LDraw.org Discussion Forums
Inconsistencies between part files for similar primitives - Printable Version

+- LDraw.org Discussion Forums (https://forums.ldraw.org)
+-- Forum: General (https://forums.ldraw.org/forum-12.html)
+--- Forum: General LDraw.org Discussion (https://forums.ldraw.org/forum-6.html)
+--- Thread: Inconsistencies between part files for similar primitives (/thread-26293.html)

Inconsistencies between part files for similar primitives - Another Mike C - 2022-05-02

Hi, new user here.

For some background, I've been looking at using the LDraw parts catalog as a starting point for 3d printing a few missing parts from some of my original 1970s space lego sets I kept from when I was a kid - set 924 for example. I could just buy the pieces of course, but trying to print them myself seems like a lot more fun :-).

I know the caveats around measurements in the LDraw catalog being approximations of the real physical dimensions, and being optimised for on-screen rendering, etc, so please say something if I'm stirring up old discussions and maybe just point me in the direction of those so I can read them.

Having said that, while I've been experimenting I've found some general inconsistencies between similar primitives in LDraw parts catalog and was I hoping someone could help me understand if they're "by design" or just historical drift between individual files, and whether there'd be any interest (or not) in trying to iron them out if they're not expected. Some of the parts haven't been modified for over a decade so I appreciate it might just be better to let sleeping dogs lie at this point :-).

There's a bunch of different types of inconsistencies I've found - I won't go into all the details unless there's some positive value to the community, but as a single example the control points for the conditional lines on the ends of part "p/1-4cyli.dat" look like the green lines in this screenshot and they appear to work as intended when the model is rotated (Note: I've attached the screenshots as well in case the Imgur posts get deleted):

[Image: 84wFt.png]

but the control points for the conditional lines at the ends of "p/2-4cyli.dat" look like this and don't really show the conditional outline on the end edges correctly when the model is rotated:

[Image: IW9I9.png]

I'm kind of new here (only been lurking for a couple of weeks), so there might be a legitimate reason for these odd-looking control points, in which case I'd be interested in what that is so I can understand it better.

In any case, if this kind of feedback is useful please let me know. And if it's not, please say so as well and I'll avoid spamming the forum with lots of irrelevant glitches :-)

Cheers,

Mike

RE: Inconsistencies between part files for similar primitives - Travis Cobbs - 2022-05-02

(2022-05-02, 20:03)Another Mike C Wrote: There's a bunch of different types of inconsistencies I've found - I won't go into all the details unless there's some positive value to the community, but as a single example the control points for the conditional lines on the ends of part "p/1-4cyli.dat" look like the green lines in this screenshot and they appear to work as intended when the model is rotated (Note: I've attached the screenshots as well in case the Imgur posts get deleted):

[Image: 84wFt.png]

but the control points for the conditional lines at the ends of "p/2-4cyli.dat" look like this and don't really show the conditional outline on the end edges correctly when the model is rotated:

[Image: IW9I9.png]

Neither of your images is correct, so your conclusions are also not correct. I'm not sure where you got the two endpoints for your conditional lines, but they do not match the data in the files.

The conditional lines on the edges of the n-fcyli files are set up to be tangent to the cylinder. That way if two cylinders are put next to each other, each will show its conditional line half the time. If a cylinder is placed against a plane that is tangent to the cylinder, the conditional line will appear as expected, and (more importantly), not appear when it shouldn't.

In LDView, enable conditional lines and then select "Show all" and "Show control points" to see the actual geometry.