F***k consoles.
Pretty sure physics break more on lower framerate.
what is CCD ?
What is adventage of using blender for colision instead unreal ? i want also do that with a sphere with quite big facet . like visible facet so lets say 3 time size of the player for each poly.
will the physics works with that ?
Ccd is an option in the physics setting - what it essentially does is it makes it so that calculations happen more than once per frame based on the setting you pick.
There is only advantages to using blender really.
I would suggest you scale the model up to the appropriate size, and then work out how to cut it up into hexagonal shapes.
With the right shape, you will only need one shape mesh in engine and can automatically instance the tiles into the closed sphere - in a way that the individual tiles are still able to occlude…
Hi, you’re right, but is there a more or less working method that is both optimized and efficient?
Dividing it in hexagons (beehive) is probably the best option you have.
I havent really seen any other structurally sound divisions that allow you to re-use the same piece over and over to generate a full sphere.
Was it possible to make not from honeycombs, but from triangles?
Yes, but the triangle won’t be the same size and shape everywhere.
Then again, The honey comb could be given a center point that generates all triangles which means that in theory you could have it be made up of all the same triangles.
However, arranging said triangles in space is going to be really a lot harder than it would be to arrange instances of hexagonal pieces.
There are math formulas for how to split a sphere in recursive shapes. You should look into that - and some sphere math basics since you will need to know it to place things around.
I never used blender but in unreal you can select how much triangle you want to use to do the sphere
so that pretty simple just select low amount like 50 and you got pretty low poly sphere perfectly arranged
Viable option, but you likely need to export and breake it apart to extract a single triangle that can be re-used - and recomposing it off of that will mosy likely involve the triangle having to shift/skew if not scale up and down to fit.
Dodecahedrons, vs icosahedrons.
The only way you have accurate triangle only is a D20.
Further extrapolations of that go in increments of how to subdivide each triangle - so thats 4, 9, 16, 25, etc.
But, the tris you generate are all twisted/rotated up/down.
It is based on an obvious pattern (one up, adjacent down).
So yea, you could probably build the math to put it togeter like that as well.
I still think Hexagons are easier to deal with though.
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.