Panda3D for Cave


I’ll be working really soon on a artistic VR project for a CAVE environment.
It consist of irix sgi machines, stereo projections for 3 walls and a floor, a pair of goggles and a wand.
Could Panda3D be used for creating this project? Or should it work together with something like freevr or cavelib?
I’m quite new to all this, so I’m sorry if my question is a bit clumsy…
Any help, advise and/or pointers would be really appreciated!


Panda should meet your needs for this.

This is exactly the sort of application we had in mind when we originally started developing Panda3D–the VR Studio has just finished several CAVE projects driven by SGI Irix hardware, and though we could see that the PC was going to be our direction of the immediate future, we imagined we would still do the occasional Irix project.

That wasn’t the way things happened, though, and ironically, you might be the first person to use Panda3D on Irix to drive a CAVE.

As the trailblazer, you might find the work is more involved than you want it to be; although I can assure you that everything should work well, I can’t promise you that it will. But we’ll be happy to be of help in getting everything going.


David, thanks for your answer!

Within our team we interpret it in different ways though…
Some read “yes! it works, but you might spend a week or so of configuring and hacking.”
Others read “no, it’s not supported. nobody ever did it. you have to do all of the window handling yourself. The effort is too risky given the timeframe”

Could you give us any indication of how much time it would take?

The project I was talking about is supposed to be presented in March, with one fulltime programmer, one artist, and some techies who can help a couple of hours a week. Maybe Panda3D is too risky for this particular project?

We’re all very excited about Panda3D, but we have to be practical… Maybe it would be wiser to use something else for this project, but reserve some time for playing with Panda, so we can use it in future projects…?

Hmm, I don’t want to be guilty of overselling Panda, but I think the honest answer is closer to the first one. It works, practically out of the box, but probably there will be small issues that will come up and will be easily remedied.

We did develop Panda on Irix first, where our expertise lay, before we ported it to any other platform. It still remembers its roots. The window handling for X and other nitty-gritty has been in there for years and it works fine (this part of the code also works on Linux, so it does get actively exercised). We do still build Panda on Irix at least every few months, so I am confident it will compile, although there might be some minor issues if you happen to have a different version of the compiler than we do.

Still, it is a pretty tight timeframe, and there’s risk involved, especially with a new graphics engine that none of your team is already familiar with. But there’s also risk with trying to develop two different solutions at the same time.

It comes down to how nervous you are about slipping and what your other options are. If you’re considering another engine that you’re comfortable with and already know will work, that might be the better choice. But if it’s between this unknown devil or some other unknown devil, I’d suggest that Panda is likely to be an excellent choice.