It could just be noisy data, it’s comparing 365 days of 2024 with ~40 days of 2025
The first few days or weeks of a new year are less accurate compared to the end of a year.
PassMark sees the first yearly drop in average CPU performance in its 20 years of benchmark results
Submitted 2 months ago by misk@sopuli.xyz to technology@lemmy.world
Comments
gex@lemmy.world 2 months ago
realitista@lemm.ee 2 months ago
It may have to do with spending more of the die on NPU and GPU features? Some of these new integrated processors have massive GPU cores on them.
xia@lemmy.sdf.org 2 months ago
Peak CPU?!? Hoping for some kind of graphene terahertz breakthrough.
Eheran@lemmy.world 2 months ago
Why is the graph not logarithmic? Urgh
source_of_truth@lemmy.world 2 months ago
Because if it was logarithmic, it would look basically entirely horizontal.
Eheran@lemmy.world 2 months ago
What? No. Instead we would be able to see steady increases of say 10 % per year as a straight line instead of this, where it appears to be ever larger increases and the first ones essentially invisible.
Enkers@sh.itjust.works 2 months ago
I wonder if this has anything to do with Intel’s big snafu with gen 13/14 processors. If the solution was to push a microcode update cuts the voltage to the CPUs, it’s basically a “stealth” nerf. Their spin doctors have been working overtime to frame this as erroneously high voltages that were being “fixed”.