Building realistic graphics

I am just getting started with JOGL.  I started doing some Java 3D programming about a year ago, but then had to stop the projects I was working on.  Now that I'm picking up 3D programming again I thought I'd use JOGL.

When I was working with Java 3D I was only working with very basic 3D shapes to start with.  For even these shapes defining the geometry was very tedious and error prone.  I will do the same thing with JOGL, that is start out with basic shapes and define the geometry manually but I was wondering if there are any tools that help you to create more realistic, lifelike looking graphics.

I'll be attempting to do geologic simulation graphics and would like the animations to look as "lifelike" as possible; i.e. real rock textures, etc.  If defining basic shapes is tedious and error prone, certainly defining terraine and lifelike rock tectures will also be error prone.  Any help here would be greatly appreciated.

Also, I may be generating a great deal of data from mathematical models of geologic processes and then wanting to use that data to simulate the process graphically.  Any suggestins on this?

I know both of these questions are very advanced for the part of the learning curve I'm currently on with JOGL, but I'm looking ahead to the real exciting stuff!!!

Thanks,
Mike

You can find some nice rock textures on Internet, even on opengameart.org.

I worked in a subsidiary of the French institute of the petroleum, maybe I can help you a bit. It's weird for me as in general we don't try to use textures on mathematically generated models even though it isn't impossible.

There are several geological frameworks supporting JOGL through Java3D or Ardor3D. Maybe you can use GeoCraft, you might need some help just to update its source code in order to use the very latest version of Ardor3D and use it with my JOGL renderer.

Thank you for your prompt reply.  I'm sure I'm going to need all the help I can get!

This is a new programming endeavor for me.  I have over 30 years of software development experience in commercial applications.  I've been programming with Java since the mid to late 1990, although I have not been doing much programming on a day to day basis lately.  This is a good opportunity for me to get back to that.

As I get started working with JOGL I'm sure I'll have many questions.  I'll post them here, in that way you can contribute.

Let me explain in a little more detail what I am doing.  My undergraduate degree is in geology and I have a masters in software engineering.  I have always thought of ways to try and combine my experience in these two fields.  I have been teaching software engineering as an adjunct for the past 5 or 6 years and just this year started teaching undergraduate introductory geology too.  I want to get involved in some geologic research where I can use my software engineering.  I plan on studying some Triassic alluvial fan deposits here in New Jersey (USA) where I live and work.  I would like to use the data I collect on these deposits as input to alluvial fan mathematical models to hopefully recreate the original environment of their deposition.  Once the model have been run with this input I would like to program an animated simulation of the formation of these very interesting deposits and hopefully publish my results.

I have a steep learning curve to climb at this point and right now I am at the bottom of the hill!

Hope to be talking to you more on this forum in the future,
Mike

Sounds like a fun project!

I think so, just hope I'm not being overly ambitious!  I'll only know that as I get deeper into it and find out what I really can and can't do.

Just don't do everything from scratch, it's not doable. I worked on a low level scenegraph during 18 months and it was time consuming. Don't imagine that you'll be able to write everything by yourself.

Thanks for the advice.  I'm sure I'm going to have many more questions as I progress.  Stay tuned!

I already took a look at GeoCraft today.  Although it's "advertised" as a framework, a tutorial I looked at was showing how to build models in the "Eclipse" environment.  I was hoping to find a tutorial on the API.  I'll keep looking.  I did download a ZIP file and was poking around in it.  I'm also going to take a closer look at this.

I also spent some time looking at one of the JOGL 2 tutorials on the JOGAMP site.  I'm going to start experimenting with some very basic code of my own tonight.  Nothing to do with my project, just basic "climbing the learning curve" stuff.

In the worst case, you can use Geocraft's importers and its code that builds the meshes if you don't succeed in using it as is.

Thanks, I'll keep that in mind for the future.

BTW, I did find what appears to be a good open source Java GIS library.  I downloaded it and will also be looking at how I might be able to incorporate some of its features in the future.

Julien,

Thanks for mentioning meshes.  This is new terminology for me so I did some "googling".

Learning the very basics of what meshes are (i.e. from JOGL and OpenGL perspectives) gave me an idea, something to look at again for way down the road.  Here is what I'm thinking.  Once I really understand all the equations that model alluvial fan deposits, I will create a component that generates the data for an alluvial fan given certain input.  Part of that component (or maybe a separate component) will convert the "raw" data generated by the equations into mesh verticies that can then be used as input to the graphics part of the simulation.  This is where I was thinking of applying the "rock textures" as I mentioned earlier.  Maybe I won't need them, time will hopefully tell.

What do you think?  Does this sound feasible?

Thanks for mentioning meshes.  It gave me some direction to take once I climb the learning curve.

Mike

Mike wrote

deposits, I will create a component that generates the data for an alluvial fan given certain input.  Part of that component (or maybe a separate component) will convert the "raw" data generated by the equations into mesh verticies that can then be used as input to the graphics part of the simulation.  This is where I was thinking of applying the "rock textures" as I mentioned earlier.  Maybe I won't need them, time will hopefully tell.

What do you think?  Does this sound feasible?

Yes that is in very high overview what many game systems use to generate terrain, although games usually don't simulate real world events but produce a data set that "looks good enough" :-)

You should check out this site if you haven't already: http://vterrain.org
Maybe not a tutorial but more of an inspirational/link site.

Thanks, I'll keep this in mind also.

I'm so glad I posted here; getting lots of good ideas and advice!

Also the positive reinforcement is great, but I'm looking for negative criticism as well.  The negative can be as, or even more, helpful since it can save one from wasting time on something not worth the effort, or just not doable!