You are reading a single comment by @JB and its replies. Click here to read the full conversation.
  • Personally I think it's a very bad thing. Huge investment required for all that additional modelling work means publishers are going to be wanting a safer ROI. Look forward to endless yearly rehashes of the best selling franchises. Of course people can still make great games with lo-fi art assets, but when your average consumer sees renders like that, their expectations are set a bit unrealistically. All IMO of course.

  • I guess the whole point of the stuff they're talking about is that it make it easier/faster(?) to import models and whatnot? I dunno.

    Impressive tech demo nonetheless - even if it is just Tomb Raider with an animated wingsuit lady.

  • There definitely seems to be some optimisations around the process of creating models at various levels of detail. I guess my point was more aimed at the impact of the general increase in graphical fidelity. While it may be easier to create various instances of models of varying levels of detail, the overall poly-count and effort on texturing is also ramping up significantly.

    I'm sure there will be some instances of amazing games that look equally amazing (CDPR, Rockstar etc.), I'm just concerned that gameplay and innovation will suffer in most cases.

    Maybe I'm just getting old!

  • It makes the process faster.

    Having done some 3D modelling (strictly hobby based though) the usual process is:

    • Model the thing you're doing up to the detail level you want it for your game
    • start baking normals/lighting/bumpmaps/textures and optimize it for lower polycount

    Actually being able to completely skip the second step is... huge.
    Now you can completely choose the detail level of a 3d model and don't need to worry how your level of detail or polycount negatively affects the rendertime.

About

Avatar for JB @JB started