TechPowerUp Test: Large cores in Alder Lake are half as powerful as small ones

Such a test is directly offered, as Alder Lake-S has up to eight P and E cores, ie powerful Golden Cove and weaker Gracemont. So far, it has been tested that weaker kernels have been deactivated, which makes sense, as the scheduler is not infallible and sometimes assigns the E kernel to a demanding process, which of course has an impact on performance. In other words, deactivating the E cores can have a positive effect on the performance of the Alder Lake-S processors, but of course only if they have a high but incomplete load, ie games that can make the cores work hard but do not need much overall.

On TechPowerUp thus, we find an inverted test in which the P cores are deactivated, which of course affects performance, especially in synthetic or rendering CPU tests. Just look at Cinebench R23, where it can be seen that the E nuclei are surprisingly not as weak as might be expected, but it should be added that the P nuclei were depleted in this test of Hyperthreading (HT), which would their performance was significantly higher and above all their beats were reduced to the level of E cores. However, in other tests (video coding for example) the difference between E and P cores is much larger despite HT handicap, even with a multiple difference.

And what about games? In them, of course, the E nuclei are also weak, which is, of course, also due to their overall low beat. Interestingly, the P cores, whose clock speed was at the same level in the test, show only about 6 percentage points lower performance in Full HD compared to the default processor. When the resolution is set to 720p, it is already 11.5 percentage points and on the other hand in 4K the difference is negligible (0.2 pp).

This would mean that if someone plays exclusively in 4K, they can easily turn off the E cores on their Core i9-12900K and underclock the remaining P cores and they will not know the difference. However, of course, if someone plays in 4K, then they simply do not need a Core i9-12900K processor and will do well with much cheaper, which is contrary to what gaming companies usually provide us in the so-called recommended specifications for new games.

This is also evidenced by the fact that the difference between the Core i9-12900K by default and operation only with E cores is only 5 percentage points. We can easily buy a Core i5-11400F and the FPS will go down on average by only a small fraction, while we will save around 13 thousand on the processor, which we can invest in our local scalper with graphics cards.


As with any test, of course, it depends on the choice of benchmarks or games, so it cannot be said that the above should apply always and everywhere. And it also depends on whether we really use the set only for games that will demand their performance, because if someone also renders or edits video, a powerful and expensive processor will of course offer them much more.

Prices of related / similar products:



Source: Svět hardware by www.svethardware.cz.

*The article has been translated based on the content of Svět hardware by www.svethardware.cz. If there is any problem regarding the content, copyright, please leave a report below the article. We will try to process as quickly as possible to protect the rights of the author. Thank you very much!

*We just want readers to access information more quickly and easily with other multilingual content, instead of information only available in a certain language.

*We always respect the copyright of the content of the author and always include the original link of the source article.If the author disagrees, just leave the report below the article, the article will be edited or deleted at the request of the author. Thanks very much! Best regards!