LDraw.org Discussion Forums
48245 and 92586 - Printable Version

+- LDraw.org Discussion Forums (https://forums.ldraw.org)
+-- Forum: Models and Parts (https://forums.ldraw.org/forum-18.html)
+--- Forum: Part Requests (https://forums.ldraw.org/forum-30.html)
+--- Thread: 48245 and 92586 (/thread-6613.html)

Pages: 1 2 3


Re: 48245 and 92586 - Max Martin Richter - 2012-11-13

There is a gap between the inner torus (t01i0714.dat) and the 4-4con12.dat. Torus is 22.7514 wide and the cone 22.75 LDU. (Could be seen, if primitive is inlined.
I'll fix the other gap in the mean time. and run edger2 before I start the holes for the handle.


Re: 48245 and 92586 - Philippe Hurbain - 2012-11-13

Quote: to create additional condlines where top torus meets outer top cone
My bad, it's between green cylinder ans outer red cone.
Quote:Torus is 22.7514 wide and the cone 22.75 LDU
I don't understand - why don't you set torus radius at 24.5 instead of 24.5007??? Then fit would be perfect!


Re: 48245 and 92586 - Max Martin Richter - 2012-11-13

The funny thing is, if I set the the torus to 24.5 then the inside would fit perfect and i would have the same problem on the outside.


Re: 48245 and 92586 - Philippe Hurbain - 2012-11-13

Adjust the ouside cyli/cone/other torus that don't perfectly match anyway...???


Re: 48245 and 92586 - Max Martin Richter - 2012-11-13

What a long thread...
I had both top tori with 24.5 with this setting the inner radius of the inner torus was correct and the outside radius of the outside one was 26.2486 (or something like this) instead of 26.25.
With setting the scaling to 24.5007 the problem is moving from the outside to the inside...


Re: 48245 and 92586 - Philippe Hurbain - 2012-11-14

Max Martin Richter Wrote:What a long thread...
Yup Wink
Actually this is a direct result of rounding in primitives!
The most outer point of the torus lies at x=1+1/14=1.0714285714285714285714285714286 but primitives are rounded to 4dp so you get 1.0714. Once scaled by 24.5, the error is -0.0007. So you get 26.2493 instead of 26.25. But this error is less than 1/1000 ldu, completely invisible. And even smooth shading in LDview accounts for some rounding errors between polygon vertices and condlines endpoints. That's also the reason why in Edger2 has a "precision" option to allow some discrepancy between matching coordinates. This precision defaults to 0.001, but it may be changed if necessary.


Re: 48245 and 92586 - Max Martin Richter - 2012-11-14

Ok, I will set the torus to 24.5 and build the rest of the bucket. Not watching the "gap".
I think, I have the German problem to be too exactly ;-)
Thanks for your help.