(2020-12-04, 20:13)Jeremie Guillaume Wrote: Hello,
Set 6824 "Space Dart I" is coming... But, may I ask a (novice?) question, for which I found no clear answer yet: for minifigs hips and legs, which parts should be used ? I mean, could I still use 3815/3816/3817, or do I have to use the newer 3815b/3816b/3817b ? Since as I understood, the former will be obsoleted in the short term (as seen in this thread), the latter being the preferred one to be used instead from now on ?
In fact I used the newest parts in my first three models, which I thought was best, but I admit that I'm really not satisfied with the slope on the back of the legs - especially when the minifig is sitting - while I think the old parts would be quite acceptable...
I've seen the particular way Willy (Tschager) addressed this in his recent models (for example for sets 6844 and 6980); minifigs look slighty better like that, but this lets me quite perplex since, in that case, the studs underneath don't align correctly with the stud holes of the minifig legs...
So, in short... is there a "best way" to modelize a sitting minifig legs ?
The new versions are supposed to be more correct…but note, for that to be so, the back faces of the legs on a seated minifig should not be parallel to the surface it's seated upon. (The default snap info in LDCad makes it so that they are parallel.)
But the fact is, the distinction between the old and new geometries is reeeally subtle, when viewed on real-life parts. (So much so as to be all but indiscernible, honestly.)
(2020-12-04, 21:07)Magnus Forsberg Wrote: Everytime I see this question I ask myself, - Did we make a horrible mistake redesigning the legs?
I think the motivation was good, but maybe there needed to be some more careful measurement—to a Cailliau-esque level of precision, perhaps?
Or maybe this is just another example of a case where the real-life dimensions of a part, and those that are idealized for LDraw, are just too small to warrant being modeled—just as we don't model the gaps between bricks. And maybe it's yet another instance where the precise dimensions of a part would make more sense being stored as a kind of metadata, to be applied when great precision is needed, but ignored by editors and viewers when over-accuracy is not desirable?