Unreal Engine 5, [UE5 Developer Availability 2022-04-05]

Not really. SF is reactionary, querying the GPU on what its sampling and you can respond to that. An algorithm is predictive, computing texture use before-hand. This is what virtual- texturing has relied upon to select which tiles to load; the better the algorithm, the fewer tiles you need pre-cache.
Not if it actually evaluating texture sampling for the current viewable frustrum.
 
EbzCW6rXkAIgrK3


FF7 remake part 2 on Unreal Engine 5?
 
Only if you want to wait until 2023 for FF7Rp2 to actually use the new tech.

Isn't that their likely timeline anyway? They weren't exactly speedy with part 1.

Also, Epic were again stressing that you can start with 4.25 and migrate to 5 fairly easily.
 
Also, Epic were again stressing that you can start with 4.25 and migrate to 5 fairly easily.

You can, but as we talked about before, you will likely need/want to have your entire content creation timeline done with UE5 to take full advantage of the two major new systems.

As to their normal speed, I assumed FF7Rp2 would be 2 years from their last release. Maybe I wrongly assumed they'd pick up speed and build momentum.
 
The funny thing geometry and texture quality is from a performance point of view a problem solve after the size of assets make it a problem but this is not a power problem anymore out of reaching a better resolution.

Realtime global illumination is now the performance limitation and asset quality is limited by size on the SSD.
 
Geometry is NOT a solved issue since Nanite is currently limited to rigid, opaque geometry. This is perfect for rocks or buildings but completely inadequate for foliage for example. Don't get me wrong : this is an exciting tech and a great first step in the good direction but still far from a perfectly general solution.
 
Geometry is NOT a solved issue since Nanite is currently limited to rigid, opaque geometry. This is perfect for rocks or buildings but completely inadequate for foliage for example. Don't get me wrong : this is an exciting tech and a great first step in the good direction but still far from a perfectly general solution.

They said this is not useful for foliage but if it is only staying foliage, hair and grass we begin to see the end of the problem, And the best it seems the system is not a performance hog. But we will wait for future iteration of Nanite technology.
 
You can, but as we talked about before, you will likely need/want to have your entire content creation timeline done with UE5 to take full advantage of the two major new systems.

As to their normal speed, I assumed FF7Rp2 would be 2 years from their last release. Maybe I wrongly assumed they'd pick up speed and build momentum.
Dev for FF7R took 3 years once they took it back from CC (and rumor has it they pretty much scrapped most of the work done thus far).

They have the main base of the next game with mechanics and gameplay set up. Definitely can see it taking two years, maybe three years if it's to be a bit longer and/more open world-ish.
 
They said this is not useful for foliage but if it is only staying foliage, hair and grass we begin to see the end of the problem, And the best it seems the system is not a performance hog. But we will wait for future iteration of Nanite technology.

Not foliage or hair and only rigid bodies so no characters. Basically it's only useful for a very particular set of environments : city buildings or rocks Most games don't take place only in the desert, so at the moment it's a promising tech but far from a general solution. John Carmack talked about infinite geometry more than ten years ago and John Olick built a demo based on his idea of raycasting a sparse voxel octree, it was already working on a G80 (albeit at a low resolution). Like this demo, Epic has solved the most simple use cases at the moment : rigid opaque geometry, this is always the same problem with raytracing/raycasting once you have deformable geometry and even after all this time we don't have a solution so I'll wait to see if Epic's implementation is applicable to more general geometry and the performance they'll achieve then before saying that the geometry issue is solved ;)
 
Not foliage or hair and only rigid bodies so no characters. Basically it's only useful for a very particular set of environments : city buildings or rocks Most games don't take place only in the desert, so at the moment it's a promising tech but far from a general solution

Last time I checked, most of a city will be rigid objects, including cars. Seems more of a question of how well the non Nanite foliage and characters will blend in with the higher polygon Nanite everything else.
 
What if you live under the sea? Surely less rigid bodies there.
 
Not foliage or hair and only rigid bodies so no characters. Basically it's only useful for a very particular set of environments : city buildings or rocks Most games don't take place only in the desert, so at the moment it's a promising tech but far from a general solution. John Carmack talked about infinite geometry more than ten years ago and John Olick built a demo based on his idea of raycasting a sparse voxel octree, it was already working on a G80 (albeit at a low resolution). Like this demo, Epic has solved the most simple use cases at the moment : rigid opaque geometry, this is always the same problem with raytracing/raycasting once you have deformable geometry and even after all this time we don't have a solution so I'll wait to see if Epic's implementation is applicable to more general geometry and the performance they'll achieve then before saying that the geometry issue is solved ;)

We need to wait like you said if the future release solves the non-supported features and if it is the case the only exception will be foliages, hair, and grass to use traditional methods this will be great. It seems at least they have some ideas for the transparent objects, non-rigid bodies and another important feature support tesselation and displacement.

But they are sure Nanite can't support grass, hair, and foliages but if all other features are working this is a great idea. Brian Karis told this has nothing to do with Geometry image and Jon Olick sparse voxel octree idea.

He said to Alex Evans of Media Molecule than there is some common idea with dreams engine.

Like you said wait and see and maybe an all in one solution will never work for geometry. ;)

:D There's no reason that Nanite can't be used for a destructible city. Might require some thought around how you prefab buildings, swap in damaged sections and debris pieces.

This is the case with building destruction during the end of the demo.
 
Last edited:
It always felt weird to me watching those demo's. Like the scene was beautiful but felt dead. And the I found out why, because nanite rendering billions of triangles could really only be applied to rigid structures.

Solve rendering richer billion geometry scenes with grass, trees, sand, water etc and ill be impressed..
 
Back
Top