Logo by MarkJayBee - Contribute your own Logo!
News: Follow us on Twitter
 
*
Welcome, Guest. Please login or register. April 16, 2014, 09:14:26 AM


Login with username, password and session length



Pages: [1] 2   Go Down
  Print  
Share this topic on DiggShare this topic on FacebookShare this topic on GoogleShare this topic on RedditShare this topic on StumbleUponShare this topic on Twitter
Author Topic: 3d fractal automata  (Read 1461 times)
0 Members and 1 Guest are viewing this topic.
Tglad
Fractal Bachius
*
Posts: 585


WWW
« on: March 30, 2012, 10:15:15 AM »

I have made a 3d version of my fractal automata finder, it is available next to the download for the 2d automata finder (which is at this page: https://sites.google.com/site/tomloweprojects/scale-symmetry/automataFinder), click the down arrow at the bottom right of the page to download 3DFractalAutomata.

No moving ones yet actually, but plenty of stationary ones, here are a couple of screenshots-


* 3dfractals.png (236.6 KB, 767x767 - viewed 50 times.)

* fractals2.png (121.58 KB, 756x759 - viewed 33 times.)
Logged

kram1032
Fractal Senior
******
Posts: 1489


« Reply #1 on: March 30, 2012, 03:26:10 PM »

Oh, awesome stuff cheesy

If you get a V7 equivalent going with those, maybe you get something like houses?
Because lots of the V7 results looked like nice house layouts...

Can you add cutting planes to reveal the inside structure of some of those more cuboid ones?
Logged
Alef
Fractal Schemer
****
Posts: 971



WWW
« Reply #2 on: March 31, 2012, 07:51:18 PM »

I still would stand for a colours. Maybe just gradient, how it developed.
Logged

P.E.A.C.H. - please examine and critique honestly.
Tglad
Fractal Bachius
*
Posts: 585


WWW
« Reply #3 on: April 01, 2012, 09:36:18 AM »

Thanks for the suggestions, kram, I'll add cutting planes to the todo, you can already see it sometimes in the big objects, they hit the front pane and you can see their insides.
Asdam, I could put in colours when I get time to make a GUI for it. I'm not sure how to colour, a lot of the results are like rocks and sponges like the pictures below, so only subtle colouring would seem appropriate, and you can get some from the colour of lights.


* rock.png (236.45 KB, 511x501 - viewed 35 times.)

* sponge.png (139.97 KB, 444x479 - viewed 470 times.)
Logged

kram1032
Fractal Senior
******
Posts: 1489


« Reply #4 on: April 01, 2012, 11:32:02 AM »

(what the hell, is there an april fool's thing happening? Either way...)

Nice! I'd actually like to see what physical properties such structures have if printed in Shapeways or alike...
Logged
Tglad
Fractal Bachius
*
Posts: 585


WWW
« Reply #5 on: April 06, 2012, 01:28:33 PM »

Thanks Kram.
The moving version is coming along ok, the main challenge is getting it fast enough to use, seven 2563 arrays all updating at once is quite heavy, so I'm thinking of adding threading so it can make use of multiple cores. I don't want to get into GPU coding though.
Here's another random example, some of them look a bit like crystals, feel free anyone to try it out.


* crystalRock.png (184.83 KB, 465x512 - viewed 43 times.)
Logged

DarkBeam
Global Moderator
Fractal Senior
******
Posts: 1650


The spaghetti formula coder


« Reply #6 on: April 06, 2012, 04:25:56 PM »

Looks somewhat similar to my Mixpinski4 plus a bit of "randomness" cool. A Beer Cup
Logged

Formulas are never too much (?)

s31415
Conqueror
*******
Posts: 110



WWW
« Reply #7 on: April 07, 2012, 12:14:17 AM »

Really awesome! I featured them here:
http://algorithmic-worlds.net/blog/blog.php?Post=20120406

Best,

Sam

Logged

kram1032
Fractal Senior
******
Posts: 1489


« Reply #8 on: April 07, 2012, 10:23:43 AM »

I wonder, since you'll use the smaller grids aswell...
Do you think, it would be more efficient to cleverly embed the lower resolutions into one big grid instead of having multiple smaller and smaller ones?
I'm really not sure... Just a random idea.

Moving stuff? yay smiley How does it work now?
Logged
Tglad
Fractal Bachius
*
Posts: 585


WWW
« Reply #9 on: April 15, 2012, 09:07:54 AM »

Hey thanks Sam! I'm delighted that this is featured in your blog.
Thanks for the idea Kram, that could help, certainly different memory layouts can make a difference. I'm using the lower res grids for early culling on the static ones.

The moving automata are in now and in the link giving in the first post. I have also added a small summary on that page.
Unfortunately I had to half the resolution to keep the frame rate decent. It would probably run faster on a fragment shader but that isn't a certainty. If you have a beefy computer and it is running too fast then let me know and I can put in a higher resolution option.

At the moment the best dynamic type seems to be type 6.
Logged

kram1032
Fractal Senior
******
Posts: 1489


« Reply #10 on: April 15, 2012, 09:04:01 PM »

Do you think you could do variable resolutions? It should be fairly easy to basically procedurally generate higher resolution levels... at least I expect it to be. It's all the same but bigger, right?
Logged
Tglad
Fractal Bachius
*
Posts: 585


WWW
« Reply #11 on: April 18, 2012, 02:13:52 PM »

If you mean varying the resolution to fit to a frame rate then that isn't feasible, the jumps are too big, when you double the resolution it takes at least 8 times the CPU time and memory. But I could put in a hi res mode, if anyone has a computer beefy enough to need it.

The new version lets you try out a few different seed shapes for a particular rule. The pictures below show a type 3 rule with an O as a seed, a C and the normal random seed. You press 'o', 'c' or 'r', its in the info. The 'o' seed gives a big variety of kind of snowflake type shapes, they're not all a perfect O like the one below.


* rockO.png (110.71 KB, 467x466 - viewed 340 times.)

* rockC.png (62.63 KB, 296x475 - viewed 334 times.)

* rockRandom.png (135.58 KB, 501x412 - viewed 23 times.)
Logged

kram1032
Fractal Senior
******
Posts: 1489


« Reply #12 on: April 19, 2012, 12:56:01 AM »

I didn't mean dynamic, no. Just set it once and the go with it or restart from scratch if you want a different resolution.
Those o/c/r are an awesome idea cheesy
What about I or .? Those seem reasonable "further steps"
Logged
Tglad
Fractal Bachius
*
Posts: 585


WWW
« Reply #13 on: April 21, 2012, 12:45:17 PM »

There is already a . and Z and X, its in the info in the command window when you run it.

I have updated to include a type 5 and type 6 static. These are quite nice in that they more or less cover the set of simple neighbourhoods.

Here's a type 6-


* crystal.png (146.24 KB, 512x510 - viewed 23 times.)
Logged

Tglad
Fractal Bachius
*
Posts: 585


WWW
« Reply #14 on: April 29, 2012, 07:23:03 AM »

Update:
- Type 1 changed to a better rule set
- New types 7,8 and 9 are asymmetric in the vertical axis, so perhaps more appropriate for earth features which have gravity
- New option: 'g' side on view of a random 4x4x2 voxel terrain.
- New option: 'h' top down view of a random 4x4x2 voxel terrain.

So for example, arid.es9 looks like this from above and this from the side:



* aridTop.png (203.22 KB, 509x508 - viewed 20 times.)

* aridSide.png (89.22 KB, 508x510 - viewed 22 times.)
Logged

Pages: [1] 2   Go Down
  Print  
 
Jump to:  


Related Topics
Subject Started by Replies Views Last post
Most Chaotic Cellular Automata Other types Timeroot 12 5323 Last post March 18, 2010, 07:29:03 PM
by kram1032
Fractal Explorer - fast fractal generator for Android Smartphones / Mobile Devices Black 6 2347 Last post November 29, 2010, 10:18:20 AM
by Cyclops
probability-based fuzzy cellular automata? General Discussion kram1032 0 654 Last post March 12, 2011, 02:47:11 PM
by kram1032
Where do the Fractal Science Kit users congregate for fractal posting. Fractal Science Kit wmauzey 4 762 Last post February 27, 2012, 12:39:58 AM
by wmauzey
Melting Cellular Automata Mathematics fractower 1 797 Last post April 24, 2012, 03:47:41 AM
by jehovajah

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2013, Simple Machines

Valid XHTML 1.0! Valid CSS! Dilber MC Theme by HarzeM
Page created in 0.29 seconds with 28 queries. (Pretty URLs adds 0.023s, 2q)