My speculation is that it’s not necessarily any harder to support Nanite in the VR rendering pipeline. They just didn’t because VR is a low priority for them. They’ve arbitrarily developed it targeting flat gaming only first.
The big game changer for VR would be if they started treated VR as any sort of priority. That’s not any particular technology, but rather it would change the status of all sorts of things across the board, large and small.
Yeah I guess I just meant regardless of the pipeline support, since AER can be used (obviously not ideal) …at least there’s a chance we could see ot in VR…god bless Luke Ross hahah
Can anyone here actually start unreal with a pimax 8kX headset? I’ve tried a lot, and it never starts. Not 4.27 or 5. Was this demonstrated at all in the presentation?
Nanite is amazing technology. I think it is very heavy on data size; it certainly is for the game designers, who must load all the assets at full res into their graphics card’s memory. To build Nanite data, the original data is broken down into hierarchal triangle groups. I’m not sure how much of the original data is discarded (if any), but the geometry data doesn’t need to be loaded into video card memory. I don’t think the Nanite textures need to be loaded into GPU memory either.
The key to Nanite’s speed is that rendering is no longer driven by the number of triangles. It can draw 10s of millions of triangles just as fast as a few thousand. Instead, rendering speed is dependent on the number of pixels drawn, which is basically determined by screen size. That hurts VR rendering, since it needs a lot more pixels drawn (than a typical flat screen monitor).
If it really is software rasterised, then it sounds like a prime target for being offloaded to hardware, through whatever added support-functions/pipeline-variations/on-gfx-card-SSD-storage-with-DMA, which it could conceivably need, but is lacking today, in upcoming GPUs… :7
Still can’t help hearing an echo of Euclideon, in all Epic’s claims… 
The article shows that there’s more unsupported than we’ve discussed on this thread so far. Not only is Unreal Engine 5 not a game changer from the point of view of VR, it is functionally less capable in VR than the prior version 4.
Now that probably changes at some point. Further development and updates to Unreal Engine 5 will enable its features for VR and VR game developers will start adopting the new platform sometime in the future.
But the fact that lack of VR support was not considered a showstopper to release makes it very clear how seriously Unreal takes VR.
A lot of the industry still doesn’t recognize the storm of change that’s coming as VR takes hold. We’re going to see new names become dominant which embraced VR early (such as Meta), and we’re going to see old names fade out which were too late to hop on the train.
By dragging its heels on VR support, Unreal is opening itself up to potentially being overtaken by some other game engine coming onto the scene which fully embraces VR in revolutionary ways and becomes the defacto standard for VR game development. That would be the real game changer.
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.