Quake3World

Hr.O wrote:This is how you should do it

You rock. I was going to do the same thing a couple of nights ago, but got sidetracked, and decided I would check back in tonight. Thanks for saving me the effort.

First off all, I'm not a fan of private messages that are about stuff that can be posted in public, certainly if it's about how to map. Second, I in theory could've gone away for months again from this forum, and thus not be able to answer your question.

How did you make the patch mesh for the top and bottom?

When i try to make the mesh, I can only cover 7 sections with max. amount of rows or columns. If I try to add more rows or columns, it will split my existing areas in half.

make brush, hit shift-p keep density at 3 by 3.
Then with patch selected hit the v button, you'll see a bunch of pink and green dots and lines. The dots are called vertexpoints, the lines vertexes.
Now take the top left vertexpoint and put it at the beginning of the top-most bevel, next you move the topright vertexpoint and move it to the end of the same bevel. last bit is the middle vertex, create an imaginary horizontal line from the first vertextpoint, and a vertical one from the second vertexpoint, the third vertexpoint goes on the imaginary crossingpoint of the two lines. That's half the work done. I think you know how to do the other half.

Here's a four step plan:

To get caps in this case you can also bend square cylinder then split it to 4 meshes if your radiant comes with bob tools.

Hr.O: Are you using 1.5? The older radiant versions automaticly adjusted to the adjacent patch, but in 1.5 it doesnt, leaving holes between the curve and the top patch.

I have been using 1.5 in the past, but went back to 1.4 mostly because the missing z-bar reduced productivity. BTW, what you say about patch handling sound very irregular to me. Simple maths tell me that vertexpoints dictate the shape of in this case patches. So similair curves should fit together. If not something would be wrong with the maths behind GTKRad 1.5
However thinking about the issue here vaguely brings something about snap-to-grid working differently, or not being on by default in 1.5 to mind.

Fjoggs wrote:Hr.O: Are you using 1.5? The older radiant versions automaticly adjusted to the adjacent patch, but in 1.5 it doesnt, leaving holes between the curve and the top patch.

the outer bevel and mesh are aligning; however, it's the inner bevel and the mesh that is NOT aligning.

You sure vertexes are aligned to grid? How about after compiling?

obsidian wrote:You sure vertexes are aligned to grid? How about after compiling?

Yep, every thing has been Ctrl + G and it's the same after compiling.

even though you say they are on grid, the picture shows other wise... can you absolutely confirm in radiant that the vertexes are in fact on the grid?

Well, I guess I can't be certain on how to answer that.

If it means that the 2 ends of the patch are on the current grid, then yes it is on the grid.

However, if you mean that at each bend in the patch is on the curve, then no.

Is this another great feature that 1.5 has?

If I select the patch and press Ctrl + G, should it not be "on the grid"?

For the patch to be considered "on the grid" does that mean just the 2 ends need to be on the grid OR does that mean that at every point in the patches that bends needs to be on the grid?

1.5 does that if the patch is 32x32

Fjoggs wrote: 1.5 does that if the patch is 32x32

I don't quite understand what you mean by that?

If you make a bevel on a 32 grid and you make the mesh on a 32 grid and you adjust them on the grid (ctrl + g) wouldn't both of them be treated the same way.

or is the mesh being adjusted to the grid by the vertex points (green & pink), not the actual edge where it is bending?

I believe I even tried to adjust the patch and the mesh to a lot small grid size like 1 or even .5 and I was getting the same results.

Pardon me, I meant 16x16 units. If you drag the corner vertice, you'll see how it "adds" more divisions. and reduces it when you've aligned it correctly. The previous versions didnt do this, making it look like a proper curve.

highlanderstl wrote:Well, I guess I can't be certain on how to answer that.

If it means that the 2 ends of the patch are on the current grid, then yes it is on the grid.

However, if you mean that at each bend in the patch is on the curve, then no.

Is this another great feature that 1.5 has?

If I select the patch and press Ctrl + G, should it not be "on the grid"?

For the patch to be considered "on the grid" does that mean just the 2 ends need to be on the grid OR does that mean that at every point in the patches that bends needs to be on the grid?

right, i was speaking without thinking

so yeah, make sure the vertexes ('v') are on the grind, and generally everything should be alright. each patch should be subdivided the same way in the editor. when you align the patches that sit on top of the curves, make sure the subdivisions line up. sometimes radiant can be finicky, so you might have to alternate between patches and do some vertex manipulation to make each patch snap into place with the right subdivisions.

When you hit "v" the green and pink vertexes are called control points. Those have to be on the grid. The actual mesh vertexes don't.

obsidian wrote:When you hit "v" the green and pink vertexes are called control points. Those have to be on the grid. The actual mesh vertexes don't.

Thanks for pointing out that my mapping terminology is a bit rusty

this thread has felt a bit nostalgic... good stuff

highlander, somewhere in your inner patch, one or more of the controls points are off grid. One way to see for sure, if you havent done it already, is to zoom in as close as possible to it and I am sure you will find the offending point(s).