[LSC] LDraw coplanarity check

[Travis Cobbs]

On Nov 14, 2007 2:07 PM, Philo <philohome at free.fr> wrote:


> > Can a math person confirm that?  (I'm doing acos(0.9) = 25.8 degrees.

> > Is this the right calculation?  If so, then for 1-degree accuracy, I'd

> > need 0.9 to change to 0.9998.)

> I agree with your calculations. Note that the problem with cosine is

> that you get very low precision on angle when it is near 0. Computing

> sine like I did would be better for that.



Good point there.  I'll think about that, although it for 1 degree
difference, we're still only talking 4 decimal places needed, which isn't
that stringent.

 > Also, what should be considered

>  > "acceptable" here.  Is 1 degree too tight?

>

>

> I made a quick experiment with ldview. The visibility threshold seems to

> be around 0.4 degree. But that's for two side by side triangles. For a

> twisted quad, even with much higher non-coplanarity I see no shading

> variation (though looking at the quad from side show clearly it is not

> flat !!!). A practical value seems to be about 1 degree indeed.

>


I forgot about the fact that while the quad is drawn as two triangles, they
both are lit using the same surface normal, so the shading will actually
make it look OK.  On that note, perhaps even 5 degrees is acceptable.

--Travis
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://five.pairlist.net/pipermail/lsc/attachments/20071114/80b43ccc/attachment.htm>