Error, limitation, paradigm shift focused on chiplets … We do not know for the moment what happens with the modes and the BMI of the new generation Intel processors, but what is certain is that they step back in speed from Comet Lake-S, maximum exponent so far.
Intel Rocket Lake-S and its 11th Gen processors reportedly limited above 3733 MHz in Gear 1
If you follow us regularly this week you will have followed the evolution of this whole question of the BMI of the new Intel processors. Without wanting to be repetitive, the Gear 1 and Gear 2 modes will bring a queue, because as standard only the i9-11900K and KF
Although, as we said, we were in the middle of speculation, the performance appears to be different in the two modes, more reduced in Gear 2, but that does not invalidate the fact that Gear 1 can be set manually in BIOS like any other setting.
So far, everything is fine and it shouldn’t matter anymore, except that by sett ing Gear 1 with a 1: 1 divider, processors significantly reduce their maximum frequency even with overclocking. In other words, with a complete overclocking without triggering, the temperatures are already reported values that until now could be achieved almost with a series voltage as soon as the BMI was really good: 3733 MHz maximum.
Gear 2 mode adds latency and frequency results aren’t much better
Apparently the switch from Gear 1 to Gear 2 unlocks the frequency a few MHz more, we can possibly get DDR4-4000 in BMI which goes really black, but in return we will have a consideration as in the case of AMD : latency.
Examples of non-technical cases are already reported where the latency skyrockets above 10ns when changing from speed 1 to speed 2, even with the shifting included. Not only is this worrying, is that if the values are correct, we are talking about a loss of high performance that places them at best at AMD level in this section, and at worst behind.
On the other hand, and it goes without saying, other cases of improvements with the new BIOS are reported, where the latency would have dropped between 10 and 15% and the write speed would have increased in significantly higher percentages, in some cases with obvious problems of + 50%. Therefore, it seems that the platform is not mature in general, since these reports are with several manufacturers, indicating that drawing conclusions is hasty and that it will be necessary to wait until the reviews have a more reliable scale.
This could explain the performance gap that has been seen in recent weeks between the leaks, which we did not publish because we already felt something was happening when prestigious sites released less encouraging data that hadn’t been published. not much sense. In any case, Intel would have already hinted that, indeed, its platform would be above that of AMD, it now remains to be seen if this is indeed correct or is a product of marketing.