Shark or Small Aircraft and how to describe it in fractal

[Mustafa Umut Sarac]

Hello ,

My name is Umut from Istanbul and this is my first post to the forum and I want to present my best wishes. If I understood correctly , people interested in input a code and creating a graphic. I am interested in input a 3D geometry and finding its code. If possible , shortest code possible.

I asked this question in a forum and writer told me every bit of detail must be entered to the code and code only iterated once. But I also read that all organisms have a fractal formula in their genes and it iterates and develop the organs.How can I find a Great White Shark or a small aircrafts fractal formula ?

Thank you ,

Mustafa Umut Sarac
Istanbul

[cKleinhuis]

hello and welcome to the forums ...

ouw, clearly everything has a fractal structure, and barnsley once stated that every image could be expressed through an iterative process, which is clearly right but by no means a sort of compression, in fact it is - in my eyes - nearly impossible to find an exact (iterative ) formula for representing a 3d object ... it would be even hard enough for just a tiny 2d picture ... but on the other hand putting something 3d as input into a formula could yield interesting results, but surely not the formula for the inputted 3d object ...

have fun!

[blob]

No airplane but an old horse cart made by abdelhamid belaid in K3DSurf:

http://k3dsurf.s4.bizhat.com/viewtopic.php?p=643&mforum=k3dsurf#643

And a shark here, not sure it's a great white one though:

http://k3dsurf.s4.bizhat.com/viewtopic.php?p=812&mforum=k3dsurf#812

However if those complex 3D shapes are solely defined by formulas they aren't fractals though.

Still very interesting I think

[Mustafa Umut Sarac]

Blob ,

Abdelhamids works interesting . I will look in to codes and ask questions. I will return here after this post to you and CKeinhuis.

Below post is seems to me partially conquered by your post .

CKleinhuis,

Some architects does similar things and tell the people all presented architecture done by the computer. That story and search for the truth wasted very much of my time but learning the truth was fun. They draw a building on computer and computer changes outlines as raw as possible. Result , millions of triangles everywhere and impossible to build buildinng.

I want to create flowing lines may be CFD optimized by natural selection of code or a artpiece where their curves optimized by flowing fluid .

Lets take a simpler shape , for example not a curvilinear shape of a shark but simpler version , an origami of a shark. Would it still be impossible to compress it in to a fractal code ?

Second question , I want to classify a sculpturer , designer Luigi Colanis works as compress every art piece in to a code or whatrver is possible than the see the truth and connection in tiny bit of codes , mathematics or statistics.

We dont need to make a war with curves , modern softwares converts ugly origami in to a flowing shape with increasing the triangle count to millions.

4 questions appeared :

1- Can we find iterative fractal code of a basic origami ? How ?

2- How can we find a common dimilar class fractal code or style of an artist or how can we create streamlined designs which transforms every 3d shape in to the evolution and selection of nature and time and how can we do without running a genetic algorithm code and a CFD software but with common wisdom of natural shapes ?

3- Is there a software to iterate forward a fractal code on to 3d shape ?

4- Some people try to catalog least squares of curves but .. how we can mathematically describe a 3d shape ?

Thank you ,

Mustafa Umut Sarac
Istanbul

[/b]

[stereoman]

tiburon2 por LLeonet, en Flickr



tiburon por LLeonet, en Flickr


Hi, this origami idea it´s really interesting, unfortunately I am not able to do the maths, let´s hope someone here will make a try.
The images show the building process of a shark mesh, I´ve started with a box and then, added new divisions  in some kind of fractal procedure since I go from big divisions to small ones in a recursive way, the last step is done by the program itself rounding and smoothing the mesh.

I don´t know if it´s possible to enter the mesh code in a fractal program and do something.

[Mustafa Umut Sarac]

Hello Stereoman ,

You did great job , your process is called Subdivision Modeling and present at every major modeling software.

I researched the net , looked to the concepts and found lindenmayer process. It is somewhat relates our problems and used to create different parts of a shark for example than we do merge them.

But subdivision modeling is faraway easier than to write differantial codes.

Another interestring concept is evolutionary structural optimization. You create a dense mesh and simıulate fluid dynamics problem or structural problem than the software removes the unnecessary mesh cells. Final shape is organic and very strong .

Lets talk about your examples , second iterated mesh have approximately 30 corners and surfaces.

We can take a complex sculpture , compress its surfaces with reversed subdivision modelling and get a low count mesh. Than we can deconstruct the mesh to create 2d pleaded paper.

Than we can find this pleaded papers iterative fractal formula .

What do you say ?

Umut

[stereoman]

Well, I don´t think L.Systems will solve the problem, if you want to do a shark this way you´ll end with a segmented mesh at best, but if you consider the time it will take to write these formulae, if possible, this explains why modeling programs don´t use L-Systems, you´ll need a formula for each part of each model,and you must know the result of each formula in the first place, polygonal modelling is far easy and accurate, if you forget the long years spent learning how to draw.
       Usually, the modelling process take the opposite way, you start with a low res mesh, and finish with a high res one, but the first low res mesh, starts with some 2D drawings, in these drawings the model is virtually contained, so, the modelling process is just a hand done fractalization.
         What I want to point is that you need the IDEA of the model, numbers are numbers, and recursion is just a process, nor numbers nor recursion can give you a shark, you need the idea, and the idea is not in numbers nor fractalization.
         Of course, this has no link whatever with evolutionism nor passing time.