unmatching primitives (4-4ndis and 4-4ring6)

[Bertrand Lequy]

Hello I'm working on 87580p02  and I noticed that there a problem here :

1 0 -13.4 0 -8.8 3.5 0 0 0 1 0 0 0 3.5 4-4ndis.dat
1 71 -13.4 0 -8.8 0.5 0 0 0 1 0 0 0 0.5 4-4ring6.dat

The 22.5 [45] do not match, is it a problem ?
for exemple : -10.16635,0,-7.46055 for the ndis and -10.1664,0,-7.4606 for the ring6

[Philippe Hurbain]

Hello I'm working on 87580p02  and I noticed that there a problem here :

The 22.5 [45] do not match, is it a problem ?
for exemple : -10.16635,0,-7.46055 for the ndis and -10.1664,0,-7.4606 for the ring6

This is an unavoidable "problem" related to limited precision of numbers in primitives combined to scaling.
See a similar issue in 3626 minifig head: https://forums.ldraw.org/thread-28146-po...l#pid53415

[Rene Rechthaler]

I think it is the same with torus prims:
for example, if you use the outer quarter prims, the 3 ends dont always match connecting prims...
(if you use it to chamfer a rounded corner)

[Bertrand Lequy]

This is an unavoidable "problem" related to limited precision of numbers in primitives combined to scaling.
See a similar issue in 3626 minifig head: https://forums.ldraw.org/thread-28146-po...l#pid53415

So, I shouldn't worry and let it go ?

[Philippe Hurbain]

So, I shouldn't worry and let it go ?

Let it go! This "problem" exists for 25 years so it must be managed by all LDraw tools... Anything below regular part rounding (0.001 per specs) must be accepted when unavoidable.