Zen 2 XT refresh initiated, awaiting further instructions from Linus.
Zen 2 XT refresh initiated, awaiting further instructions from Linus.
I understand this request may seem somewhat self-centered, but it’s important to get the information needed for informed choices ahead of time. I’m not sure if there’s a more effective method to reach out to the team, but I’m confident Linus and others will soon be getting the AMD XT chips for evaluation. I’m hoping the assessment covers FCLK enhancements, should there be any. Early whispers suggested a 2,000 FCLK target or possibly 1,900 now is possible for XT chips. As someone who uses RAM XMP, knowing if a 2,000 FCLK is feasible with 4,000 MHz RAM would be extremely helpful and could justify the additional expense compared to the 3700x model. Thanks.
I'm wondering if newer CPUs can achieve 2000 FCLK while keeping the same performance-to-RAM ratio as before. My test samples at 3600 and 3700X didn’t quite hit 1900, though 1800 is feasible but not much higher. I think checking overclocking communities could provide more insights on this. RAM is already reaching 4000, so that’s not a major issue, even though results can vary between tests.
In any case, values above 3600 don't significantly boost performance. Likely around a 5% improvement. Probably not full FCLK upgrades like out-of-the-box; it's more about improved interconnects that deliver results. I’ll need to learn more about these XT chips—I assumed they’re just a simplified version of the X chips.
Well, if you're considering a scenario where bandwidth is strictly limited by RAM speed and performance improves linearly, moving from 3600 to 4000 would only represent an 11% rise. For typical situations, I believe you're being overly optimistic with a 5% figure. From what I understand, these processors mainly offer a slight improvement in clock speed compared to earlier models. If you treat faster RAM as an overclock, some will still seek the absolute best results, even if the actual gain is minimal.