Athlon64 3000+, 1GB Dual Channel RAM, Radeon 9800 Pro 128MB.
Cool, let us know where to get the latest when it’s ready cause I’d love to redo the benchmarks. Currently, just pulling up the Java3d, Xith and jME versions I have on at 800x600 my work machine (P4 3.2 with a 9800 XT card) gets me 343, 345 and 165 FPS respectively after a 30 sec burn-in at the initial spawn point.
I found the problem with the textures ;D. It was a problem with the parameters, but with the filename parameter ;). The textures are expected to be in a certain folder structure, which was not the case. So I searched the filenames in the bsp file and recreated the folder structure. e voila… the level looks just great, even if there’re still some textures missing. the biggest missing texture is the sky. Maybe someone has this texture or another matching one, then he could commit it to CVS please.
I proceeded to code the org.xith3d.util.EgoInputAdapter and now I find it could be useful for many people writing FPS shooters. Please have a look at it and tell me, if there’s anything missing to have a common tool for this purpose.
Enjoy the demo 
Marvin
Checkout your CVS. It’s just become ready ;).
Marvin
Is that right? Xith out performs Java3D? And JME has less than half the framerate?
Kev
Sigh, my brain is on ice in this freezing cold office… it’s jme: 345 java3d: 343 xith: 165… Java3d does better than jME though if you go to the top of the scene and look down at everything at once.
I’ll have a look tomorrow at CVS when I thaw out… 
Memory footprint, CPU usage and FPS are what I look at for the testing.
Just to check, the version of the BSP viewer that Qudus checked in is based from David Yazel’s original stuff right? It’s not based off the Tom’s benchmark code which is almost certainly what Renanse is running?
Probably not a fair comparison if so.
Kev
Yes, it’s based on David’s code. Is the Tom’s benchmark code faster (I mean after the level has loaded)?
Marvin
About the missing textures. I tried another map and basically there were the same textures missing. It seems like we need some kind of “Quake 3 map kit”, where all the base textures are located in. Or if I remember right, they were in the base folder in a Quake 3 installation. Are they freely available somewhere?
Marvin
Looking at the numbers you were quoting compared to Renanse’s I’m not convinced which version would be faster. It would seem Tom’s was quicker.
However, I think it’d be a clearer comparison of scenegraph core performance if the code bases were essentially pretty similar. The webstarts and related resources were referenced in this thread:
http://www.java-gaming.org/forums/index.php?topic=11748.60
Kev
I’ve had a look at the source of Tom’s one. The sources are dammn similar. There won’t be a difference in the performance because of the loader code.
Marvin
It would be very nice if the test could be redone when I’ve optimized the RenderBin thing. The two test results would be of a very high interest.
Be even nicer to see ongoing performance tests as you do the optimizations. Would help us all understand which ones were the most useful.
Kev
Well, so far the only optimization visible in performance is the one concerning the display lists. This has done a 22% boost in MultiCubeBenchmark and 16% in BSPLoaderTest. The other “optimizations” were just to make the code better understandable and cleaner.
The next big boost is to expected from the RenderBin optimizations.
Marvin
I pulled down xith from cvs and paired it up with the code I have here… I had to massage a lot of the texture stuff to get it to work, and the end result is that the performance is actually slightly worse than before. Maybe you could edit the xith version of the joint test code and post that somewhere?
sometimes I’m inclined to believe you’re a lier paid by NCSoft to diffuse false infos about jME’s concurrent engines…
but I do know it’s not true at all… so please don’t flame again, FPS is FPS, OK ok I know.
Wow, I assume thats the language barrier coming in there because otherwise thats fairly insulting. Also, there doesn’t appear to be a flame anywhere?
Kev
What do you mean? The cluster-visibility test? This is the same as in David’s original and in the version I found in the original benchmark thread (posted above).
And I can’t believe the FPS are worse than before ,since we have display lists now and the rendering code is not less efficient than before.
Marvin
Oh yeah I know shouldn’t have posted that, it’s just it seems the FPS of Xith3D in the Q3 bench really can’t go up. Now it seems even that it has gone down since display lists implementation. Unbelievable.
@renanse : don’t pay attention I said it wasn’t true… I just feel a bit sad
Well there have been a lot of other changes going on with Xith aswell. Not to mention as Renanse pointed out, the code provided here isn’t the same code base as he tested before - because this is based directly off David’s initial code, rather than Tom’s benchmark (of SquareHeads - the optimised Quake 3 level renderer - fame).
Either way, Renanse is nice enough to bother running the tests - no need to get defensive
Kev