Cyberpunk 2077 Hotfix 1.05 would improve performance of certain Ryzen CPUs

Published by

Click here to post a comment for Cyberpunk 2077 Hotfix 1.05 would improve performance of certain Ryzen CPUs on our message forum
https://forums.guru3d.com/data/avatars/m/94/94596.jpg
Moderator
kapu:

I wonder if anyone can explain poor 9900k performance .... It completely sux vs almost any AMD offering.
upload_2020-12-20_19-22-32.png
I agree. And it's not that it sucks against AMDs offerings, its that it should be right up there with them, as should the 8700K, not 20% behind. With this HEX edit fix and the AVX fix it makes me think it really is the compiler they are using. This next part is a rant, it's like they used the latest Intel supplied x64 compiler and it intentionally nerfs all CPUs before the 10 series. And another rant, you can't even disable AA (TAA) in this game because its tied to Nvidia DLSS. Now that I'm done talking shit I have to say I am have a great time playing this. 😛
data/avatar/default/avatar39.webp
rflair:

I agree. And it's not that it sucks against AMDs offerings, its that it should be right up there with them, as should the 8700K, not 20% behind. With this HEX edit fix and the AVX fix it makes me think it really is the compiler they are using. This next part is a rant, it's like they used the latest Intel supplied x64 compiler and it intentionally nerfs all CPUs before the 10 series. And another rant, you can't even disable AA (TAA) in this game because its tied to Nvidia DLSS. Now that I'm done talking crap I have to say I am have a great time playing this. 😛
Im on radeon and taa is decend in this game. But i would like some msaa additionally as i have gpu power to spare
https://forums.guru3d.com/data/avatars/m/220/220626.jpg
rflair:

I agree. And it's not that it sucks against AMDs offerings, its that it should be right up there with them, as should the 8700K, not 20% behind. With this HEX edit fix and the AVX fix it makes me think it really is the compiler they are using. This next part is a rant, it's like they used the latest Intel supplied x64 compiler and it intentionally nerfs all CPUs before the 10 series. And another rant, you can't even disable AA (TAA) in this game because its tied to Nvidia DLSS. Now that I'm done talking crap I have to say I am have a great time playing this. 😛
I'm not too sure. Looking at that pic, the 8700K stock is on par with the 3600, which is typical. The 8700K (or 8086K) at 5.0/5.1 GHz should perform similar to the 10600K. Same logic applies to the 9900K, an OC should move it up the stack significantly. That being said, when you use higher settings in Cyberpunk it very quickly becomes GPU bound. Even the high preset at 1440p should pretty much flatten that graph.
https://forums.guru3d.com/data/avatars/m/88/88871.jpg
Enjoying the game and working OK on my old Intel xeon 1680v2 8 core @ 4.8gz and Palit 2060 super 8gb on 1440p with gysnc module
https://forums.guru3d.com/data/avatars/m/273/273678.jpg
JAMVA:

My sandy Bridge from 2011 supported AVX , Instead of actually fixing it , they just ripped the code out for everybody 🙄
There was no code to begin with, all users were running SSE4.1 paths even with 1.01.
data/avatar/default/avatar32.webp
Game's performance is still trash... 5950x + 3080oc here and the fps dont fit the settings... Max settings looking at the city 69fps -> disable RT I get 64fps (wtf??) -> disable DLSS I get 70fps (wtf????) none of this makes sense...
https://forums.guru3d.com/data/avatars/m/94/94596.jpg
Moderator
Cyberdyne:

I'm not too sure. Looking at that pic, the 8700K stock is on par with the 3600, which is typical. The 8700K (or 8086K) at 5.0/5.1 GHz should perform similar to the 10600K. Same logic applies to the 9900K, an OC should move it up the stack significantly. That being said, when you use higher settings in Cyberpunk it very quickly becomes GPU bound. Even the high preset at 1440p should pretty much flatten that graph.
Yes but the difference from the 9900K to the 10600K makes no sense. Edit. I'm just commenting on that one graph also (Gamer Nexus).
https://forums.guru3d.com/data/avatars/m/123/123760.jpg
rflair:

Yes but the difference from the 9900K to the 10600K makes no sense. Edit. I'm just commenting on that one graph also (Gamer Nexus).
Intel 8K, 9K & 10K series should have barely any difference when running the same amount of cores and same amount of clock speed.
https://forums.guru3d.com/data/avatars/m/72/72830.jpg
Do the optimal setting as Digital Foundry has recommended for some FPS boost Standard Optimised Settings Contact Shadows: On Improved Facial Lighting Geometry: On Local Shadow Mesh Quality: High Local Shadow Quality High Cascaded Shadows Range: High Cascaded Shadows Resolution: Medium Distant Shadows Resolution: High Volumetric Fog Resolution: 1080p Ultra, 1440p High, 4K Medium Volumetric Cloud Quality: Medium Max Dynamic Decals: Ultra Screen Space Reflections Quality: Low but High if you find the amount of grain distracting. Subsurface Scattering Quality: High Ambient Occlusion: Low (there is barely a difference) Colour Precision: Medium Mirror Quality: 1080p High. 1440p High, 4K Medium Level of Detail: High Recommended Ray Tracing Settings Ray Traced Reflections: On Ray Traced Shadows: Off Ray Traced Lighting: Medium If that is not enough for you, Turn off reflections - the RT lighting is more important, on balance. Image Quality Recommendations: 1080p DLSS Quality, 1440p, DLSS Balanced, 4K, DLSS Performance [youtube=pC25ambD8vs]
https://forums.guru3d.com/data/avatars/m/69/69564.jpg
Netherwind:

Slight difference with less lows perhaps but no major difference with a 5900X. Do you agree that the game is more demanding on your CPU than GPU? I'm running my 3080 undervolted but cannot really say that GPU is stressed to the absolute maximum, like in Control for example, rather sitting at some 96-97%. Feels like the game is crazy heavy on the CPU which is understandable with all the things going on at the same time.
Completely GPU unless using 1440p DLSS balanced without any RT is used (aka low resolution) where the CPU can occasionally be the bottleneck. No difference between 3900x and 5900x either. (while using my normal maxed all settings+dlss)
https://forums.guru3d.com/data/avatars/m/132/132389.jpg
Installed 1.05, which indeed removes the memory pool modification files. Loaded up where I'm at, and without SMT it's loading slowly but... it actually IS using some SMT threads, just not all? What? I'm not sure if this is a bug/unintended or the intention from the start, but either way it is using more threads than before. I'll report back once I have the patience to do some proper testing if there's anything noteworthy to report, but for now all I can say is, yes it definitely loads slower when all SMT threads are not used, fixing it again to enable all SMT makes it load much faster.
asturur:

I think they know theyr game more than you or some other website do.
Then go play it as they're giving it to you instead of coming at me, wiseass.
data/avatar/default/avatar16.webp
On my system performance is flawless, but patch changed noting as i was on hot fix. I never seen fps, drop below 70 fps. Max settings - chromatic aberration - motion blur - film grain. +30% radeon sharpening
https://forums.guru3d.com/data/avatars/m/132/132389.jpg
kapu:

On my system performance is flawless, but patch changed noting as i was on hot fix. I never seen fps, drop below 70 fps. Max settings - chromatic aberration - motion blur - film grain. +30% radeon sharpening
Gonna have to be more specific bud. Max settings? What? On a 1060? What res, and what res is "+30% Radeon sharpening" running it at? At 1440p this game is programmed so poorly it makes a 3080 weep even with DLSS which at "quality" runs at 2.25x lower res, 1.5x lower res on each axis. So 4K would be rendering at 1440p.
data/avatar/default/avatar03.webp
Neo Cyrus:

Gonna have to be more specific bud. Max settings? What? On a 1060? What res, and what res is "+30% Radeon sharpening" running it at? At 1440p this game is programmed so poorly it makes a 3080 weep even with DLSS which at "quality" runs at 2.25x lower res, 1.5x lower res on each axis. So 4K would be rendering at 1440p.
Sorry my profile was outdated , totaly forgot to update it . 5600X + R6800.
https://forums.guru3d.com/data/avatars/m/218/218363.jpg
ViperAnaf:

Game's performance is still trash... 5950x + 3080oc here and the fps dont fit the settings... Max settings looking at the city 69fps -> disable RT I get 64fps (wtf??) -> disable DLSS I get 70fps (wtf????) none of this makes sense...
Which resolution? I agree that performance isn't fenomenal. RT off is definitely an improvement for me and some amount of DLSS does improve shadows at a distance compared to DLSS off.
Kaleid:

Do the optimal setting as Digital Foundry has recommended for some FPS boost Standard Optimised Settings Contact Shadows: On Improved Facial Lighting Geometry: On Local Shadow Mesh Quality: High Local Shadow Quality High Cascaded Shadows Range: High Cascaded Shadows Resolution: Medium Distant Shadows Resolution: High Volumetric Fog Resolution: 1080p Ultra, 1440p High, 4K Medium Volumetric Cloud Quality: Medium Max Dynamic Decals: Ultra Screen Space Reflections Quality: Low but High if you find the amount of grain distracting. Subsurface Scattering Quality: High Ambient Occlusion: Low (there is barely a difference) Colour Precision: Medium Mirror Quality: 1080p High. 1440p High, 4K Medium Level of Detail: High Recommended Ray Tracing Settings Ray Traced Reflections: On Ray Traced Shadows: Off Ray Traced Lighting: Medium If that is not enough for you, Turn off reflections - the RT lighting is more important, on balance. Image Quality Recommendations: 1080p DLSS Quality, 1440p, DLSS Balanced, 4K, DLSS Performance [youtube=pC25ambD8vs]
I use these and they give me around 65 fps in the city at 3440x1440.
Alex13:

Completely GPU unless using 1440p DLSS balanced without any RT is used (aka low resolution) where the CPU can occasionally be the bottleneck. No difference between 3900x and 5900x either. (while using my normal maxed all settings+dlss)
That's interesting. I also tried running the GPU at 2GHz but didn't feel like it did any difference.
data/avatar/default/avatar19.webp
5950X + 3080 Performance trash ? What are you smoking . I'm on 5600X + R6800 and NEVER saw FPS below 75 or something close. True im on 1080p , but 1440p wouln't change that much , i would still be above 60FPS , it is smooth as it can be , perfect frametime , zero stutter.
data/avatar/default/avatar25.webp
angelgraves13:

Psycho reflections are stupid. They cost as much performance as RTX reflections. Try Ultra.
It doesnt matter. One scene max settings with rt gives me the same as lowest settings no rt =70.... Something is broken on this game period. Looking at the floor fps burly goes up...
data/avatar/default/avatar37.webp
kapu:

I think they lied a bit on that 8 core part. Tomshardware claims fix is working . But check that 3300X and 5600X performance 😀 AMD 6 core matching intel 8 core.
What are those MODIFIED Ryzen configurations? Just PBO adjustments with 3800 or faster RAM?
data/avatar/default/avatar15.webp
Valken:

What are those MODIFIED Ryzen configurations? Just PBO adjustments with 3800 or faster RAM?
Moded game exe . Not OC.