Intel Warns That 12th Gen Non-K CPUs Aren’t Designed for Overclocking

Tsing

The FPS Review
Staff member
Joined
May 6, 2019
Messages
12,595
Points
113
intel-12th-gen-core-processor-bottom-1024x576.jpg
Image: Intel



Intel has sent out a statement clarifying that its non-K Alder Lake CPUs are not meant for overclocking purposes.



The statement, which was presumably prompted by der8auer’s recent successes in overclocking 12th Gen Intel Core non-K processors on select motherboards, warns that doing so could result in damage to a CPU or a reduction in its lifespan. These problems are not covered by warranty, as Intel points out.



“Intel’s 12th Gen non-K processors were not designed for overclocking,” the statement reads. “Intel does not warranty the operation of processors beyond their specifications. Altering clock frequency or voltage may damage or reduce the useful life of the processor and other system components, and may reduce system stability and performance.”



That said, der8auer has already shared multiple videos that have piqued the interest of Intel...

Continue reading...


 
Are any really designed for overclocking?

I mean, if it were "designed" for it, wouldn't they just boost that to the stock speed.

This is strictly Intel wanting you to pay more for an unlocked processor, without going to the length of forcing the motherboard manufacturers to disable BCLK overclocking (again)
 
Are any really designed for overclocking?

I mean, if it were "designed" for it, wouldn't they just boost that to the stock speed.

This is strictly Intel wanting you to pay more for an unlocked processor, without going to the length of forcing the motherboard manufacturers to disable BCLK overclocking (again)
Well, actually they do. Outside of LN2 overclocking, I haven't really seen any recent Intel CPU's that can go past their rated boost clock speeds on all cores if that. These CPU's are being run at the edge of what the silicon can do. The only benefit to all core overclocking is locking in your speed at all times, but even so it comes at a cost of increased heat output and power consumption.
 
Well, actually they do. Outside of LN2 overclocking, I haven't really seen any recent Intel CPU's that can go past their rated boost clock speeds on all cores if that. These CPU's are being run at the edge of what the silicon can do. The only benefit to all core overclocking is locking in your speed at all times, but even so it comes at a cost of increased heat output and power consumption.
I consider Boost and other factory-supplied algorithms to be “stock”, since it comes equipped and enabled by the factory default settings.
 
I consider Boost and other factory-supplied algorithms to be “stock”, since it comes equipped and enabled by the factory default settings.
That's what I mean. You aren't really able to overclock modern Intel CPU's past their boost frequencies.
 
No CPU was designed for overclocking.

Overclocking means running the CPU outside of official specifications.

I'm sure no one at Intel intended me to run my 486 sx25 at 50Mhz, bit I did, and it worked great. (To this date it is still my only 100% overclock.)

At some point Intel and AMD decided to multupliet lock the chips to prevent people from buying cheap CPU's and having them perform like high end CPU's, thus cutting into high end CPU sales.

Heck, my motivation for the first 10 years or so of overclocking was not to get the most out of a top end CPU, but rather to get better performance out of the limited hardware I could afford.

The Duron I bought in 2000 was cheap. I got the second from bottom model, the 650Mhz version, because it was what I could afford. I ran that thing at 950Mhz for almost the entire time I had it.

This is all about trying to protect sales of high end chips, and has nothing to do with warning about damage or anything like that.
 
This is exactly how intel responded last time Asrock added overclocking for non-K CPUs. It might even had been this exact headline.
And a few weeks later they pressured them to remove the option. As soon as they feel back in the saddle they go back to their old ways.
 
Then what do you think Intel is referring to?
I think you are over thinking this. Yes, overclocking is pushing past the limits. It's just that with a 12900K, you aren't going to be able to do much more than make them run at their maximum boost frequency on more cores than it might ordinarily run at that frequency, for longer than it normally would. You can also increase the E-core clocks a bit as well.

You can get more out of them, but not a lot given that they are binned so close to the edge of what the silicon can do. At least, not with ambient cooling. But, a 12400 is a different animal than the 12900K. It maxes out at 4.4GHz boost clock. Pushing that to 5.2GHz is possible as you can see in the Der8bauer video. That's definitely overclocking and the largest increase I've seen on a CPU in several generations.

But, as the article says, Intel's not referring to the 12900K or any K series CPU. It's referring to the non-K parts which tend to have slightly lower clock speeds and do not allow for multiplier based overclocking. So if we are talking about those, traditional overclocking is exactly what you are getting as they will boost higher than their maximum rated clock speeds under boost conditions. However, you are likely only going to see the biggest gains out of CPU's like the 12400 which have a decent amount of headroom. The 12600K didn't overclock as far in the video and CPU's like the 12900 non-K aren't likely to do much at all.

Intel's saying "non-K CPU's aren't designed for overclocking" for a couple of reasons. They do not want to undercut sales of the higher end CPU's. They also recognize that the non-K parts are likely not the best examples of the silicon and overclocking them carries certain risks as it always has. Taking a 12400 non-K and pushing it from 4.4GHz to 5.2GHz is a significant jump. If you over estimate your cooling or your particular CPU requires a lot of voltage, you may degrade the silicon or outright kill them.

That being said, it's early days with these CPU's. We have no idea how well these will handle that kind of overclocking over a long period of time.
 
Intel's saying "non-K CPU's aren't designed for overclocking" for a couple of reasons. They do not want to undercut sales of the higher end CPU's
This was all I was saying. I never made it any more complicated than that right there - money.
 
This was all I was saying. I never made it any more complicated than that right there - money.
It's not just money, it's also reliability, if the chip could handle more they would sell it as such and make more money.

Now some inexperienced user is going to see what Debauer does, and try the same and then start complaining that his 12400 can't do what Derbauers can, what he does not know is that Derbauer picked one specific CPU out of a whole bunch and found 1 or 2 that can go this high.

Those guys from silicon lottery did not stop selling CPU's for no reason, there is no headroom left.
 
It's not just money, it's also reliability, if the chip could handle more they would sell it as such and make more money.
This isn't always the case. Market segmentation is about getting more money. If there were no cheaper CPU's to sell, then people would either run what they have for a lot longer or buy used. Intel needs cheaper CPU's to make up the volume. Intel would sell fewer CPU's overall if all it offered were high end parts. That's why CPU's like the 12400 and even cheaper ones exist in the first place. It's a balancing act to get the most money possible.
 
Last edited:
It's not just money, it's also reliability, if the chip could handle more they would sell it as such and make more money.

Now some inexperienced user is going to see what Debauer does, and try the same and then start complaining that his 12400 can't do what Derbauers can, what he does not know is that Derbauer picked one specific CPU out of a whole bunch and found 1 or 2 that can go this high.
That's always been the case with overclocking
 
Become a Patron!
Back
Top