Holes pattern over a curved surface

Become a member of the CGSociety

Connect, Share, and Learn with our Large Growing CG Art Community. It's Free!

THREAD CLOSED
 
Thread Tools Search this Thread Display Modes
Old 05 May 2013   #16
Depending on the version of max you have, you could always use the conform brush with your model on a version of your model that doesn't have the holes in it, to match the correct curvature for each vertex. I think the conform brush was added in 2012. If you have an earlier version, I believe you can download scripts from scriptspot that have similar tools for conforming and retopologizing.
 
Old 05 May 2013   #17
Hello,

Well, I gave Populate Panels a shot and it wasn't bad. Actually ended up using it on something else. Very useful and easy to use. On this model 'though I couldn't achieve the desired outcome. The main problem was that it scatters one object (in this case a hole) for each face which meant that if I wanted a lot of holes all together, I needed a very high poly mesh. The problem with that was that I lost smoothness when I subdivided as a result of the high number of edges next to one another.

I still haven't discarded it 'though. I'm thinking that maybe I can subdivide de original mesh first, and then give it a try with the holes. We'll see.

The trick with the conform brush was a pretty smart idea. Again, as the perforated mesh was higher poly than the original one, even after the conform it wouldn't smooth perfectly.

It is highly probable that any of these methods fits this problem perfectly and I am the one messing it up, but that is why I am trying them all and moving back and forth with this, so I can learn from the experience.

In this process I found out about Parametric Array and figured that it was exactly what I needed. (At least that was the kind of process I first imagined I would use to model this). Too bad it's a 3rd party script. :(

Anyway, I'm mostly posting this because from what I've seen on the net this is a somewhat complex problem that many people come across with, and for any other newbs like me out there it may be useful to read about some of the mistakes I've done on this, so... I'll obviously update when I find the best solution.

Regards ^^
 
Old 05 May 2013   #18
Oh yeah! Parametric array! Forgot about that goodie!

Yeah, you've git a more challenging thing going on there... It's almost like the holes need to be placed on the mesh but kinda in the way of UVW unwrap mapping. Just dont know a way to do something like that.
 
Old 05 May 2013   #19
maybe this will help:
http://torabiarchitect.com/parametric-array/?p=817

well worth the money!
 
Old 05 May 2013   #20
Yeah this definitely looks like a challenge and there are lots of ways of going about this. In my opinion, you are probably going to need a higher resolution mesh to get this to look smooth, and will probably require some work. The way I would approach this to get even smoothing for everything is-

- First model the general form of your object, with evenly spaced quads.
- Make the hole patterns out of extruded shapes.
- Subdivide the form until its roughly the same resolution as the holes.
- Copy the object.
- Use ProBoolean to cut the holes out of the copied mesh.
- Clean up any vertices that are only shared by 2 edges and any vertices sharing the same space.
- Remodel edge flow in areas along the holes and around the holes, add edge loops around the holes, and try to keep everything in quads.
- Use the conform brush to reset the positions of any vertices that aren't on the suface of the original mesh.
- Add shell modifier to add thickness last, and delete original mesh.

Last edited by Ian31R : 05 May 2013 at 09:16 AM.
 
Old 05 May 2013   #21
Yes timd1971 that's it. It's very powerful and I donít think it's even that expensive when compared to many other 3rd party softwares (considering what it does). But this thing is just a side project for me, so I donít plan on going out of budget with it. Anyway, turns out it was free for download just a few months ago, and now itís not anymore. :(


About the model, I have been trying not to have to rebuild it from scratch because it is part of a piece that ought to fit on top of another just like it. So it took a bit of patient modeling to make sure there were no intersecting faces, which I would like not to throw to waste now. But given the fact that I am not going anywhere with the way it is now, I may have to do just that.

Ian31R, how would you position the holes objects around the mesh? Also, any ideas on how to achieve an evenly spaced quadrified mesh on such an organic shape?


Well, thanks once again to everyone who is posting suggestions. I really didnít expect so many replies and would probably have given it up already if it weren't for all the solutions that you presented. Appreciated.

Last edited by Octavarium : 05 May 2013 at 11:48 AM.
 
Old 05 May 2013   #22
Hello,

I had to put this aside to work on some other things, but didn't want to give it up, so I caught up with it a couple of days ago.

Basically I did the following:

Created a plane and used Generate Topology on it to get the pattern. Then used conform to apply it to the mesh; deleted the correspondent mesh faces, attached the plane to it, and then corrected the geometry with few hand cuts, welds and caps.

The topology was a bit of a problem. Couldn't have too much edges because of the subdivision as I mentioned earlier, so making it all quads was an issue. I still haven't figured it all out, and it's looking pretty ugly, but the objective was mostly to understand whether it was a viable way of doing it or not (see, problem with us newbs is that only after two days of work we realize the mess we're making).

This is the mesh with shell and subdivision


Without subdivision


And these are the smoothing issues it has right now.



So overall I think I will proceed with this method. From what I've been trying on this model and what I searched about the "holes on curved surfaces" thing, I would say the best way to do it is probably with something like Parametric Array. The Panels Plug-in is very cool but for something like this only works if you have a high poly mesh, and preferably with evenly shaped poly's. For this specific model, this method was the one with a better amount of work\rigorous outcome relation. I hope any of this can save anyone some time in the future.

Many thanks to everyone once again.

You can see the WIP thread of this model here if you want. :P
 
Old 05 May 2013   #23
Yeah, those smoothing issues are what is always going to be a problem. When you have larger polygons overall it's more apparent, which is why you have to increase the polygon count to reduce how visible those issues are.
__________________
The Z-Axis
 
Old 05 May 2013   #24
Thread automatically closed

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.
__________________
CGTalk Policy/Legalities
Note that as CGTalk Members, you agree to the terms and conditions of using this website.
 
Thread Closed share thread



Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
CGSociety
Society of Digital Artists
www.cgsociety.org

Powered by vBulletin
Copyright ©2000 - 2006,
Jelsoft Enterprises Ltd.
Minimize Ads
Forum Jump
Miscellaneous

All times are GMT. The time now is 08:17 PM.


Powered by vBulletin
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.