BiArc - each arc rendered differently when lofting??

Hi All,

First post, nothing serious but more annoying. Just doing the tutorial #6, running through the BiArcs, and when I loft them, they do not display the same rendered surface unless I reverse one of the curves in the loft panel.

I probably wouldn't mind but it doesnt seem that Rutten had to do this to achieve a uniform surface.

I haven't changed any settings from default.

Thanks in advance for any tips.

Attachments:

Capture.PNG, 527 KB

Replies to This Discussion

Permalink Reply by Pieter Segeren on April 3, 2015 at 5:42am

Hi Matt, you could just loft the list of complete biarcs coming from output B.
(What tuturial are you talking about ? - please provide a link)

Permalink Reply by nikos tzar on April 3, 2015 at 10:14am

Hello matt, Pieter,

This is David Rutten's "getting started" videos. (http://vimeopro.com/rhino/grasshopper-getting-started-by-david-rutt...).

Matt, If you mean you have to reverse one of the initial curves, then that's because you haven't designed them both with the same direction in the first place. If you mean something else, please upload your file.

cheers, nikos

Permalink Reply by matt poot on April 3, 2015 at 11:41am

Heres a copy of my file as I completed it, without the loft reversed.

Pieter, I experience better results using the loft straight from the bi-arc output, but I was just copying Rutten's moves, and got a different outcome so was a bit perplexed.

Nikos, when following the tutorial, I had to reverse the z-value for one arc as specified, to avoid the s-curve and get the single shaped arc.

Thanks for helping out.

Attachments:

Getting Started 6.gh, 11 KB

Permalink Reply by nikos tzar on April 3, 2015 at 12:08pm

I was talking about the original curves that you created in rhino (which are missing from your GH file btw). If you start one from left to right and the other from right to left then you will have to reverse one of them.

Can you also upload the rhino file so that I can check it?

Permalink Reply by matt poot on April 3, 2015 at 12:47pm

Yup, sorry bout that. I think I created them with the same orientation, but who knows.

Attachments:

Getting Started 6.3dm, 50 KB

Permalink Reply by nikos tzar on April 3, 2015 at 6:23pm

Ok, the initial curves seem to be ok.

This is what I get just by opening your files:

I don't see anything wrong. The only problem is that [join brep] doesn't seem to join the 2 surfaces (it still outputs 2 breps) but this happens often in GH.

Could you explain again the problem you are facing?

Permalink Reply by matt poot on April 6, 2015 at 8:05am

Hey Nikos,

So you have what I am expecting. But instead, I recieve this:

If I reverse one of the lofts, it gives me a uniform surface like yours, but it seems like I shouldnt have to do that. Right now, the darkly shaded arc is shiny underneath.

Maybe there is some rendering setting?

Thanks!

Permalink Reply by nikos tzar on April 6, 2015 at 8:10am

Could it just be that you have turned on the visibility in both the loft components and the brepjoin? Have you tried turning off the 2 loft component's visibility?

Permalink Reply by matt poot on April 8, 2015 at 12:37am

Yeah, I've tried turning off the loft visibility, (and everything else) and it still turns out that way.

very weird.

Permalink Reply by nikos tzar on April 8, 2015 at 4:41am

I just noticed we probably have different preview settings (Ctrl+Shift+P):

Other than that we are both getting the same result: one of the two breps has normals facing in the other direction:

This possibly starts from the initial curves, because they are not planar. It can be fixed in a couple of ways: First is reversing the curve order in one of them, like you did, or you can move around the divide curve slider (on some values the breps will be ok). or finally you can set the loft type to loose (the most stable solution):

Permalink Reply by Andres S on June 5, 2015 at 4:26am

Ok. Had a same problem and did some analyzing.

Both Case_1 and Case_2 represent baked brep_join. As you can see from the selection in Rhino command prompt case_1 brep_join failed and case_2 succeeded.

The logic behind these two geometries is the same except:

- case_1 used two separately drawn curves before lofting (coplanar)

- case_2 used one curve, copied and pasted and then lofted (coplanar)

Any edits on curves via control points seem to mess up with brep_join. I haven't found out which operation actually makes a problem. If I copy and paste a curve, and then move it completely in x,y and z axis the brep_join succeeds. But as soon you mess up with each individual control point result of join becomes unpredictable.

Join also fails in Rhino afterwards.

Also ran a scan on surface normals and in both cases they face a different direction on each arc segment of bi-arc.

RSS