Proud to say this was also my first thought.
here's some older videos showing some of the improvements from 5.1
View: https://www.youtube.com/watch?v=yKg6NxrWOgI
View: https://www.youtube.com/watch?v=lorFbw_dRqY&t=248s
Yeah the Nanite trees are also moving and not static which is really cool. The games at the end of this gen are going to look bananasā¦this needs to be included in OP as Nanite in foliage has been like one of the TOP features people have asked for
on skinned Nanite meshes, I've seen a dev say that Nanite's software rasterization makes it more difficult, while hardware rasterization makes it easier. so I wonder if Epic is making nanite a hybrid to take that into accountthis needs to be included in OP as Nanite in foliage has been like one of the TOP features people have asked for
It will make it easier for developers to precompile PSOs which would solve compilation stutters.
This one is for the editor and it will likely introduce stutters there when being used. But it should allow for faster design iterations.
I thought posting this here would be the most appropriate. The below twitter thread demonstrates Lumen and Nanite running in VR on UE 5.1, which is extremely impressive.
View: https://twitter.com/iBrews/status/1562543743504244737?s=20&t=LZVpieZYTGVIv5lGGvKWKQ
Can't wait not to have stutters in 2026 then.At least we can hope that most if not all UE5 titles will be using v5.1+ from release.
View: https://www.youtube.com/watch?v=9paHxiFW5_0
Nanite foliage is looking just as good as I dreamed. We may finally see some proper forests in games.
I wanna see a field of Nanite grass too tbh.
What about shader compilation stutter? Is it gone? That's the most important part. 100 FPS doesn't mean anything if the game is constantly stoppingFinally got to messing around with this and the results are as good as I dreamed.
Performance is 10 fps higher with Nanite on as well. Give me a Crysis on UE5 now. It's time to kill LODs.
Edit: Whoa, I enabled Nanite for everything and FPS went from around 60 to more then 100fps.
If this is in editor, shaders are compiled upon loading the project. The project will have to be compiled into an executable and tested fresh to determine this. And it might still have it because since there's not immediate auto-precompilation step added to ue5 yetWhat about shader compilation stutter? Is it gone? That's the most important part. 100 FPS doesn't mean anything if the game is constantly stopping
For future games made in UE5 that use this version or newer..
I love this
Hyperbole if I ever saw it.People cheering about 60fps Lumen are in for a rude awakening.
My prediction is that 60fps Lumen will be a massive downgrade from 30fps. Like seriously noticable and those cheering for it will be switching back to 30fps lumen or atleast begging developers to go back to high res static offline lightmaps instead.
Refers to my Comment as Hyperbole and then instantly makes a blatant Hyberbolic statement...ok.Hyperbole if I ever saw it.
The differences will be so minute that outside of tech analysis channels, nobody will be able to tell the difference, nor care about it.
They are optimizing around the "High" setting for Lumen, rather than the "Epic" setting.
The differences between those 2 in practice are very hard to notice, and one of them gives a significant boost to performance.
The only ones in for a rude awakening are the people who think they know this engine better than the people who built/develop it.
UE Engineers: We will get this running at 60fps on Console, and it will look stunning.
Arm-chair Developers: No you won't.
š
curse you, I thought we had a more dedicated thread for UE5.1 so I was late!Unreal Engine 5.1 is now available!
Find out how this latest release makes the groundbreaking feature set released in Unreal Engine 5 more robust, efficient, and versatile.www.unrealengine.com
It's up !
you can control which light contributes to the Lumen RTDoes Lumen support different setting per light, or is it an engine-wide switch? You could use the faster version on user-controlled dynamic lights (like a flashlight), and the slower, better version for more static lights like the sun or other fixed artificial lights where the extra time to compute won't really be seen by the player.
if a game is early enough, they could probably move.I wonder if games that started development on UE 5.0 could easily make the transition to 5.1, or would this be a case where only games that start development in 5.1 will benefit?
Either way, I'm excited to see the massive improvements with Lumen and Nanite. I know the coalition is happy lol.
I wonder how close we are to having skinned meshes using Nanite?
Another random question, with an increasing number of developers moving over to UE, are mesh shaders ever going to be relevant?
I wonder if games that started development on UE 5.0 could easily make the transition to 5.1, or would this be a case where only games that start development in 5.1 will benefit?
Either way, I'm excited to see the massive improvements with Lumen and Nanite. I know the coalition is happy lol.
I wonder how close we are to having skinned meshes using Nanite?
Another random question, with an increasing number of developers moving over to UE, are mesh shaders ever going to be relevant?
In theory, not a huge deal. In practice, it's a huge "it depends". End of the day, it's still work - and hinges enormously on how far along something is.I wonder if games that started development on UE 5.0 could easily make the transition to 5.1, or would this be a case where only games that start development in 5.1 will benefit?
Either way, I'm excited to see the massive improvements with Lumen and Nanite. I know the coalition is happy lol.
I wonder how close we are to having skinned meshes using Nanite?
Another random question, with an increasing number of developers moving over to UE, are mesh shaders ever going to be relevant?
Fortnite is already UE5.Forgive my ignorance, but what is looking like the first UE5 game to ship, either on PC or console?
A new PSO precaching mechanism was introduced as Experimental in 5.1 to improve PSO hitching in DX12 titles. Improvements to this system in 5.2 include:
- We've improved the performance and stability of the system. There were various corner cases that needed to be addressed.
- We now skip drawing objects if their PSOs aren't ready yet. The system aims to have the PSO ready in time for drawing, but it will never be able to guarantee this. When it's late, it is now possible to skip drawing the object instead of waiting for compilation to finish (and hitching).
- The number of PSOs to precache has been reduced due to improved logic that omits ones that will never be used.
- We've improved the old (manual) PSO cache system so that it can be used alongside precaching.