Thanks again! It is quick and comes with a structural solution too, but I still cannot use it to create the whole surface. Yet I can use it for creating structural details on a single surface.
Here is what I could do
found a plugin called M+, fed the previously generated plankton mesh to ‘m+polyp’ component. It’s only the envelope (not separate panels and members) which can develop further. But with the time, probably have to use that only for renderings and create separate small section using above V1C for structural component samples.
BTW: Given the opportunity … Voronoi (done on “an one shot” basis) is 1000000% unsuitable for AEC stuff (even for Academic purposes). See what Lila reports on the V1D (strictly internal: does a lot of things more than V1C - and 500% faster).
For instance and for a minSize of 1m (for such a large structure is rather an unrealistic number: I would suggest something >3 m) this pass is 100% useless (small deviations are - obviously - expected on the Morphed segments … but that doesn’t make any real-life difference anyway).
BTW: no matter the N of passes … still you are long way from home:
You already got the premium “Asian Corporate” license with T-splines, why not just using it then? Despite that @Michael_Pryor already pointed into the right direction, instead you begin the classical odyssey.
“Don’t bother me to learn doing it the right way! I want to do it overcomplicated using Grasshopper”. Now, after 6 days, you are finally at stage 2. Congratulations. You found a plugin and you’ve got a script which somehow scratches your initial idea and now you wonder how you make anything useful and good looking out of it.
One aspect of good design is controllability and detail. And nothing is more suited for this as manual modelling. No matter if Nurbs, Sub-D’s or Polygon.
You can learn Basics about Sub-D modelling within (intensive) 3 days. Maybe this is already enough to model your desired shape. Lets invest 4 weeks and another week for modelling your idea, and you receive something really good.
Doing some rational (even remotely) envelope with SubD … well … I always admire the brave (and envy the absurdly rich clients).
But I don’t see in his question where he wanted anything rational, especially looking at the reference image he sent. Sub-D is the answer to his question
Well … at least (but not entirely) for convex blobs you can do the following:
Loop storing output [the current state of a Brep] to a parameter (volatile > persistent) and getting input [ditto] from the very same parameter (the other thing). The fact that GH frequently fails to internalize complex things … just adds to the pleasure.
Start with something boxy (and preferably stay there, he he).
For each loop “chop” portions of the brep by interactively placing a plane “around” the thing and then pressing the massacre button (the red one).
AECOSim has a similar ability … meaning Generative Components has it as well (they call it Draw on Face). Since we are talking 100% solid modeling … any impossible solid opp with R is plain easy.
The point is that what you present is probably way to complex. I’m not saying its wrong doing it like this.Its just, well, complex. Because if you never climbed a mountain, it doesn’t make sense to show him how to climb 7000+ m mountains. I can model this in Nurbs/Bezier within 1 day. But I’m not assuming he, or someone being beginner can do this. That’s why I’m a professional in my field of work, and you are at yours.
From my experience I just know that sub-d modelling is the easiest,fastest and most detailed option for someone having no clue on how to model such shape.
With the limited time ( only 6days more) I think I have to go with this for now. I prefer to go with Sub-D but too late now. Let’s say I got a very rich client and go purely conceptual as an academic project
I suspected that. Here’s my proposal: introduce me to the Holly Man (or to his daughter) and let’s do some business together [0.01 - 99.99%] > rational? > what rational? > only for the losers > dollars/zillions here we come. Sirens (a certain Tom) could/may offer you more … but trust only the Lord.
I understand now… problem is the form I have generated is way too complex. Instead, I should have started with a simple form and go into the details. This polysurface was the problem since the beginning otherwise I could have implement most of those above and developed in to something pretty comprehensive. Well… this makes a good start for the next time.
The problem is that the white thing that you posted is not at all complex (or is a nightmare) depending on your point of view and/or the methods in use in order to do it. Kinda like doing oversteer on proper cars: either easy or impossible.
For instance and for the Method described a few posts above (the “chop” - interactively - in a loop shifting from volatile to persistent data (and back) starting from a box) a result like this could be on hand in a couple of minutes - nothing to do with meshes et al (I hate meshes like my sins AND the Holly Random as reigns supreme in the contemporary Architecture). That said working on BrepFaces (boss stuff etc) is also easy. The only difference is that YOU do the thing instead of some code that attempts (in vein) to mimic … er … something/whatever.
In tricky designs like these (see that Austrian disaster) aesthetics are everything. When AI arrives to the masses (soon) … a computer could do it for you but right now you must do it “by hand” (kinda) - as Tom and Michael stated already.
I’m going to provide soon a C# that does that interactively and having the ability to store the history of your variants attempted (this means that when you reopen the definition - and provited that GH can internalize the “states” (per variant) of breps made insofar - you’ll get the sum of the states of things made in your previous work session). Kinda like recording (and recalling) history, so to speak (all feature driven solid MCAD apps [CATIA, NX etc] work providing this ability - the so called history based modelling).
But I hesitated to post a solution like this in the first place in an attempt to steer you into some more rational ways to cut the mustard: you see only the form (that overrides function) … but I’m a pro … meaning that I can instantly translate any “vision” to a myriad of nuts and bolts (and cats and dogs) that are required in order to do it in real-life (plus the dollars). That way of thinking is 1B miles away from what your tutors teach you.
But anyway always remember that the only walk worth taking is the hilly one.
… minus the interactive part (makes all the difference): the coding requires at least 1 hour to do it properly - but I do hope that you get the gist of the approach:
Just marry a human [aesthetics] with an idiot [computer] and be a happy bunny…
BTW: Lila said: if you want (interactive) “windows” this means using our internal offset Polyline stuff (Rhino offset is crap). Lord said: out of question, do some other thingy no matter how faulty/slow is (life sucks).
That said the only way to do a thing like this is this way: because the form is one thing (1%) but the other things required are faaaaaar more important:
Real-time checks with regard lot’s of Topology matters (otherwise: bananas on hand).
A way to do the massacre by adapting the form to the spaces available (or the spaces to the form) is obviously paramount: after all the scope of the envelope is to engulf actvities derived from your space schedule charts.
A Load Bearing Structure concept is a must (Or contact The Lord for the antigravity thingy: just released in Beta3 phase).
Continue … until the End of Time.
All these are impossible to do parametrically if you split the job in 2 phases: SubD in, say, Modo (the best by far) and then import the mess (i.e. the mesh) to R/GH and try to shoot your own feet (use a proper Magnum 45 [loaded] for that ).
Moral: regardless the form, try to do something that could answer to some basic questions (avoid the why).
Final offer: I raise your percentage to 0.02% (that’s double the previous): just mail me the phone of the Very Rich Client.
Now, after 6 days, you are finally at stage 2. Congratulations. You found a plugin and you’ve got a script which somehow scratches your initial idea and now you wonder how you make anything useful and good looking out of it.
