FlowAlongSrf Error

What causes this type of weird distortion with Flow Along Srf?

FlowAlongError.3dm (117.1 KB)

Hi Eric - it does not like that the curves are joined - not sure why, yet, but if you explode the input curves and Match the bits for Curvature with Merge set, they work OK. There is something about that fully multiple knot at the join in the curves that is giving it fits.

So far V6/WIP likes these curves fine.


I didn’t even think about there being a joint there. That probably explains all the other times I’ve come across this.

I usually see it when I’m flowing _TextObject (bold) curves from a plane onto a cylinder. It really threw me when it had problems flowing relatively light curves from a cylinder to a plane.

Well, now I’m getting more weirdness. I unjoined the crvs and they flow fine. But then I tried to flow some block objects from the cylinder to the plane and they jumped off the plane as well.

FlowAlongError2.3dm (243.4 KB)

Hi Eric - I see that - weird - also fixed in V6/WIP but I’ll see if I can figure out a workaround for 5.

Well… as far as I can see any block instances that are on the counter-clockwise (negative U direction) side of the middle of the cylindrical surface are incorrectly mapped. No idea what is going on… if you rotate the cylinder to put the seam just to the counter clockwise side of the left most instance (as the file opens) it works.


1 Like

Hey @pascal,

This is still a problem and it’s becoming very persistent.

FlowAlongError.3dm (1.1 MB)

Hi Eric- this does work in V6/WIP and the workaround for V5 is the same - move the seam (SrfSeam) on the base surface around to one side of the objects:


I rebuild your curve, and it is ok.

hi @EricM:
May be they are blocks, after I Explode them, they are ok.

@pascal, can you see if this happens with previous v5 builds? I’ve used Rhino v5 since the beta, but I’ve only recently had problems. It’s a major PITA. While the workaround fixes the issue, it blows a hole in my rather complex history tree.

I also use blocks extensively, but they have always acted weird. I email about them some years back. When you mirror a block, the closed polysrfs/meshes can render inside out. The interior backfaces which should never be seen are rendered. The normally rendered exterior is not rendered (i.e. invisible).

Hi Eric - In your last example blocks are not involved, correct?


The first two files I uploaded show FlowAlongSrf getting wonky with crvs, then blocks. The latest file I posted shows it going wonky with srfs. I’m wondering if a recent build introduced this bug.

The bit about blocks not rendering correctly is separate. That bug has to do with _Mirror and has been in V5 since day one. It’s only a problem when I need to show a model to someone else. That is the only time I explode blocks.