Submitted 1 month ago by recursive_recursion@lemmy.ca to technology@beehaw.org
https://inv.nadeko.net/watch?v=s-lFgbzU3LY
Apologies for the late vid, still emotionally recoving from the recent US news
it’s the new gaming king! Still riding high on my 5800X3D personally, that was a GOAT purchase. I know the AM5 counterparts are beating it out on all the charts, but videos like this highlight how it is still better than the majority of CPUs when it comes to gaming performance.
Is there already a suicide hotline for userbenchmark?
While I think that userbenchmark needs to find a better hobby, I’d like to ask everyone to refrain from making suicide jokes
Fair…
PS: Wait… that’s a hobby and they don’t get paid for lying? That’s even worse than I thought.
recursive_recursion@lemmy.ca 1 month ago
One ridiculous outlier is Baulder’s Gate 3 on the 9800X3D with a 26.9% performance gain over the 7800X3D. Truly ridiculous
Toes@ani.social 1 month ago
wow! did they figure out why?
CmdrShepard42@lemm.ee 1 month ago
Because 9800 is 25.6% larger of a number than 7800.
recursive_recursion@lemmy.ca 1 month ago
no idea
Steve from GN and their team re-ran their tests multiple times and cross-verified with 2 other reviewers as they were also in disbelief/sketched out(I’m inferring their reactions here) by the results.
So at this point it’s hard to say what’s causing the significant performance uplift.
It could be due to something in the cpu like the relocation of 3D v-cache, the larger cache buffer but unless we see other reviewers provide test details with some differences it’s difficult to say what exactly is causing this beneficial performance.
theangriestbird@beehaw.org 1 month ago
like what was the tech that caused such a leap? They talk about how they switched up the x3D cache configuration - it used to be stacked above the main part of the chip with some sort of insulation in between, but they found a way to remove the insulation and put the 3D cache on the underside of the chip. I’m no expert and i watched the video yesterday, so sorry if that explanation isn’t clear. Effectively, they found a way to bring the 3D cache “closer” to the cores, which shortens the time it takes for the cores to communicate with the cache.