I don't see how it's a disaster. All I see is lots of polygons. Maybe some are oddly organized, for example, on the body.
Hello. I also like looking at illuminated screens as well as pressing buttons. We have a lot in common!
Ryan Roye
@Tanadrine-Studios
Age 41
Animator
Tanadrine Studios
Joined on 12/31/06
- Rank:
- Scout
- Global Rank:
- 39,525
- Blams:
- 33
- Saves:
- 200
- B/P Bonus:
- 4%
- Whistle:
- Normal
- Trophies:
- 10
- Medals:
- 280
- Supporter:
- 8y 1m
Comments
In modeling for 3d packages in general, more polygons means the mesh becomes harder to edit. Poor polygon flow increases the difficulty of getting the deformations (bending around) desired, and makes it far more difficult to do things like facial expressions without weird "kinks" surfacing and it added extra steps to the work in order to keep it from happening. The body's geometry was also so poor that it made bending the body at all difficult without ugly artifacts or noticeably-bad visual blemishes.
Then there's an issue that's more specific to Lightwave: Lightwave doesn't handle dense geometry well when modeling things, so even though the original mesh didn't look all that complex, the density of the polygons meant choppy and sluggish editing. There's also the fact that efficiently modeled objects will render a lot faster, which is a concern to everyone regardless of 3d software preference. We aren't yet up to the point to where we can get the real-time rendering performance that the average 3d game provides... I can't wait for things to get to that point though! :)
Hm. It seems the way the figure is modeled actually affects animation and how easy it is to work and to edit, not just skeleton rigging. Maybe this is why models tend to look weird when their body parts are turned to large angles offset from their default position.
But is there any common principle to properly model characters, to avoid oddness during animation?
Yup. And mostly... and these principles/challenges extend across every 3d animation software package in existence. There are a few tough areas that have been notorious for their difficulty for all artists; namely, the shoulders, elbow joints, and wrists where geometry needs to converge, but not appear to intersect, pinch, etc. Fortunately, Delura's models are simpler in scope so it is a bit easier to get right, but the same is not true for projects that demand a much higher level of detail.
The main thing about modeling is ensuring good poly flow (organizing the geometry so that bending can happen predictably), and efficient usage of geometry. It is a huge topic; not exactly easy to explain in text.
Of course, I forgot to mention things like "Sculpt-Modeling", where brute-force GPU power is used to allow the user to shape up their model as if it were made of digital clay. Incredible levels of detail are possible going this route, but the resulting models are never animatable (cannot be bent in any way without ugly artifacts). The sculpted model can be used as a reference for the final product, and some programs have "retopology" tools which crawl over the model and make a workable version of it in a lower polygon density.
VicariousE
It's not easy being a digital puppeteer, is it? Now that you've got a good machine to work with, I'm sure you'll be compelled to fill Delura with plenty of bodies.. hope it doesn't get too hairy!
Tanadrine-Studios (Updated )
Heh yeah... that's the big "gotcha" of 3d animation regardless of the software used. The process of animation itself is faster, but building things is often slower so it balances out especially for shorter projects.