T-Splines and Grasshopper - useful together?

So, I'm a developer for T-Splines. I've been looking at Grasshopper to see how T-Splines and Grasshopper can be used together. One simple way I've seen them used together is to panel T-Splines on surfaces.

I also made a quick test that takes in a Mesh and converts it to a smooth brep. There's an example video here: http://screencast.com/t/MTA0MTRmMTkt

Most grasshopper techniques I've seen concentrate on using surfaces and parameter space, and breps are much harder to use there. The breps we output are not trimmed, so they have that going for them.

I guess my question is whether it would be useful to release something like this, or if breps are not very useful for further construction, in which case you could just bake the mesh and do a single conversion using our existing command once the mesh looks good.

More long term, it seems like adding mesh modeling tools like extrude face, thicken, weld, etc. with appropriate inputs might make it possible to do some interesting branching designs. I've held off on things like that so far because David Rutten says that the SDK is still being changed so much, and because I'm not sure how useful it would be.

Is this kind of integration useful?

----

Thanks for all the supportive feedback. I ended up making the component shown in the video available on the T-Splines forum. Please let me know either here or there if there what you think of it, suggestions, etc.

Thanks

Replies to This Discussion

Permalink Reply by to] on February 25, 2010 at 2:21pm

Hi Tom

i looked @ your video and it looks very promising
so the component just calls the convert to t-splines command or isn't it?
working with untrimmed surfaces is of course a must for really great parametrical modeling otherwise i have to recreate them on and on which is no problem to script but your tool does it in a much better way

so for me but i tested only the trial version i think it would be great to get access to your t-spline inside grasshopper and also to some of your additional commands(creases,weight,subD)
so the interesting part is not to work with the brep it is interesting to work with the surface and their parametrical setup... so we can divide and rebuild the surface without loosing the information of the geometry and curvatures ...... that would be really perfect
but what i would find very interesting ,is it possible to work on the whole t-spline surface so maybe a linkage of t-spline with their surface class to GH so that we get access the single t-spline surface to work not with the bunch of surfaces after conversation?

btw.. do you have any news when the first 64 bit t-spline beta will be published?

hope to establish a great discussion
and looking forward to the first t-spline component

-to]

Permalink Reply by Mårten Nettelbladt on February 25, 2010 at 2:46pm

Hi Tom, I see great potential in using T-Splines with Grasshopper! Thumbs up!

Permalink Reply by Wieland Schmidt on February 26, 2010 at 4:33am

Hey Tom,

I think this is a great idea! I work with T-Splines and Grasshopper a lot and the combination of both would be very inspiring and has a lot of potential. Please keep us posted about your work, looking forward to see some T-Splines definitions in Grasshopper.

Wieland

Permalink Reply by Andrew Heumann on February 27, 2010 at 8:50am

Agreed, this would be extremely useful!!

Permalink Reply by giorgio gurioli on March 2, 2010 at 2:19am

Hi Tom
I made some examples with your definition and those of Piacentino
http://mixexperience.ning.com/forum/topics/campo-unificato-uf-unifi...

Permalink Reply by Wieland Schmidt on March 2, 2010 at 4:07am

Hi Giorgio, looks really cool! Where did you get the definition from? Would like to test it myself...

Permalink Reply by Tom Finnigan on March 23, 2010 at 9:20am

I gave giorgio an early copy for testing, but I'm looking at doing wider distribution soon. I needed some help from McNeel about how to reliably find the T-Splines plugin from inside the grasshopper component.

Permalink Reply by Roger Cooper on March 22, 2010 at 5:57am