Up until now, I've mostly used other people's computer models when I've wanted to make something. Making a computer model from scratch just wasn't in my skill set.
I'm trying to get better at it. So I started on one of the Kelvin. It's ""simple"" enough that I don't fear it will be beyond my ability, while also challenging enough to learn new techniques and practice. Here is the top superstructure completed so far.
3d modelers, I got a question if you can help me figure it out. I have an area that needs holding edges to keep a sharp panel line (marked in red). On the other side, I have a large cylinder that needs to smooth properly, so it can only have one edge running out (green). My solution for edge loop termination (blue) produces pinching and folding of the geometry after subdivision, as you can see. Any suggestions?
[COLOR=""#FF0000""]posting this here as f3d is giving me grief[/COLOR]
Placing any edge on a curve will distort no matter what you do. It sees that panel line as a fixed length and width so when you divide it it is a fixed detail so it will stretch around it. It acts like a bone in a tent. You can bodge it to appear like it isnt affecting the roundness but there will always be some level of distortion there.
here is what I think you have above. I am not sure why you have the 5 sided poly on the panel loop endpoint with the one loop radiating in the outer groove. (red in my image) As on it's own it would be a quad. Now my sample is a bit lazy as loops are not straightened out all pretty and are a bit exaggerated. Using 3 iterations on the base mesh. Removing the red loop for me resolves in a smooth result albeit there is a hitch due to the panel acting like a spar.
subdiv with that edge removed.
Both seem to smooth OK sans any folds. Might be some loose or doubled verts in there. It could be I only have one panel groove and adding in the others causes something so maybe increasing the geometery in the outer rim so you can distribute all those edges along the rim would avoid any pinching.
Like so.
Personally I would leave the panels and any crossing geo for after you collapse the subdiv and model them in minimizing poly count and any possible distortion the uvwing will gain in division. I am guessing your doing the unwrap on the base mesh then applying the subdiv ontop. Anyhow I could not reproduce the pinching and pulling in your example.
Lets see if I can attach the file. I saved it out to max 2010. I am sure your using something much newer than I am so it should open. Likely will vomit errors as I have preset backdrops etc.
also sample of the smoothing groups applied. this is a way to tighten curves without having to use multiple loops just make sure the smoothing group falls on a edge that is not changing the geometery or you'll get a hard edge in the smoothed surface.
Anyhow hope this sorta helps. What you have SHOULD be working unless there is some loose vert or edge in the mesh.
Lets see if I can attach the file. I saved it out to max 2010. I am sure your using something much newer than I am so it should open. Likely will vomit errors as I have preset backdrops etc.
also sample of the smoothing groups applied. this is a way to tighten curves without having to use multiple loops just make sure the smoothing group falls on a edge that is not changing the geometery or you'll get a hard edge in the smoothed surface.
Anyhow hope this sorta helps. What you have SHOULD be working unless there is some loose vert or edge in the mesh.
Thank you so much for that post, and the one before it. I will attempt those tonight. I am at work right now, so I can't really play with for about 10 hours. I really appreciate it.
Yes, I will be going back to this. The person I get reference from was busy for about 3 weeks, and couldn't get back to me. I needed reference of the shuttlebay. I have that reference now, but got sidelined on another project while I was waiting for it. I'll swing back around to finish up the Kelvin.
Posts
3ds max
Oh sorry, that's the bridge. Nice :thumb:
Placing any edge on a curve will distort no matter what you do. It sees that panel line as a fixed length and width so when you divide it it is a fixed detail so it will stretch around it. It acts like a bone in a tent. You can bodge it to appear like it isnt affecting the roundness but there will always be some level of distortion there.
here is what I think you have above. I am not sure why you have the 5 sided poly on the panel loop endpoint with the one loop radiating in the outer groove. (red in my image) As on it's own it would be a quad. Now my sample is a bit lazy as loops are not straightened out all pretty and are a bit exaggerated. Using 3 iterations on the base mesh. Removing the red loop for me resolves in a smooth result albeit there is a hitch due to the panel acting like a spar.
subdiv with that edge removed.
Both seem to smooth OK sans any folds. Might be some loose or doubled verts in there. It could be I only have one panel groove and adding in the others causes something so maybe increasing the geometery in the outer rim so you can distribute all those edges along the rim would avoid any pinching.
Like so.
Personally I would leave the panels and any crossing geo for after you collapse the subdiv and model them in minimizing poly count and any possible distortion the uvwing will gain in division. I am guessing your doing the unwrap on the base mesh then applying the subdiv ontop. Anyhow I could not reproduce the pinching and pulling in your example.
also sample of the smoothing groups applied. this is a way to tighten curves without having to use multiple loops just make sure the smoothing group falls on a edge that is not changing the geometery or you'll get a hard edge in the smoothed surface.
Anyhow hope this sorta helps. What you have SHOULD be working unless there is some loose vert or edge in the mesh.
Thank you so much for that post, and the one before it. I will attempt those tonight. I am at work right now, so I can't really play with for about 10 hours. I really appreciate it.
greetings
Yes, I will be going back to this. The person I get reference from was busy for about 3 weeks, and couldn't get back to me. I needed reference of the shuttlebay. I have that reference now, but got sidelined on another project while I was waiting for it. I'll swing back around to finish up the Kelvin.