Support NeoGAF

[Radical_3d]

I just watched this. It's 2 months old but I haven't seen it posted anywhere. With love to my Graphical Fidelity I Expect This Gen thread peers.

 

[PaintTinJr]

I just watched this. It's 2 months old but I haven't seen it posted anywhere. With love to my Graphical Fidelity I Expect This Gen thread peers.

It was an interesting video, but the premise of his testing was completely flawed and the lack of comment regarding the wave like LoD transition noise from the LoD solution when moving indicated they were probably predisposed to prefer LoDs.

Had he arranged the one reused asset with random heights and orientations - kit bashed, even as nanite intends - rather than unrealistically as an orderly grid, used depth cuing fog and got in close - as a scene would be - the loss of detail from the LoDs, the performance impact, shadow quality and signal/noise would have favoured nanite IMO.

edit:
They also try to demonstrate with far out image comparison at the start that lowering the LoD on the asset makes no difference, but under advanced lighting or material shader use and closeup, nanite's polygon per pixel light interactivity would easily be seen as a huge difference, dispelling that argument too.