist you ended up with a scratched chip?
 
What I want to know is what happens when actual damage is done to the CPU... like pulling the heatsink out and having the CPU stick to the heatsink, leaving some pins in the socket??? I did have the CPU stick to the heatsink once, but I was lucky and none of the pins were left behind or bent. :) In a case like that, I feel it is partially my fault... but not entirely. The old lidless CPUs rarely had that problem because there was not enough surface area making contact with the heatsink. All the new CPUs have lids, and I have not figured out a way to easily remove the heatsink.

That's why you twist the cooler out, not pull it up. Suction force on A64s is already crazy with AS5 at least, but at least you know the contact was great, if you can lift the whole computer up from the heatsink xD Once I go to s775, might just do a CPU burn test first and then shut off and straight taking cooler out, as I need my Zalman to Core 2 Duo. Not so fun twistin cnps9500 too long as it's so sharp you ed up hurting your self.

edit: overcast was faster :)
 
My guess is it has nothing to do with clocks.


It's got a couple of voltage triggers (could be physically destroyed by too much or something) and a temperature sensor.


That way if you're running way beyond spec and heat kills it (OCing without proper cooling) you can immediately tell and read the data at any time barring massive physical damage to the CPU (which would be denied anyways).
 
That's why you twist the cooler out, not pull it up. Suction force on A64s is already crazy with AS5 at least, but at least you know the contact was great, if you can lift the whole computer up from the heatsink xD Once I go to s775, might just do a CPU burn test first and then shut off and straight taking cooler out, as I need my Zalman to Core 2 Duo. Not so fun twistin cnps9500 too long as it's so sharp you ed up hurting your self.

edit: overcast was faster :)

Made the same mistake myself the first time I took the HS off my A64. Both HS and CPU came off at once, and then when I prised it clear (it was stuck really firm) it flew across the room and hit my curtains before falling to the floor :eek:

Luckily it still works fine to this day, no pins damaged (that was almost 2 years ago), and I've learned to always twist the HS - and that A64s can take a beating!
 
Made the same mistake myself the first time I took the HS off my A64. Both HS and CPU came off at once, and then when I prised it clear (it was stuck really firm) it flew across the room and hit my curtains before falling to the floor :eek:

Funniest thing I've read in a long time :D Good that it works still, your heart might have given you couple missed beats :)
 
i thought if you twist you ended up with a scratched chip?
Don't twist it like an animal then? And even if you do scratch it up a little bit who cares. If you're that paranoid, take some Mothers Mag polish and buff it out. This is electronic equipment, you shouldn't be wrenching anything in or out.
 
That's why you twist a heatsink, not pull.

Okay... now I am going to have to take advantage of the knowledge you all have. :)

I guess I never really looked at sites when it comes to installing the heatsink, it seemed rather obvious to me. I do not use the high end CPU coolers... just the basic one that comes with the AMD (socket 939 or 754) CPU. Once installed, I am only able to turn the heatsink about .5 cm because the plastic support attached to the motherboard gets in the way. It also prevents me from sliding a credit card or something like that in between the CPU and heatsink to break the seal. It seems as though the only option I have is to slowly pry it up.

What am I doing wrong? Any help with this would be greatly appreciated!!! :respect:
 
Last edited:
good! why should intel have to replace your CPU when you were too dumb to overclock it correctly.
 
can they also detect undervolting? or not, because your cpu undervolts itself when it drops mulitpliers to save energy...hmm i wonder.
 
undervolting isn't likely to damage the cpu though
 
Back
Top