Ledge under slope bricks (very minor details)

[Jean-Philippe Ouellet]

All slope bricks have a little ledge so that the 4LDU space is left for the stud to fit in without intersecting anything.

However, 3675 (this brick's convex counterpart) does not:

I find this odd because in every other slope brick I can find, the inner slope and outer slope are parallel, except 3675:

This could just be because the ldraw model was directly derived from a scaled version of 3045:

(which by the way, has the ledge in real life, although it may just not be worth modeling due to its size.)

Compare the following photos:



The ledge of 3675 seems to be smaller than that of 3298 (left) (smallest modeled in the library) but larger than that of 3037 (right) (not modeled).

Since this part is the only one I've seen where the under side of the ramp is not parallell to the top side of the ramp, is it okay to assume that it should be there, and update the part?


Also worth noting is the fact that the ldraw version of 30363 has the ledge come all the way to the edge of the underside tube:

but the real-life version does not:

and the top and bottom seem to already be parallel:

What is the proper solution?

[Tore Eriksson]

Talking about very minor details, I dimly recall a discussion some time ago about the inner "ceiling" of bricks, plates, and tiles aren't exactly 4 LDU below the top, but closer to 3 (or was it 5? No I don't think so, then there wouldn't be room for studs inside tiles and plates) LDU.

Can anyone confirm this?

[Tim Gould]

I'd like to see the parts made consistent.

So perhaps you could fix the concave one to match the others. And check the new convex one too.

Tim

[Jean-Philippe Ouellet]

Gladly, but how long should the ledge be? That is also inconsistent, and from the real life parts, the 3x3 convex doesn't seem to have the same ledge width as the 3x2 33 degree.

If by new one you mean concave one? Then yes, it has a ledge, but it also has overlapping pollies in the underside tubes, which I don't think is proper.

[Philippe Hurbain]

There are a lot of compromises in LDraw parts compared to reality, and especially in slope bricks. While some of them can be solved relatively easily, others can't be fixed without rewriting all the slope bricks!
The two main discrepancies are:
- As mentionned by Tore, ceiling of bricks is thinner than 4ldu. This means that the minimum height to receive a stud is about 5ldu (5.25 according to LDD parts)
- The straight foot of slope bricks is also higher, modelled as 4ldu in LDraw, but about 5 LDU in real life.
When you combine this with slopes, you get things like ledges that are too long or disappear completely...
Note also that even true LEGO parts are not always consistent with each others

it also has overlapping pollies in the underside tubes, which I don't think is proper.

This is discouraged but not forbidden by LDraw standards:

Note: Primitives that overlap/intersect other primitives or polygons (line type 3 or 4), provided that the overlap is small, are considered acceptable. The intention is not to force part authors to in-line primitives. However, there may be some cases where it is better to use (say) a 3-8cyli and a quad than a 1-2cyli; case-by-case decisions are left to the parts review process.

...and this may give better rendering sometimes!

[Tim Gould]

Given Philo's response you could make them both (convex and concave) consistent with the regular slopes

Please note that I'm *not* putting pressure on you at all. Only do it if you feel like it. I'd like to see them consistent but it had never bothered me until you pointed it out.

Tim

EDIT: Whoops. That was really quite the opposite of what I wanted to say.

[Jean-Philippe Ouellet]

Thanks for the feedback.

Done.

[Tim Gould]

Thanks. My inner anal-retentive is happy.

Tim

[Greg Teft]

I noticed the ledge some time ago when sorting parts back into their sets, and also that the thickness of the material except for most walls is thinner - about 3 LDU. This is consistent across all part variations. I think studs are also slightly taller than 4 LDU as well.

Also a number of elements have variations in wall thickness and use of stiffeners such as the 1 X 4 brick.
For this element, there are many different combinations of wall thickness and ribs or raised surfaces. I'm guessing this is to subtly change the weight of a commonly used element in order to verify contents of a bag or set by using a scale. I think the little open rectangle on some 1 x 1 'headlight' bricks is due to this - a combination of thinner wall and accommodating the connecting stud on the bottom. I have some where this is feature is closed, but very, very thin.

I don't think these variations are necessarily important for LDraw as the 4 LDU material thickness seems to have been around a long time, and established parts probably ought not to be updated.

Any feature that would cause an incorrect collision should be modeled however. I would say that this should include the inside ledge on longer slopes in as much as the visible notches on longer inverted slopes. Also, I noted once that an axle-hole should accommodate a 4LDU diameter tube, I don't know if the axle-holes had been updated to correct this or not.

[Philippe Hurbain]

I don't think these variations are necessarily important for LDraw as the 4 LDU material thickness seems to have been around a long time, and established parts probably ought not to be updated.

Yes, the main goal is to keep LDraw system self-consistent imho.

Also, I noted once that an axle-hole should accommodate a 4LDU diameter tube, I don't know if the axle-holes had been updated to correct this or not.

Good point (I guess you mean 4LDU radius tube . Problem is that thickness of axle arms are modelled at 4LDU, the real thing is closer to 4.5LDU. Changing axlehole primitives would be easy(*) but that would impact the construction of many bricks around holes. Perhaps just blunting a bit axlehole teeth?

(*) Actually the primitives would be simpler since we could use a simple eighth division of 6 ldu hole! (2*6*sin(22.5°)=4.6)

[Greg Teft]

Yes - I meant radius - it was late.
I thought blunting the teeth with a single face chamfer, or 2 or three faces if greater detail is desired for the roundness would suffice rather than make the axles & axle-holes larger.