Observations show unexpectedly minimal standards.
Observations show unexpectedly minimal standards.
Someone could assist you with your low CPU scores during benchmarking. Your Geekbench 5 test and 3Dmark time spy scores are available here: the links provided. Your system details are also listed, including the AMD Ryzen 9 5950X processor, NZXT Kraken X73 cooler, Gigabyte 6900xt GPU, MSI MAG X570 motherboard, Corsair memory, Sabrent Rocket Q SSD, and a CORSAIR HX1000 power supply. You've confirmed testing without the XMP profile and still achieved similar results.
In TimeSpy your GPU score is fine but your CPU score is very low. My stock i9 10900k does 14084 in the bench and your CPU does 11586. That is less than my stock i9 9900k does with half the cores. Use HWiNFO64 to see if you are throttling in the test. My i9 10900k at 5.3ghz all cores with a 360mm AIO throttles in that bench but does not in Firestrike or Cinebench. Even without HWiNFO64 you can tell since the bench will stutter.
Your single-core rating is only slightly below expectations, yet your multicore results are significantly lower. Verify the boost voltage and temperatures while under load; adjust accordingly. Cl18 3600 performance is average but shouldn’t drastically impact scores. If you left it stock, the default BIOS settings might be the issue. Likely scenario: PBO is applying excessive voltage and damaging the DOCP timings. Consider using a negative CPU voltage offset or a curve optimizer with Ryzen RAM calculator.
The bios show PBO is active, but peak voltages fluctuate to 1.49 without any clear cause.
You're about to purchase a dual-rank RAM module with 16GB each, priced at CL14. I'll check if there are any notable differences for you.