Great article ! Interesting and fun
Quote from the article:
" Throwing more cores at the problem isn't the solution, which is a bit surprising given the game will spread the load quite well across even 12 cores"
Maybe on a stretch here, but could we say that this game is badly optimized for multithreading ?
Being able to utilize all cores/threads for +90%, doesn't necessarily mean all threads are actually doing something useful
During performance testing for CAE/CAD software I've seen cases where the CPU usage was higher, but it took longer to complete the test.
In one example, when a process was configured to use only 1 core/thread, it finished quicker than when using 8c/16t. Another example was quicker in 4c/8t than 12c/24t.
In all these tests the CPU usage was +90% (for the available threads). Sometimes the difference in time was not much, but when taking efficiency or power usage into account, it became a different story:
10sec of 90% CPU utilization for 4 cores, or 10 sec of 90% CPU utilization for 12 cores is a big difference.
(For a game on it's own, this probably doesn't matter to much, but other applications running in the background could be affected ...)
I'm wondering if there is something similar going on in this game ?
I would guess, if the 12900K or 5950X are set to a lower (but reasonable) core configuration, that the performance would still be the same, but the power usage will be lower ?
(Freeing up resources for different tasks at the same time ...)