Sign in to follow this  

What's a proper workflow/method modelling for animation?

Recommended Posts

I'm a beginner with 3D, so there's this tutorial series I'm following which totally throws me off. He models a head, but then leaves out the part where you create the mouth flaps and eyeballs for animation eventhough he kept talking about topology for animation throughout the video. He creates a full body model, combined all the body parts, but then on the next video for UV Mapping, he has the model uncombined again and only have half the body without explaining it because its easier to do UV mapping with a separated body parts I guess. I'm kinda lost here, so any help would be very much appreciated.

Share this post


Link to post
Share on other sites

I'm a beginner with 3D, so there's this tutorial series I'm following which totally throws me off. He models a head, but then leaves out the part where you create the mouth flaps and eyeballs for animation eventhough he kept talking about topology for animation throughout the video.

Yeah I'm not sure what he was doing there (maybe he had a separate video where he showed people how to model them or assumed his audience was knowledgeable enough to know by that point?).

As for the UV mapping part, the separation of the body parts is actually very important for texturing the model. Long story short, if you use an imported material (like the paint on a car) and you haven't adjusted your UV Map properly, it'll either look weird or completely wrong. 

Share this post


Link to post
Share on other sites

 

I'm a beginner with 3D, so there's this tutorial series I'm following which totally throws me off. He models a head, but then leaves out the part where you create the mouth flaps and eyeballs for animation eventhough he kept talking about topology for animation throughout the video.

Yeah I'm not sure what he was doing there (maybe he had a separate video where he showed people how to model them or assumed his audience was knowledgeable enough to know by that point?).

As for the UV mapping part, the separation of the body parts is actually very important for texturing the model. Long story short, if you use an imported material (like the paint on a car) and you haven't adjusted your UV Map properly, it'll either look weird or completely wrong. 

 

So what do you suggest about UV Mapping? I combined the body parts so that I know how they'll look like while still modeling. Because from his video the arm totally looks like its attached to the body but he could still highlight the arm like it's its own object.

Share this post


Link to post
Share on other sites
Posted (edited)
He models a head, but then leaves out the part where you create the mouth flaps and eyeballs for animation eventhough he kept talking about topology for animation throughout the video.

The human eye topology is complex as modeling a full human face, you would need a full tutorial to explain so it's no wonder that part was skipped, just look for tutorials on eye topology; same for nose and mouth.

A even better way of learning it, is studying anatomy. Topology is a anatomy term for relations between various structures, learning anatomy will improve your 3D topology.

 

but then on the next video for UV Mapping, he has the model uncombined again and only have half the body without explaining it because its easier to do UV mapping with a separated body parts I guess.
As for the UV mapping part, the separation of the body parts is actually very important for texturing the model. Long story short, if you use an imported material (like the paint on a car) and you haven't adjusted your UV Map properly, it'll either look weird or completely wrong. 

Don't cut this short, it would only cause confusion and disaster!  :blink:

During texturing and UV mapping the mesh, of a organic model ,should not be split this will cause baking problems and seams.

While UV mapping you won't physically split a model, instead you will make seams. This will keep normal intact while allowing you to work with the model as if it's split.

Your 3D software will treat a seam as a split in all ways except in regards to normals, allowing you to select a piece and hide other pieces; these are known as UV islands.

In other words, you should not split the mesh during UV mapping or texture baking.

Because your 3D software will treat a island as if it is a mesh on it's own, there is no reason to split the mesh.

Without seeing the tutorial, I am going to say it only looked like the meshes where split. Chances are the rest of the body was only hidden. If it was really split, it could be the artist is inexperienced with the software or using a low level 3D modeling software.

 

Now the reason for dividing the model into more than one UV map:

It would be ideal if a game charter model only used one texture, this rarely possible because of texel density.

Texel density is the amount of pixels a texture has per polygon. If you used only one texture you, wouldn't have enough pixels for the face to include small details; unless a texture of 8K or more is used.

The common textures per character is: Head_texture, Body_texture, Arms_Texture, Legs_texture. Now because these are four textures they can be made as one large texture, some 3rd person games do make them as one.

For most games the body that has twice the volume of the other parts, often more if there is a lot of clothing, the head in most games will also have a larger texel density because it's the part that is most focused on so your textures end up looking like this: 4096 * 2048 Extremities_texture and a 4096 * 4096 BodyAndHead_Texture.

This will mean you need 2UV maps for baking, however you still shouldn't split the mesh.

All you need to do is unwrap extremities to one texture and the body and head to a other texture. Any 3d modeling software, that can bake textures for games will allow you to use more than one UV map.

Using multiple UV maps will allow for smooth textures with no visible seams.

Edited by Scouting Ninja

Share this post


Link to post
Share on other sites
Posted (edited)

Don't cut this short, it would only cause confusion and disaster!  :blink:

During texturing and UV mapping the mesh, of a organic model ,should not be split this will cause baking problems and seams.

While UV mapping you won't physically split a model, instead you will make seams. This will keep normal intact while allowing you to work with the model as if it's split.

Your 3D software will treat a seam as a split in all ways except in regards to normals, allowing you to select a piece and hide other pieces; these are known as UV islands.

In other words, you should not split the mesh during UV mapping or texture baking.

Because your 3D software will treat a island as if it is a mesh on it's own, there is no reason to split the mesh.

Without seeing the tutorial, I am going to say it only looked like the meshes where split. Chances are the rest of the body was only hidden. If it was really split, it could be the artist is inexperienced with the software or using a low level 3D modeling software.

 

Now the reason for dividing the model into more than one UV map:

It would be ideal if a game charter model only used one texture, this rarely possible because of texel density.

Texel density is the amount of pixels a texture has per polygon. If you used only one texture you, wouldn't have enough pixels for the face to include small details; unless a texture of 8K or more is used.

The common textures per character is: Head_texture, Body_texture, Arms_Texture, Legs_texture. Now because these are four textures they can be made as one large texture, some 3rd person games do make them as one.

For most games the body that has twice the volume of the other parts, often more if there is a lot of clothing, the head in most games will also have a larger texel density because it's the part that is most focused on so your textures end up looking like this: 4096 * 2048 Extremities_texture and a 4096 * 4096 BodyAndHead_Texture.

This will mean you need 2UV maps for baking, however you still shouldn't split the mesh.

All you need to do is unwrap extremities to one texture and the body and head to a other texture. Any 3d modeling software, that can bake textures for games will allow you to use more than one UV map.

Using multiple UV maps will allow for smooth textures with no visible seams.

 

Whoa there's a lot of terminology I don't understand, guess I have a lot of learning to do. 
Here's the video to the tutorial, https://youtu.be/HLhazEa8wmw?t=1m10s

is only the arm being highlighted because he was using seams? Upon watching all the way through he does seems to have a method for full body UV mapping, but is he using the multiple UV maps like you mentioned? I have no idea.

Edited by Zrifepsych

Share this post


Link to post
Share on other sites
Posted (edited)
Here's the video to the tutorial, https://youtu.be/HLhazEa8wmw?t=1m10s

I can see why you are confused, he did use separate meshes that he then joined into one after UV mapping. The reason he did was because he thought it was easier.

The way he maps and then joins islands is slow and rarely used, because it results in strange errors.

The way he shows in the end, with the full body, is how UV mapping is done. He only used one map and in the end.

Some of his unwraps was less than optimal, for example the toes overlapped because he was using a cylinder unwrap when he should have used a cube unwrap.

Where he adjusted the texture to get the chess tile pattern the same size, that's texel density.

 

I think you jumped into the deep end too fast, it's not a bad idea. The problem is that with his tutorial he is expecting that you know the basics, so there is a lot that is missing from the tutorial. In fact it's more of a "how to" than a tutorial.

Start with a beginners UV map tutorial, then move to that one after and it will make more sense.

Also all 3D shapes and forms derive from basic shapes like cubes, start by learning how to unwrap them first.

 

Using more than one UV map to bake seamless textures is a advanced technique, you wont see it in UV mapping tutorial. You will have to look into texture baking before you notice it.

Edited by Scouting Ninja

Share this post


Link to post
Share on other sites

 

Here's the video to the tutorial, https://youtu.be/HLhazEa8wmw?t=1m10s

I can see why you are confused, he did use separate meshes that he then joined into one after UV mapping. The reason he did was because he thought it was easier.

The way he maps and then joins islands is slow and rarely used, because it results in strange errors.

The way he shows in the end, with the full body, is how UV mapping is done. He only used one map and in the end.

Some of his unwraps was less than optimal, for example the toes overlapped because he was using a cylinder unwrap when he should have used a cube unwrap.

Where he adjusted the texture to get the chess tile pattern the same size, that's texel density.

 

I think you jumped into the deep end too fast, it's not a bad idea. The problem is that with his tutorial he is expecting that you know the basics, so there is a lot that is missing from the tutorial. In fact it's more of a "how to" than a tutorial.

Start with a beginners UV map tutorial, then move to that one after and it will make more sense.

Also all 3D shapes and forms derive from basic shapes like cubes, start by learning how to unwrap them first.

 

Using more than one UV map to bake seamless textures is a advanced technique, you wont see it in UV mapping tutorial. You will have to look into texture baking before you notice it.

 

I see, I do need to understand more about texturing in general. Thank you for the explanation, it helps a lot.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this