« Help, when is a number a Number? | Main | Searching for Bob Crimi »

VRML...not quite dead yet.

When SGI dropped VRML, years back, it looked like the end of VRML.

Well, somebody's got to be using VRML because ParallelGraphics is still making VRML authoring tools.

I have a soft spot for VRML, and would love to see a revival.

ParallelGraphics (a 3D VRML company)


I've used vrml a few times...Among other things I used it to visualize strange attractors in 3D...I have one here (600k)


Not much really...but you asked...


Hey Bo, great to hear from you!

I tried viewing your VRML file, without luck. I downloaded the cortona VRML client, but it wouldn't open your file :( It gave me a message saying that I needed a converter (for VRML 1), so I downloaded that...also no luck. I'll try again later, though. The auto-install may not have worked right.


I don't have a VRML plugin installed here, but had a look thru yer wrl file... Is that basically a huge list of points in 3D space? If so, it wouldn't be too hard to write a parser to read in the wrl, extract the coordinates and plot them in Flash :) Ofcourse writing a Strange Attractor generator in Flash is more fun, but hey :) What program did you use to generate the wrl?


Yes, I'm using the Cortona viewer...I have no clue why it wont work for you...Mayby some sort of formatting issue? I got it to work on Cortona and figure it would work in general. Could there be an issue with the formatting of the pointlist and your browser? I seem to recall, that not all browsers will render a point list the same way, but the details escape me...


The file IS simply a huge point list. How (and why) I generated the point list is a longer story...but since you asked :-)

I am an architect. I discovered the world of strange attractors through Krawczyk http://home.netcom.com/~bitart/ (I think we briefly discussed his work on Bit-101 some time in the past?) and was immediately taken by the beauty and complexity of the suggested surfaces. But everything I could find was 2D finished plots...no one seemed to have explored or published 3D models of strange attractors that you could rotate and study in real time...

So - I decided I needed to find a way to do that.

I wrote a small LISP application that I could use to generate strange attractors as 3D models in AutoCAD. It gave me the advantage of being able to use the AutoCad tools for viewing 3D models in real time and I found that if I used a very simple graphic (a point), I could plot about 5000 iterations of an attractor (or more) and rotate the resulting 3D model in real time. I could even apply perspective to the model view and make it slowly spin around....I could fly through it...around it...It was beautyfull!

But I still needed to find a way to make these models available for online viewing (as best I know, the latest dwf viewer from AutoDesk support 3D orbit for mechanical models only, not generic AutoCad dwg files). So I re-wrote my LISP program to write the point list to a text file, added a bit of formatting to make it a vrml file and - voila! I had a model of the strange attractor I could post online with a reasonable filesize and let others enjoy and explore the world of a 3D strange attractor in real time.

That was a long story...

You are right - it would be manageable to extract the point list from the wrl file (a simple edit in notepad) and use that in Flash...but my goal was not only to render a 3D model, but to let the user spin it around and view it from different angles. VRML renders the pointlist very efficiently. I've done some models with 20000 points and only had a slight performance issue...



Yeah, performance-wise Flash isn't good enough, I agree...

BTW, if you want dozens of formula's to generate 2D & 3D (even 4D) Strange Attractors (with heaps of sets of coefficients for each that lead to an SA), check out the excellent (also downloadable) book Strange Attractors: Creating Patterns in Chaos by Julien C. Sprott (http://sprott.physics.wisc.edu/sa.htm), if you haven't already... Robert's page is indeed wonderful (I used his spirolaterals-page as inspiration to make my own in Flash), but this book is the bomb when it comes to Strange Attractors :) I used it to build my Strange Attractor generator in Flash. No flying thru an SA tho :)

BTW, with 'parsing' yer wrl-file I don't mean re-writing the file in Notepad so it's 'optimized' for Flash, I mean reading it into Flash as-is and using it :) From what I can tell there's only (for Flash) redundant information at the front and the back and the points are listed inbetween... Doesn't matter :)

Have you considered using Java/Processing? Since more people have a JVM than a VRML-plugin installed?

Which SA-formula's did you use?

Sorry I'm asking questions, just exciting more people are exploring these wonderful things :)


No need for apologies! Questions are cool!

I got Sprott's book as pdf, but thanks for the link - it certainly is a good, if not the best, resource when it comes to exploring strange attractors.

For the SA posted, I used one of the formula's from Poul Bourke's site http://astronomy.swin.edu.au/~pbourke/fractals If I remember correctly it was the Peter DeJong attractor:

xn+1 = sin(a*yn) - cos(b*xn)
yn+1 = sin(c*xn) - cos(d*yn)

I then added the third dimension on my own (man - I was SO thrilled when this worked...)

zn+1 = sin(a*(xn*yn))

Basically, that just "boost" the atractor...nothing major about that. By using a sin vawe I got a nice 3D model. No specific thoughts on this other than it looked good...

I have considered using proce55ing for this and you are probably right that jvm is a better choice then vrml (sorry, Kristin). I'll have to explore java one of these days...in my infinite sparetime... :-)

I did recently come across a very nice app to explore 3D strange attractors though - You can download it here: http://www.btinternet.com/~ndesprez/index.htm

Of course, building your own SA rendering engine is where the fun is at, not hitting F4 repeatedly to get yet another attractor rendered. Still, it is a neat app - Got a lot of Sprott's formulas in there too. Check it out, if you have the time.

Spirolaterals are on my long list of topics to explore...I'll get there eventually...I have seen the ones you have posted and they are very nice!



haha...I agree that jvm may be better for rendering than a vrml browser. I just like vrml from the authoring perspective. :-)

Ah yes, Chaoscope :) Unfortunately my videocard isn't good enough to run it in any decent form, but I've seen that page a few times while 'researching' SA and such ('researching' in this case means more like finding all these amazing websites and wiping drool off my keyboard, haha)... Paul Bourke's page is indeed another endless source of drool for me :)

That's quite cool that you added a 3rd dimension to that formula, congrats :) And 'No specific thoughts on this other than it looked good...' sounds like a good enough reason to me :)

BTW, I wasn't trying to say VRML was/is a bad choice... Hell, I tinker around in Alambik every now and again and compared to that VRML is ubiquitous :) Too many exciting technologies!