I'm curious if anyone has taken a crack at porting PolyVox to Actionscript 3. Since the Flash 11 Incubator has started, you can make use of full hardware acceleration within Flash now. From what I have read, its performance is rather good.

I am considering taking a look into the development of such a library myself, however, I figured I'd check and see if anyone has looked into porting PolyVox.

Are you talking about expossing the library so it can be called from Flash (can Flash call C++ libraries?) or actually porting the code to ActionScript?

There is no actual port as far as I know. But there are some basic SWIG bindings which have been used to access PolyVox from C#, for example. I don't know if SWIG has ActionScript support though - I doubt it...

A port would be required for it to run in Flash. There are two ways that it could be ported. It could be converted from C++ to AS3 using Adobe Alchemy or it could be ported from scratch into AS3.

I'm not sure if the benefits of converting are that great mostly due to Alchemy being a seeming abandoned project by Adobe.

I don't know a ton about C++ so I'm not sure how well I could port it myself. I was thinking about taking a crack at implementing an voxel cube extractor from scratch but figured I'd see if anyone else is interested in a Flash version of PolyVox and perhaps we could try take a crack at porting it together.

To be honest I suspect a direct port would quite difficult. PolyVox's heavy use of template will probably get in your way here, and you'll also be forever playing catchup as the library evolves.

You might find you are better off just trying to understand the algorithms and then implementing them from scratch. The CubicSurfaceExtractorWithNormals is the simplest extractor, and for testing purposes the Volume class with compression/paging/etc could be replaced by a simple 3D array.

Yeah, most likely that is what I will end up doing if noone else seems interested. Like I said, I figured I'd gauge interest and see if anyone else was doing something like this.

The focus for the future seems to be browser games and apps. Having an engine akin to PolyVox for Flash seems like a worthwhile project.

Yeah, definitely would be worth playing with.