• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Cyberpunk 2077 Ultra performance

It is but some are happy with that level of detail. The conversation then reverts to people posting how great the game looks. I could care less about RT on/off comparisons. It's obvious the game has a gimped LOD.

Yup it definitely does and it can be rather distracting at times! That and the noise on reflections are my only graphical gripes.
 
All valid points but try to remember this is the performance thread, not the gameplay thread. So if we remain objective then there is less ammo for the CDPR defence league to fire.
I fully understand why some don't like the game and regardless of what hardware they are using. Bugs galore are a plenty in CP and it really does have a lot to answer for on console systems. Turning off RT, it runs great on a 2080Ti at 1440P UW and bugs aside, I am quite surprised to see the negativity towards the visuals but again, each to their own. What I have seen here and the RDNA thread though is a dislike for it because NVidia and that is quite sad. I loved TressFX in TR when that came out and applauded AMD for such a good job and that really did add some decent realism to an already gorgeous game.
 
Its the new Crysis isn't it. What did you expect? Its a punishing game.

2080 is hardly going to cut it at 4K with RT lol :eek::eek::eek: could have told you that years ago?:D
You have a GPU that's almost 2.5 years old and you want to run 4k with RT with limited DLSS.

Had you been clued up you should have ditched the 2080 before Ampere came along, and got 3070 or 3080.

you cant even run cyberpunk at 4k with RT on a 30 series card and get 60fps best your getting is 55fps and thats with a 3090.
 
We wouldn't be having this conversation if you had 3080, you would be playing it in RT.

"Regressed regression (proof already given)" like what?

O
We wouldn't be having this conversation if you had 3080, you would be playing it in RT.

"Regressed regression (proof already given)" like what?

Sorry, check my post again as the images I used as proof did not link correctly. RTX 3080 or not the point is many feel the RT is not worth it and prefer the better FPS because the game gives over 50% performance impact when enabling RT (even with DLSS on balanced). From 83 FPS to 38 FPS is one hell of an impact for marginal improvements in most areas. That is my point (and that of many others), you get some definite eye candy improvements (sometimes subtle as shown above) for a significant performance impact.

https://www.techpowerup.com/review/cyberpunk-2077-benchmark-test-performance/4.html
 
Last edited:
It is but some are happy with that level of detail. The conversation then reverts to people posting how great the game looks. I could care less about RT on/off comparisons. It's obvious the game has a gimped LOD.
Well good thing the high def textures are still there it really looks nice that way and again, if some are missing let's hope we will see improvements from the devs or mods. There is some potential in this game and again, if it brings multiplayer then it will be a big success because you can have enough fun and sometimes even more fun if there are a lot of bugs and glitches in the game. :)

I'll repost your link because you made a mistake there and the address needs to be fixed.

https://imgsli.com/MzQ4MDA

It really looks nice with that cinematic mode.
 
O


Sorry, check my post again as the images I used as proof did not link correctly. RTX 3080 or not the point is many feel the RT is not worth it and prefer the better FPS because the game gives over 50% performance impact when enabling RT (even with DLSS on balanced). From 83 FPS to 38 FPS is one hell of an impact for marginal improvements in most areas. That is my point (and that of many others), you get some definite eye candy improvements (sometimes subtle as shown above) for a significant performance impact.

https://www.techpowerup.com/review/cyberpunk-2077-benchmark-test-performance/4.html

that makes a bit more sense:D:)
 
Have you got Cyberpunk2077?

And that is relevant how? You don't need to own the game to tell those kind of glitches are not acceptable. The AI driving is definately abysmal and there is barely a play sesison where the AI cars arent't crashing into barriers or each other.

The world looks flat and uninteresting to me, it has sometime great levels of detail but overall it does not reach the level of detail as that of London in WDL for example. Just looking at the intracacy and detail in the architecture in WDL puts CP to shame. The fact it is based on a real vs fictional location does not change the fact that CP looks very basic in comparison. Note, I am not referring to gameplay here.
 
Last edited:
Well unless something has changed you need NVidia RT series to run raytracing in this game as AMD doesn't support it :confused: Did you not know that?
Dodge the question then.
You can play the game without RT , did you not know that?

And I would assume most of the people posting screenshots of RT on Vs off have an nvidia GPU otherwise they would not be able to turn it on to take the screenshot of RT on, did you not know that?

Are our opinions only valid if we have an nvidia RT GPU then?
 
Last edited:
O


Sorry, check my post again as the images I used as proof did not link correctly. RTX 3080 or not the point is many feel the RT is not worth it and prefer the better FPS because the game gives over 50% performance impact when enabling RT (even with DLSS on balanced). From 83 FPS to 38 FPS is one hell of an impact for marginal improvements in most areas. That is my point (and that of many others), you get some definite eye candy improvements (sometimes subtle as shown above) for a significant performance impact.

https://www.techpowerup.com/review/cyberpunk-2077-benchmark-test-performance/4.html

That link shows :

Code:
RT Off                   83 FPS
RT On DLSS Performance   84 FPS
RT On DLSS Balanced      74 FPS
RT On DLSS Quality       65 FPS
RT On DLSS off           38 FPS


I've got my 3080 installed in an old Ivy Bridge system, 3770k, DDR3, SATA SSDs. I'm playing with RT Psycho, DLSS Quality and everything else maxed except film grain and motion blur at 1440p/60Hz. My FPS in such an old system is still 45 - capped 60FPS. Sometimes it can drop lower but as I tend to be moving forward I don't notice it.
 
And that is relevant how? You don't need to own the game to tell those kind of glitches are not acceptable. The AI driving is definately abysmal and there is barely a play sesison where the AI cars arent't crashing into barriers or each other.

The world looks flat and uninteresting to me, it has sometime great levels of detail but overall it does not reach the level of detail as that of London in WDL for example. Just looking at the intracacy and detail in the architecture in WDL puts CP to shame. The fact it is based on a real vs fictional location does not change the fact that CP looks very basic in comparison. Note, I am not referring to gameplay here.

Your post has no relevance to this thread
 
That link shows :

Code:
RT Off                   83 FPS
RT On DLSS Performance   84 FPS
RT On DLSS Balanced      74 FPS
RT On DLSS Quality       65 FPS
RT On DLSS off           38 FPS


I've got my 3080 installed in an old Ivy Bridge system, 3770k, DDR3, SATA SSDs. I'm playing with RT Psycho, DLSS Quality and everything else maxed except film grain and motion blur at 1440p/60Hz. My FPS in such an old system is still 45 - capped 60FPS. Sometimes it can drop lower but as I tend to be moving forward I don't notice it.

You are aware that DLSS can be used without RT to give a significant performance increase. So at 1600p (as per that link) the FPS with DLSS on performance could be well over 150 FPS? Unfortunately the test does not compeletly test apples to apples and if it did then the performance hit would still be ~50% like for like. Hence my reason to compare only RT on vs off without DLSS added to skew the results as DLSS was not tested without RT.

So the question still becomes do I take a 50% performance hit to get the additional RT eye candy. The problem is this is a suibjective issue and will very from person to person.
 
Back
Top Bottom