openmw.org

Dont mind me, I am a bit confused these days and should stay away from the internet.
But I meant the four or three artists already working on the example suite - if they produce something to show it might get others interested. Or they just finish the whole thing.

One piece of feedback from the models I have seen (cant open the osg files), the human mesh could use some serious optimization. Especially the feet, the fingers, the ears and the nipples could be simplified a lot. Details like nipples or finger nails should be done via normal maps which works better and in some cases even looks better: normal map nipples > modelled nipples

Ah, understood!

I was wondering about the level-of-detail. Destined mention that the high-poly count wasn't a problem, and I'm not sure I'm 100% on-board with that.

I want ES to push the envelope to show what OpenMW can do. That includes have high-poly count models (if it adds any visual benefit).

The human model can be used by others as a basis for all other humanoid models which can be low-rezzed (is that a word?) as required.

A low quality version could be back-ported to the Template to replace Moon-boy for example? Is that possible?

You can make a low-quality version, attatch it to the same skeleton and have it work.

In an ideal world, we'd decimate the model and then bake the original onto it as a displacement map, then use hardware tessellation to recreate the highest detail version an individual machine can handle at runtime. However, we can't do this yet (and if Scrawl needs to take a break it might be a while before this is possible at all, if ever), so we can make do with normal maps and parallax maps.

psi29a wrote: The human model can be used by others as a basis for all other humanoid models which can be low-rezzed (is that a word?) as required.

This sounds like a good use case for multires.

With how things have progressed, we might have a release soon. Could be 1st August or 4th August.

very nice to hear that