The player lost 4k points following a 3090 upgrade.
The player lost 4k points following a 3090 upgrade.
I moved to the 3090 from the 2080 Ti. My CPU score on 3DMark went from 17525 to around 13700, and it's running at 5.7GHz all cores. Every test is consistent. Can someone clarify this? I'm curious about the full OC 3090 in water with a 5.7GHz i9 13900K, but I don’t get the same results. Also, no overheating issues.
Did you fully uninstall the previous drivers using DDU? If problems persist, a clean Windows installation might be necessary. Usually, CPU Stuff reflects sustained clock readings. Also, confirm it's Timespy Extreme, not regular Timespy. We have a comparable rating—my system has a 5800x3d + 3090ti.
yeah, it was ddu wiped actually . I had to send the card out to get repaired. just got a fresh driver installed. and no actually it was regular timespy. should i be running extreme for benchmark? is it cuz its a 3090 lol. but its weird cuz then why did my cpu score go down 4,000 points and on my 2080ti it was 17,000 cpu score. bottleneck or something?
Attached are the 2080ti and 3090 results. I just rerun the test since it wasn’t saved, and it finally hit the 14,000 threshold. Odd, right? With a score of 17,000 you’d see an average closer to 21,000 with a 2.3dmark-result.
They don’t function properly on the site. Do you have a regular webpage instead? Just share it so I can update it.
Open 3D Mark, navigate to Results, click Load, then use the files. This method functions correctly. Using it as a link will not work.
Still doesnt work lol. Just get me the URL like com/spy/33411434 number thing
From these outcomes it seems confusing at first. However, when comparing your data to many others, some patterns become clear. The 3D Mark results appear inconsistent, likely because hardware is currently very diverse. This makes sense given the wide range of systems available today. It probably stems more from adjustments, settings, and OS cleanliness rather than anything fundamental. Are you running 3D Mark in a clean environment with no background processes? Ideally, use tools like HWinfo64 and Rivatuner Statistics with an overlay to monitor usage and temperatures during testing. A clean score is best achieved without any background applications running.