The topic that is bugging me a bit lately, is what should be really put in priority - precise geometry, or cutting down on that a bit, to have model be shaded way more nicely? Example of that would be the cut cylinder on the 50921. It features cylinder cut by the other one. In the current version, it looks like this:
[
attachment=8765]
The thin triangles lead to a pretty unpleasant shading, which essentially keeps edges very clear
[
attachment=8766]
It can be fixed by merging some of the vertices, but that results in cylinder not being really circular anymore:
[
attachment=8767][
attachment=8768]
I am not sure which approach is favorable. First result has pretty precise shaping, but loses in appearance. Second one is more fine in how it looks, and precision loss, while present, is not that dramatic
The best solution here would be to use interpenetrating cyli/cyls/cyls2 primitives. This way you get both and this is the only way to get correct hires primitive substitution.
(2022-12-04, 7:33)Philippe Hurbain Wrote: [ -> ]The best solution here would be to use interpenetrating cyli/cyls/cyls2 primitives. This way you get both and this is the only way to get correct hires primitive substitution.
In this situation, cylses can't be applied unfortunately
(2022-12-04, 9:34)Max Murtazin Wrote: [ -> ]In this situation, cylses can't be applied unfortunately
I've not checked closely, but in this configuration they should work easily. Keyword is "interpenetrating": since this part is not transparent you can hide unwanted portions of primitive geometry inside the part.
(2022-12-04, 15:37)Philippe Hurbain Wrote: [ -> ]I've not checked closely, but in this configuration they should work easily. Keyword is "interpenetrating": since this part is not transparent you can hide unwanted portions of primitive geometry inside the part.
It depends on what we want here.
A smooth surface is possible as Philo say, using cyls primitives. Also in hi-res.
A smooth surface
and a non-crooked edge line between them, is not possible, due to the cylinders beeing of different dimensions.
(2022-12-04, 15:37)Philippe Hurbain Wrote: [ -> ]I've not checked closely, but in this configuration they should work easily. Keyword is "interpenetrating": since this part is not transparent you can hide unwanted portions of primitive geometry inside the part.
I tried to do that, but it still doesn't fit too well. Cyls needs to be extended more than what length of the cylinder is, leading to condline ugliness like this:
[
attachment=8770]
(2022-12-04, 18:29)Max Murtazin Wrote: [ -> ]I tried to do that, but it still doesn't fit too well. Cyls needs to be extended more than what length of the cylinder is, leading to condline ugliness like this:
Longer cyli, shorter cyls maybe? Or cyls2?