Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Discussions Rules and Guidelines
Report this post
CPU change should not change overall scores much. CPU test score is only about 10% of the overall score. Fire Strike going up more makes sense as it is already somewhat CPU limited on a 3080. 200 point difference in Time Spy tends to be within normal variance, would need multiple runs to see if this is just an outlier.
Since Time Spy is GPU limited on 3080, would not expect substantial changes in overall score from just a CPU swap.
Thanks for the insight. Sorry, I should have linked some results in my first post. I wasn't thinking. I didn't do several runs of each but here's Firestrike and Time Spy for both processors, unfortunately; just before and after results.
3900x
Firestrike - https://www.3dmark.com/fs/25284496
Time Spy - https://www.3dmark.com/spy/19456660
5900x
Firestrike - https://www.3dmark.com/fs/25387838
Time Spy - https://www.3dmark.com/spy/19775948
-------------------------------------
In both benchmarks, your GPU scored slightly lower in the second than the first.
In Fire Strike, the physics and combined scores were so much better, presumably from the new CPU, that they pushed the overall score up.
In Time Spy, though, the CPU score was only a couple of hundred higher than before and didn't counteract the downward pull from the lower GPU score.
Without more information about your system and settings, I can't say why your GPU is scoring slightly lower less than a month later in both benchmarks. Maybe its overclock needs reworked? Maybe you should reinstall the latest driver after uninstalling it with DDU? Maybe it was just temperamental during those runs and would do better if they were run again?
Your CPU performed remarkably poorly in that Time Spy run. But, it was pretty good in Fire Strike. An unusual result?
I used DDU to nuke my current video driver and reinstall it just in case. I did two new fresh runs back to back (hope two was enough). I don't run benchmarks that often or competitively so forgive my ignorance on some of this stuff. The links should show what my clock speeds are for CPU and GPU. Everything feels like it's running smooth but I was just curious and wanted to see other opinions/advice. Thanks for taking the time to have a look.
New scores
Firestrike - https://www.3dmark.com/fs/25405664 / https://www.3dmark.com/fs/25405682
Time Spy - https://www.3dmark.com/spy/19832403 / https://www.3dmark.com/spy/19832599
As for my system, everything is stock except I do have D.O.C.P enabled in BIOS for my RAM. I have EVGA Precision installed that has my power limit maxed but no overclocks on anything. I do have G-Sync monitors but I have it disabled during all the testing and I don't change anything in the benchmark settings. I don't choose "Custom" and make any changes; I just click "Run".
Detailed Hardware Specs
Imo, if that AIO is attached to your CPU and not your GPU, try increasing the motherboard's Precision Boost Overdrive (PBO) setting for your CPU to try to get some more power out of it.
As far as the GPU, have you tried flipping the vBIOS switch to the high-power version instead of the quiet version? If not, do that when your system is shut down, and then boot like normal. Test with that running, and see if you like it.
The switch on my GPU was actually already set to "OC" instead of "Normal" out of the box but I did hop into the BIOS and just found Precision Boost Overdrive and set it to "Enabled". I wasn't too sure on the other options it had under "Advanced" so I just left it at that. Also enabled another one that just said "ASUS Performance Enhancement" because...I guess why not? Here are some more results that saw slight improvements without any major hits to temperatures. Again, these runs are just back to back after tweaking those settings you mentioned
Fire Strike - https://www.3dmark.com/fs/25406124 / https://www.3dmark.com/fs/25406145
Time Spy - https://www.3dmark.com/spy/19833633 / https://www.3dmark.com/spy/19833682
As both of those products are largely stock, scoring near the median in Time Spy is fine. The scores don't tell us that anything is wrong. And, as long as you're happy with the in-game performance, I wouldn't worry about it. They'll likely last you for years. :-)
But, I feel certain that there's headroom there for both products though if you're willing to take the time to massage the settings into shape. It would probably result in only several percent of performance improvement (vs. 10 or 20 percent), so whether that's worth the time involved is up to you. For example, I use an ASUS 3080 Strix, and this is the Time Spy result I got about two weeks ago after reworking my GPU's undervolt and opening my 5900X's PBO range by a bit: http://www.3dmark.com/spy/19609712 . And, this was with a legitimate GPU overclock, but lower CPU settings, back in January: https://www.3dmark.com/spy/17198384 . I spent days-worth of time optimizing my settings after getting my hardware last year and have spent many more hours since trying to keep it all in really-good shape.
If that all sounds appealing to you, I suggest researching how to improve performance with those products, plus your system RAM. If not, I completely understand. :-)
Cheers, and good luck!