We tested Intel’s 12th generation Core overclocking tool with the new Raptor Lake platform and were surprised to see that the results are… pretty boring
Sometimes the end users get to see aids that were developed with the aim of improving system performance, even if in somewhat dubious ways such as removing the cover of processors by inserting the processor into a device and applying power with a key or screwdriver.
Shortly after the launch of Intel’s 12th generation Core, many in the overclocking community found that due to the way Intel built the processor mount, closing the processor exerts great pressure on the processor which actually creates an arc in its shape and it “gets a belly”, something that does not particularly contribute to heat dissipation.
As a result, products were born as you see here in the picture. This is the “Corrector Frame” made by Thermalright, a company that is mostly known for the heatsinks it produces. It is important to note that the manufacturer of overclocking solutions Thermal Grizzly was the pioneer of the idea.
The idea behind this product is simple – replacement of the standard closing mechanism for the processor holder in a frame that holds the processor more securely and evenly on both sides. The cost of this product was $5, and it was ordered directly from AliExpress. the screws? The same ones that come with the original mount, using the same original back plate that comes with the motherboard, this is the part that catches the mechanism. All we did was screw them out using the screwdriver that comes in the kit, and screw the frame in place.
It is important at this stage to note that Thermalright’s kit does not come with any mechanism that prevents the user from over-strengthening the frame to the motherboard, so consideration must be exercised in closing and using emotion and logic in tightening strength. If it is possible to define in words – simply close until there is sufficient resistance.
After the launch of the 13th generation Core and the new motherboards with the Z790 chipset like the ASUS ROG Maximus Z790 Hero you see here, we wondered how the move would look with the new processors, as they are especially known to be power hungry and very hot.
We checked whether a modern system would benefit from any improvement, since tests conducted in the 12th generation showed that by using these metal frames, the working temperatures can be improved dramatically. We have developed a test methodology for this purpose. Noctua’s NH-D15 Chromax.Black cooling was used and it was determined that the speed of the single fan installed in it would be 75 percent, which is about 1,200 rpm. This speed will not change during all the tests so that the cooling output remains the same.
The overclocking tool that Intel developed, the XTU, is a perfect tool for this kind of testing, as it allows segmentation of heat envelope types and speeds, real-time changes and now supports Intel’s 13th generation. We created five profiles for each test mode, where they include one mode without thermal envelope restrictions and four modes with thermal envelope restrictions to create a segmentation in the amount of heat produced by the processor in each mode.
In each test, we let the Classroom test in the Blender software renderer work and embrace all the cores until the processor reaches its electrical envelope, or in the unrestricted case, the heat envelope.
At Zon we believe that even boring results are important results, because they have consequences for end users who want to experiment with these things. Completely surprisingly, we saw that between the state of using the frame and the state of using the normal bracket mechanism we saw almost no differences at all.
The rendering times remained the same, the operating temperatures remained the same except for the 140W mode, where you can call it a standard deviation. Yes, there is data that is completely a bonus here, and that is the ratio between the power consumption and the rendering performance of the processor. It is interesting to see that compared to unlimited mode (usually 255-275W with such cooling), 190W mode loses less than 5 percent of performance. We note that the gaming performance remained the same, since the few cores that are normally active can move freely within the framework of this heat envelope.
We will add that after all this, we replicated the test system and methodology to MSI’s Z690 Carbon WiFi motherboard and there we observed almost identical results. We were a little surprised, as we assumed that the bracket mechanism would be slightly different.
We don’t have much left but to cautiously come to the conclusion that there is no need or use within the 12th generation repair framework, and that Intel made sure that in the new processors the contact between the heatsinks and the processor cover would be healthy. It seems that now we have to continue researching ways to tame the beast, and tame the Core i9 13900K and its challenging power consumption.