Open AndreaMissaggia opened 2 months ago
Same here in God of War (2018). FSR 3.1 using optiscaler looks bad compared to both the original far 2.0 implementation and optiscaler's fsr 2.2.1, which, BTW, looks slightly better than the original fsr 2.0.
I am using it with SteamOS in the steam deck, btw.
Same here in God of War (2018). FSR 3.1 using optiscaler looks bad compared to both the original far 2.0 implementation and optiscaler's fsr 2.2.1, which, BTW, looks slightly better than the original fsr 2.0.
I am using it with SteamOS in the steam deck, btw.
That is strange, do you have any screenshots of it happening? Can you offer more information? I tested GoW 2018 on Windows when implementing FSR 3.1 for DX11, and other people tested that game too, with none experiencing those issues.
Which Proton version are you using, and what SteamOS branch are you in? Can you send an OptiScaler.log and a proton log of the game?
Edit: I have also personally tested OptiScaler on Linux with two different games (Final Fantasy XIV and Deep Rock Galactic) and FSR 3.1 performed correctly there, so while I have not tested GoW on Linux, at least with other games, FSR 3.1 was working correctly there, too.
Same here in God of War (2018). FSR 3.1 using optiscaler looks bad compared to both the original far 2.0 implementation and optiscaler's fsr 2.2.1, which, BTW, looks slightly better than the original fsr 2.0. I am using it with SteamOS in the steam deck, btw.
That is strange, do you have any screenshots of it happening? Can you offer more information? I tested GoW 2018 on Windows when implementing FSR 3.1 for DX11, and other people tested that game too, with none experiencing those issues.
Which Proton version are you using, and what SteamOS branch are you in? Can you send an OptiScaler.log and a proton log of the game?
Edit: I have also personally tested OptiScaler on Linux with two different games (Final Fantasy XIV and Deep Rock Galactic) and FSR 3.1 performed correctly there, so while I have not tested GoW on Linux, at least with other games, FSR 3.1 was working correctly there, too.
I'll do it as soon as possible and will open an Issue. Maybe on Monday. It may be related to GoW being DX11?
I am using Proton Experimental, deck OLED, Main branch (SteamOS 3.7 ATM).
I added this related issue: https://github.com/cdozdil/OptiScaler/issues/70
Can not reproduce this on SteamOS 3.6.17 with Proton 9, game looks identical to Windows when running FSR 3.1.1 with Direct3D11on12. Though note that FSR 3.1+ shimmers more on high frequency textures compared to FSR 3.0/FSR 2.2, this is normal and is an issue on FSR itself.
Can not reproduce this on SteamOS 3.6.17 with Proton 9, game looks identical to Windows when running FSR 3.1.1 with Direct3D11on12. Though note that FSR 3.1+ shimmers more on high frequency textures compared to FSR 3.0/FSR 2.2, this is normal and is an issue on FSR itself.
Then it's just that I am very sensible to shimmer. Thanks again! )
EDIT: After testing your fix in v0.6.8-pre1, FSR 3.1 feels way better in GoW. Maybe I'm just making things up but now it feels definitely better than FSR 2.2... Anyway, thanks for the great support! You rock guys!
When using FSR 3.1 with this mod in Cyberpunk 2077, there is significantly more shimmering during camera rotation compared to both FSR 2.1.2 (modded) and the native FSR 3.0 implementation in the game.
This issue is more noticeable when using a controller due to the smooth and linear camera movements it allows. Both the native FSR 3.0 and modded FSR 2.1.2 implementations do not exhibit the same level of shimmering.
I have tested this with both reactive masks enabled and disabled, as well as with sharpening on and off, and the excessive shimmering persists in all configurations.
Additionally, the same issue with excessive shimmering appears when using FSR 2.2.1 from the mod. Interestingly, Cyberpunk 2077's native FSR 3.0 implementation, which is based on the same upscaling technology as FSR 2.2.1, does not have the same shimmering problem. However, with the modded FSR 2.2.1, shimmering is very noticeable during camera movements, while the native FSR 3.0 implementation handles these transitions better.
That said, CD Projekt's native implementation of FSR 3.0 also has issues, as there is noticeable shimmering even when standing still in certain areas. This suggests that reactive masks may not be properly configured in some parts of the game.
I have tested FSR 3.1 also in Alan Wake 2, and the same issue with excessive shimmering during camera movements appears in that game as well, suggesting that the problem with FSR 3.1 may not be isolated to Cyberpunk 2077.
I used DLSS Enabler to test OptiScaler, and I experimented with OptiScaler version 6.6 and subsequent beta versions.