Question Using slower graphic card on lower resolutions can result higher fps than much faster card?

litwicki23

Distinguished
Sep 19, 2009
573
2
18,995
I have an question. Example Cyberpunk 2077. If i am testing 4090 with 12900K and using 1440P with DLSS QUALITY i have the same fps on DLSS PERFORMANCE. 80FPS DLSS QUALITY, 80 FPS DLSS PERFORMANCE. Because i am bottlenecked.
But when i use slower card ( like 2080 ti ) then i can have higher fps than 4090 on DLSS PERFORMANCE? <because i am not bottlenecked with 2080 ti ). Just using slower card on lower resolutions can result higher fps than much faster card?
 

Karadjgne

Titan
Ambassador
No. It's nothing to do with the card. It's the resolution.
1080p = 1920x1080 = 2,073,600 pixels.
1440p = 2560x1440 = 3,686,400 pixels.
4k = 3840x2160 = 8,294,400 pixels.

The gpu gets the same amount of frame packets from the cpu, regardless of resolution, but then has to render each frame it can, populating every single pixel with a color or none.

So if a cpu sent 500 packets per second, a 4090 might render all 500 at 1080p, 300 at 1440p and 100 at 4k.

Swap to the 2080ti, same 500 packets but 400 at 1080p, 250 at 1440p and 50 at 4k.

So comparing a 4090 at 4k to a 2080ti at 1440p, or a 4090 at 1440p to a 2080ti at 1080p, the weaker card gets more fps at the lower resolution.

You'd need to compare the different cards at the same resolution to get parity. Comparing to a lower resolution is always going to mess things up.

The DLSS is going to change things, Performance being higher fps at losses to graphical correctness, where Quality will have the opposite affect, sacrificing the amount of boost in favor of the picture.

You are not only comparing apples to oranges, but throwing potatoes and carrots into the mix. Doesn't make for a good salad no matter how much dressing you use.
 

litwicki23

Distinguished
Sep 19, 2009
573
2
18,995
Remember that posted benchmarks/videos of gaming FPS scores are usually the top .1%. They are not the norm - they are the outlier.
No one's going to post 'meh' benchmarks/gaming FPS videos. Benchers/reviewers tweak, bench, tweak, bench, tweak, bench, etc., until they get a score that they deem 'worthy' and only then does it gets posted. These are usually people who work with technology for a living and know the ins and outs of settings and components. Many of them also know how to 'cheat' the system by running the benchmarks with unrealistic settings. They change everything from CPU priority, to benchmark config changes, to doing borderless/alt tab crap. In short, don't worry about hitting published benchmark/FPS numbers.
 
  • Like
Reactions: Karadjgne

Karadjgne

Titan
Ambassador
Videos, benchmarks, basically any hard number that any person puts up for public viewing only does One thing. It shows what's Possible. It does not show what's Probable.

Usage is misleading because it's misunderstood from the start. Usage is Not how much of the cpu is Used, but how much it Uses. There's a difference.

You assume the 12700k/3080ti has the same settings, you'd be wrong. There's a lot more settings involved than just the game settings, there's global settings in nvidia control panel, GeForce Experience optimizations, bios, other hardware such as ram speeds, NVMe, Windows etc. All those can and do change what you see on the screen.
 

Karadjgne

Titan
Ambassador
You don't have a bottleneck. Forget that word entirely. It's being misinterpreted, misused and misunderstood. As is your usage. Lower usage does not imply lower performance, just as higher usage does not mean you'd get more performance.

If you ran down the street as fast as you possibly could, that does not mean you are using all the strength in your back, or all the strength in your arms, but you are still running as fast as possible. That's usage. It means there's still room for you to carry a backpack and some water and not slow you down a bit.

If you hit 100% usage, then you Are using all your strength in your entire body, and any additional weight or uphill slope Will slow you down.

Stop worrying about what others get, your performance is where it should be.
 
Last edited:
  • Like
Reactions: litwicki23

litwicki23

Distinguished
Sep 19, 2009
573
2
18,995
You don't have a bottleneck. Forget that word entirely. It's being misinterpreted, misused and misunderstood. As is your usage. Lower usage does not imply lower performance, just as higher usage does not mean you'd get more performance.

If you ran down the street as fast as you possibly could, that does not mean you are using all the strength in your back, or all the strength in your arms, but you are still running as fast as possible. That's usage. It means there's still room for you to carry a backpack and some water and not slow you down a bit.

If you hit 100% usage, then you Are using all your strength in your entire body, and any additional weight or uphill slope Will slow you down.

Stop worrying about what others get, your performance is where it should be.
Oki. Thank you Karadjgne for explanation. Really greatful.
 
Good comments above. ^^^
You should not want to see any resource like cpu or gpu running at 100%.
It would suggest that resource is a limiting factor.
CPU and GPU do not run fully overlapped.

What counts is how well YOUR setup performs.
If it is inadequate, do something; otherwise enjoy.