When comes the UnrollSrf ?!?!?!?!

When comes the UnrollSrf ?!?!?!?!

I would really need it ; )

Best,

Jesse

Replies to This Discussion

Permalink Reply by Shawn Fernandez on August 13, 2010 at 12:35pm

We were discussing unrolling curved surfaces HERE

Try the definition at this ... LINK Just be sure that your input is actually a developable surface. You can tell by cranking up the number of segments and watching the accuracy difference. If it's over 1% then your surface probably isn't developable to start with.

▶ Reply

Permalink Reply by leon on August 16, 2010 at 3:03am

Hi Shawn,

Thanks for your reply.

I have seen this link before and found it interesting. Strangely enough I studied with Urban Future Organisation and although this method is interesting it still has the same problem.

What I am particularly interested in is Marc's system of using a vb script to call in the rhino command which has a more stable and accurate output. The only drawback with this is the inability to reference a point or curve to align the developed and curved surfaces together.

Ultimately I want to modify Marc's vb script to utilise the full rhino command including the 'select curves on surface to unroll' part which would solve my issue. If I can do this then the entire process (of which this is a small part) can be completed in grasshopper without the need to switch out to rhino. The problem is not the unrolling it is locating the 2 geometries around a common point.

Thanks
Leon

▶ Reply

Permalink Reply by Shawn Fernandez on August 16, 2010 at 8:34am

The problem with the accuracy in the link I gave is based on two things. First is that the surfaces being unrolled are often NOT developable to begin with, (without meaning offense, the authors of the link I gave did not appear to know the difference between developable and ruled surfaces) and the second is the number of elements used in the approximation.

Do you happen to know if the rhino command is a mathematical transformation mapping the equations of the surfaces to 2-Space? If so then it wouldn't be an approximation at all, but an exact derived transform, something that I have been seeking after for a bit as well.

I think the unroll script was not intended to produce developable surfaces that crease or fold over as if generated from a single developable sheet, but rather allow for the creation of segmented forms that could not be created from a single sheet, but where each face was a separately developable surface that would be stitched together. While that does give some added functionality to the command to create shapes that otherwise could not be made, it doesn't really help those of us desiring to use a single developable surface.

▶ Reply

Permalink Reply by leon on August 16, 2010 at 8:54am

Hi Shawn,

I don't actually know the process that the rhino unrollsrf command uses but the results are more consistent than any other method I have used. And since this seems to be such a big topic about implementing the rhino command in a scripted form I am guessing that others consider this the most reliable method of unrolling surfaces too.

In the instance that I am using this command there is no requirement to create all parts from a single developable piece but rather to make singular panelisations that can be unrolled individually as you suggested. Marc's vb is the closest I have found to replicating the rhino function although misses out a crucial part for my requirements. This link explains the process in rhinoscript but I am not experienced enough in vb to translate this into Marc's component.

After the select surfaces line I would need to insert this part of script into the vb which would enable to full rhino command to be used through the grasshopper interface.

' Format curve string
Dim i : i = 0
Dim strCurves : strCurves = " _Enter"
If IsArray(arrCurves) Then
strCurves =
For i = 0 To UBound(arrCurves)
strCurves = strCurves & " _SelId " & arrCurves(i)
Next
strCurves = strCurves & " _Enter"
End If

Any suggestions how to incorporate this???