Good luck, hope it Works - -
You might want to check this out: Digital Emily
Were you involved?
@rocko I’m hoping selling my work will facilitate that…
@James_Unknown.64 Not sure I follow, but I assume you’re referring to the screenshots I posted here. The second one shows the modified mesh after it’s been subdivided, which is perhaps a bit misleading. The mesh is just about as sparse as I can make it without losing necessary topology:
4K textures would be ample for most cases using the face-centric UV map.
@TheAwesomeGM Already on my radar. I wasn’t involved, no.
Because it seems like it! sciency glasses on Let’s have a look.
Yes, ok, now the next one,
Yes, they certainly have some similarity.
But thats just a theory.
A ̶G̶A̶M̶E̶ 3D THEORY!
P.S. that got longer and much more unexpected then i thought.
Hey Chris, did you sculpt first all the figure and then make a retopology inside blender with a shrinkwrap modifier to keep the sculpted shape on the lower res mesh?
He says he imported it in from LightWave. so probably, no.
If memory serves, I modelled the base mesh in Lightwave, then used that to sculpt refinements in Sculptris, then back into LW to manually adjust the base to fit. It’s been back and forth between modelling and sculpting ever since (except now I do it in Blender).
Ok, thanks for the info.
How are you going to make the wrinkles for the expressions?.., have you made any progress with the tension map?
Also, what are the advantages of making a rig with a low res mesh with a SubD Modifier instead of going directly with a hi res mesh with No SubD Modifier?
I haven’t had a chance to revisit wrinkles, but I’m still hoping tension maps will be the go.
Low res with SubD makes manipulating and animating it more manageable. If it’s a high res mesh to begin with, you don’t have the option to lower the res to speed up viewport performance while still rendering at high res. It also makes weight painting and UV editing easier, and other things besides.
As an animator I can tell : a good low res mesh with a well thought topology makes animating a heavenly work.
If you don’t have that, you know you will have to animate in slow motion because it will kill your performances, it will look bad while you animate and you’ll have to kind of ignore that when checking your work, and require you to fix the wrong deformations after you’re done animating.
I haven´t yet rig anything in blender, but i bring a medium res geo from maya some time ago and did a quick rig with rigify and i was surprised by the performance.
Of course it did not have any corrective shape or any other deformer, it was just the skin weighting, but nonetheless it was very smooth… i wonder how much of a performance slowdown it will suffer with all the deformers and such…
Yes, performance is essential!.
Speaking of performance, what’s the overall playback frame rate with all rig features enabled and which features of your rig are the heaviest speed killers?
edit: oops, this question is related to Chris and not you yolao. Sorry for the mistake.
@yolao For reference I just did a quick test, and subdividing the mesh a few times brings the performance from 35fps down to around 2. Also selecting anything takes a second or more, and it hangs every so often for about 15 seconds. And that’s only at 350K faces… I dare not attempt to subdivide it into the millions!
@SidV I keep adding things to the rig, and amazingly nothing seems to slow it down (except the eyelid bulge warp modifiers, but only when SubD is active). That’s with 270 bones, IK all over the place and 180+ shape keys (most of which are not being animated though).
As for frame rate:
Interesting, thanks for the info Chris.
Since the eye bulge is perfectly spherical, why not use a cast modifier instead of warp. It’s faster and easier to setup. I picked this up from Helmut_S and i use it in many different places in my rigs ever since.
The Test below maybe a tad to overexaggerated (put this together in 5 min) but you can be quite suttle and precise with it, have a look…
EyeBulgeTest.blend (648.6 KB)
Not a bad idea, except that turning up the subdivisions reveals a hard edge, whereas the warp modifier has a user-definable falloff.
Thanks for your efforts though.
How did you start this project?
But wouldn’t you subdivide after the deform modifier?
I don’t recall exactly, probably by modelling the face.
Good point, that does improve it a lot even though it doesn’t have the falloff control that Warp does. However there’s another issue: the lip of the eyelid deforms in all directions, instead of just outwards.
As in what gave you the thought of making a realistic head?