Is SUB-D Pinching Okay For A Product?

PolyCore33

I'm trying to make a body topology that is anatomically accurate, and good for muscle deformation with no self-intersecting loops. So far most of it is done (very happy with it btw) and just need to lower the polycount on my previous head model.


Currently, there is only one, or should I say two areas with what seems to be subdivision pinching, and that is on the vastus medialis. I am not quite sure how to go about fixing it without creating intersecting loops.

sprunghunt

that's not really what I think of when I hear sub-d pinching. Those poles are probably fine. 

Alex_J

all you got to do is use transpose or make a zsphere rig to deform that area to check and see for yourself.

Bend the leg all the ways it could during animation. Does it look bad or not? If not, no problem. This is something worth practicing so that you can get quick at it if you are going to make a lot of models to be animated. 

Personally for human characters I prefer a quick rig in maya, but honestly just masking and using the transpose tools with a soft mask is probably all you need to see how some topology deforms.

I also think it's a good idea to create your topology in loops and an even grid as much as possible. Makes making edits later easier, makes unwrapping easier, minimizes texture distortion, and so on.

PolyCore33

Thanks for the responses! I am actually using Maya's Quad Draw. Although I have no rigging experience yet, I have had a lot of practice with topology, and try to keep it as clean as I can. I have no doubt the knee area will bend well, but after further thought, I guess the vastus medialis topology probably wouldn't be suitable for a smooth skin deformation (plus I don't like the way it smooths when subdividing), so I've been trying to rework it from the calves up.

carvuliero

I wouldn't worry about topology ,there are  bigger problems to worry about like hungry creature luring around you

PolyCore33

carvuliero said:

I wouldn't worry about topology ,there are  bigger problems to worry about like hungry creature luring around you

Lol, thanks. I guess the rectus femoris muscle does need to be deeper, but maybe not that deep (well, the bulk of it anyway). I think I should point out that this is based on female muscle anatomy lol.

carvuliero

There is no 'vs' here, its construction problem half of thigh muscles are doing absolutely nothing right now shes paralyzed it happens of something bite 4-5 tendons of your leg ->>>> check Pes Anserinus

PolyCore33

carvuliero said:

There is no 'vs' here, its construction problem half of thigh muscles are doing absolutely nothing right now shes paralyzed it happens of something bite 4-5 tendons of your leg ->>>> check Pes Anserinus

Sorry, but I'm not seeing it. Maybe it was just the angle in the image that confused you. The area you want me to fill in is the same area where you can see the curve with the vastus medialus blocking most of the skin in that area in the RL images. The straight leg stance is a good comparison to the 3D model images. If anything, it probably just needs slightly less curve depth towards the knees.

carvuliero

So nice of you let me be the confused one , to actually see "it" you have to know what to look for first

Using contracted muscle as reference to sculpt relax one is not a very good idea

Obviously you know what you doing so I will let you figure out if your thing and image below have any major differences

PolyCore33

carvuliero said:

Using contracted muscle as reference to sculpt relax one is not a very good idea

Obviously you know what you doing so I will let you figure out if your thing and image below have any major differences 

Thanks for your concern. 

If the area you are using as a reference is in similar position to that of a T-Pose, then it should be fine. getting muscle contraction references isn't my intention, but so long as there are no joint bending or lifting(etc) involved, there shouldn't be any issues. You would only end up sculpting finer muscle detail that can still be obtained while relaxed. If I can create the topology in a way that creates a smooth transition with a non-muscular morph, then it can be done. So far, the only issue I have is the front thighs, but am now showing progress.

I also want to point out that white muscle, especially tendons don't bulk up like red muscle does, which explains why that area has less girth, or looks like it has a bite mark as you say. As you can see in the image you posted, that area is all white. 

There is detail there, but minor and more visible when moving the model, but because they are tendons, they can't be bulky and are supposed to look like they aren't doing much. The topology in that area is also rather simple, so morphing in a layer of fat for a less slim muscular morph wouldn't be an issue. The only muscle I need to worry about in that area is the sartorius, which is already there. If you watch some videos, you can barely see any action in that area.

The major differences with the image you posted is that it is an ecorche, and the muscles are of average size. With a skin layer, everything would be smooth, and all you need is very basic topology. So getting an ideal transition would be to use a muscular model/reference to build topology on. I want to make this model as flexible as I can.