0
$\begingroup$

I'm a graphics programmer but I like to dabble in modelling just to better understand how artists use the tools. For the first time, I've modelled and rigged a human character. So far I've modelled them in a SFW way, lacking genitalia, but I'd also like to use the model to render realistic nudes. How can I get both uses from my character? Ideally I'd like to be able to use the same pose on both variants.

I can think of a few options, but they all have pros and cons.

  1. Model the genitals, then use a shape key to flatten them into the mesh. This seems most flexible, but my SFW version will have bad topology in that area because you can't change topology with shape keys.

  2. Have the genitals as a separate object. This keeps the SFW version clean (pun intentional), but the realistic version will have intersecting geometry. Maybe that's not hard to handle if it's parented to the rig well.

  3. Just save the two versions in different files. Then when I want to put them into a scene, I can link whichever version I want and switch between them. It seems like doing it this way I wouldn't be able to make any further changes (e.g. to fix up rigging problems that show up when I pose the character) because I couldn't make the exact same change in both files. This is a big issue for me because I'm not confident I've done a good job of the modelling and rigging.

Is there an option I've missed, or have I misunderstood one of the options? Maybe some pros or cons are much smaller than I've described and one of the options is clearly better to an experienced artist.


To answer comments: it's not an incredibly realistic model because I'm just not that good, but I'm modelling for offline rendering with a particle system for hair so I'm not worrying about polygon count. Baking the shape into a normal or displacement map might be fine for a female character but male genitals have more effect on the silhouette. I'm not really thinking of animations - more stills of static poses - but I'd like to keep open what I can try out as I go.

I know that workflows aren't really designed for going backwards and I have to commit to the polygons at some point, but I do keep having to go back when I find errors so I'd like to minimise that pain even if it limits me in other ways.

$\endgroup$
  • 1
    $\begingroup$ Intersecting geometry isn't a problem if you tell Blender not to render it. Create a driver controlling camera (+ shadow, etc) visibility and you can easily swap between the two versions. But the reality is, at a certain point you have to feature freeze things like rigging; IMO, it becomes a lost cause to perfectly preserve non-destructive editing, you end up wasting more time than you'd save. $\endgroup$ – Nathan Jul 6 '18 at 16:48
  • 1
    $\begingroup$ You may want to state the context of your model. High realism or not? High polygon count acceptable or not? Fluid and realistic animation or not. How much scrutiny will the unspeakable area receive? $\endgroup$ – atomicbezierslinger Jul 6 '18 at 17:00
  • $\begingroup$ Polygon count, in certain circumstances, has no direct connection to hair. Now that you have brought up an unspeakable topic, hair may actually work to minimize discontinuity. Hair placement will require UV Map. Hair can be painted on anywhere on the 3D mesh or texture and shaped. Hair placement is finer than polygon placement. Eyelashes for example can be placed hair by hair if necessary. $\endgroup$ – atomicbezierslinger Jul 6 '18 at 17:25
  • $\begingroup$ You have expressed somewhat the desire for the perfect model for all occasions. There is no need to worry about that. I can give you a money back guarantee for the purchase price of Blender, that your human mesh model and all other human models on BSE are not perfect with perfect unspeakable areas. $\endgroup$ – atomicbezierslinger Jul 6 '18 at 17:31
  • $\begingroup$ What you need is remove doubles modifier, that would enable you to switch parts of mesh and "stitch" everything together, keeping only 1 master mesh with 1 set of blend shapes etc.. This is very likely coming in 2.8 in model nodes, but for now this is something other 3D apps offer atm like Houdini. $\endgroup$ – Jaroslav Jerryno Novotny Jul 6 '18 at 17:34
1
$\begingroup$

Regarding point 2.

  • It helps to know what is the center of focus and pick your battles. More close view inspection may require more finesse.

  • Intersecting geometry is not necessarily an issue. If you do not create excessive geometry.

  • If the area is not subject to intense scrutiny then no one will observe the minimized discontinuity of the clip-on curls .. I mean ummm ...embellishment.

  • Clearly game development would use textures and normal maps to simulate geometry and the audience does not protest very much

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.