Question Question and maybe an issue with my 3dmark Speedway score with a 4090 GPU ?

Feb 3, 2023
3
0
10
So how does the test work exactly? what parts of my entire computer are actually being tested in the speed way score. it says its a GPU test, but from reading other peoples messages, it seems bios options, geforce options, and someone said things like motherboard and other things can SLIGHTLY effect the end score.
My issue is i have a MSI 4090 Gaming X Trio, which from reading others with this card i should be getting roughly 9950-10050 out of the box. But does the system matter?
Im getting 9650 which i find is on the lower end.

Does anyone know what i can do to increase my number without overclocking?
How big of a differnce is 9650 to 10,000? is it concerning?
Does anyone know what things i could try to increase my score?
Do certain parts of my computer help with this score?

Thanks for any response.
 
Feb 3, 2023
3
0
10
So it does matter in the speed way test? my CPU shouldnt cause bottlenecks, its a 5800X3D
which in a 53 game comparison on 4k is only 1.3% slower than a 13900K and on 1440p 4.7% slower i believe was the difference
let me find the link
https://www.techpowerup.com/review/rtx-4090-53-games-core-i9-13900k-vs-ryzen-7-5800x3d/2.html

this is why i think my number should be at least 9900.
the average is 10,050, but i know that alot of people with the card have the 13900K/13700K etc, plus DDR5 ram (makes a big difference) - also i assume alot of people overclock.
But to be 350 points behind the average seems alot.
 
all of your components figure in these benchmark scores.

of course the majority of those mid to higher scores have mostly or all higher end components.
while the lower scores probably are using slower disk drives to run the software, lower speed memory and less available, power supplies that cannot provide a proper steady load of power, lower end motherboards that don't take full advantage of the available technologies, etc...

without knowing your exact system specs, no one can offer you any real suggestions.
 
Feb 3, 2023
3
0
10
Whats the best way to show you what im running? a dxdiag?
also, when i put it on my 4k monitor it went to 9k. i understand 4k versus 2k is a big difference, but when it says average for my video card is 10,074, is that in 2k? 1080p? 4k?
like should i be doing the test in 1080?
ive been doing them normally in 1440p (2k)
 
The people who run these benchmarks over and over and over are usually those overclocking and trying to squeeze out extra performance. This skews these comparisons as the average is artificially inflated. Unless you have a heavily factory overclocked model or are overclocking yourself it is normal to get results below average. Ultimately synthetic benchmarks have a very limited usefulness. Try your games and see how they compare to game reviews/benchmarks, just ensure you can use the same settings and you know if the reviewer states they are not overclocking. If your results are within 10% then there is nothing wrong.
 
  • Like
Reactions: GarrettL
The people who run these benchmarks over and over and over are usually those overclocking and trying to squeeze out extra performance. This skews these comparisons as the average is artificially inflated. Unless you have a heavily factory overclocked model or are overclocking yourself it is normal to get results below average. Ultimately synthetic benchmarks have a very limited usefulness. Try your games and see how they compare to game reviews/benchmarks, just ensure you can use the same settings and you know if the reviewer states they are not overclocking. If your results are within 10% then there is nothing wrong.

When you run a test in 3Dmark go online to compare the results. As sizzling stated, look at the high scores and select one to view and they have overclocked their gpu's and likely cpu when possible. You can see the frequencies listed.

Same thing with my 4070 TI and the 3800x3d, scores extremely well in 3DMark. Those with slightly higher score have overclocked.