OMG Blender 2.5 is out!

…with a wildly different structure for plugins.
Has anybody finagled Chicken to work with Blender 2.5 yet? I’m sure it’s just a matter of time. I want to start learning the new 2.5 interface as soon as I can without having to worry about backwards-compatibility to 2.4.9 to export. Or did the file format not change that much?

first of all… blender 2.5 is NOT out yet. what is released is alpha 0.
it is NOT recommended for use in’s just an early testing version. daily builds for 2.5 where available for month now. a really the stable version is to be expected around mid of 2010. chicken is beeing worked on to work with 2.5. looks like the python api changed drastically so it requires some work.
in the end, it’s a matter of time. for both, the real 2.5 release, and chicken for blender 2.5. more informatino might be available in the chicken-thread

Chicken will be reworked for 2.5, but its basically going to have to be a rewrite - the differences are that big. I won’t be working on that till Christmas at the earliest though, and certainly not with the current version, as the Python API is not yet locked down.

Ouch. I was afraid of that. Maybe the plugin API changes gives the same reaction that the UI changes do:
“This looks like it makes way more sense than it did, but I’ve no idea how to use it.”

it’s been quiet a year now, blender2.5 has reached beta status.
any new news from chicken exporter?

The plan for Panda is to depreciate egg in favour of Collada, which is a standard file format that Blender already has support for. Only problem is that, at this time, both the Blender Collada exporter and the Panda Collada importer are horribly limited, making that pipeline mostly unworkable, but nobody want to port Chicken knowing it will be depreciated at some point in the near future. Basically, right now we have a rather unfortunate gap in service, which is really not good as everybody is moving to 2.5x (Who wouldn’t?).

There is little to be done however other than to wait until both side of the pipeline are working - for Panda that is a lot of work (The current dae2egg will need to be replaced by a system that takes a dae directly to Pandas internal format.), for Blender its more of a few missing features (That and a recent Blender update means I can no longer compile Blender with Collada, and can’t work out why - I had been submitting bug reports to get those last few omissions fixed.).

On one hand its different API of Blender 2.5 and Python 3, which is a new language, on the other is Disneys idea to switch to Collada.
For Blender, I dont know if all these changes were worth it, I’m not impressed by the new UI or the new movie which is meant to showcase its features.

For Panda, I highly doubth the Collada exporter will have all the features of egg format, unless we imrove it ourselves and contribute back.
Like curves, stencil textures, collision solids, LODs and stuff like that.
That will take some time. I dont think Panda’s collada importer supports all that right now or anytime soon.

Disney has an idea to switch to Collada? I didn’t know that.

The whole point of switching to Collada is the fact that it supports a lot more things than egg (cameras, lights, shaders, to name some very basic things)

Change it to “community”.

Like I said, “unless we imrove it ourselves”. I know that its a powerful format. But I highly doubth Blender’s exporter has all that features.
We already talked about this with lethe.

Blender exporters main omission at this time is tangents/bi-normals, which prevent normal maps from working. It also doesn’t export arbitrary properties of any kind, which are of course very useful for game assets. Beyond that there are various issues, but all the normal stuff is there - meshes, uvs, materials, animation etc. It also has camera and light support. Current status is pretty irrelevant however - we are talking about the future here, and there is nothing stopping us submitting patches to Blender.

Switching to Collada is an obvious decision to make - it is the more capable (If somewhat harder to handle.) file format and this is not just about Blender - almost every 3D modelling program on the planet has Collada support in some form or another these days, and most of the competing game engines out there also support it, so supporting it means all those esoteric programs can work with Panda. There is no good argument to not do it, other than the problems of transition, and in the long run those are insignificant.

I don’t see where we disagree.

If EGG files were to be replaced, would we still have BAM files (Collada->BAM instead of EGG->BAM)?
Also I would be sad to not have all of the great egg*.exe tools.
To be honest I’ve grown quite fond of EGG format and expecially the EGG/BAM workflow.
Has this already been discussed to death somewhere and I missed it?

Yeah. I think. Unless there are binary collada files.

Well they will be pointless since your modeller most likely supports Collada in some degree.
Though there could always be collada converters instead.

Yeah, this thread, page 17, starting from my first post: … &start=480

Don’t worry its gonna be great when its done. But that might be next year and thats the reason we wont see newer versions of Chicken. Only rewarding in the end.

I thought Disney had this idea, because the official versions are going to remove egg support as I heard, I know Disney uses its own version of panda, but they still contribute stuff. One day they might add something we cant use with our version because we dont use egg files anymore. Thats all I thought.

Actually I was referring to egg-palettize, egg-optchar, etc.
It sounds like this is just more of an idea rather than something that has been decided and being worked on, am I correct?

Most likely a rewrite for collada format.

It has been decided and is being worked on. I dont know if it was obvious from that topic, but lethe, rdb and drwr have mentioned this several times in different threads.

I was frustrated by this from the beginning, before lethe explained everything. If you feel the same then tell me why you think like that and I’ll try to explain it to you.

I don’t want to steer off the subject, but could you perhaps summarise your objections against switching to COLLADA in Panda3D 2.0? It’s supposed to only have benefits.

rdb, you seam to misunderstand me in these threads… I’m not against it, quite the opposite. I said I was against it before lethe explained all to me (few months ago).
So I have no idea how you came to that conclusion.

OK, then I misunderstood you, or was mixing up several posts of yours. I apologise.


Moving to Collada seems to make sense but there may be a problem: the modelers. Having tried Wings, Blender,
Sketchup, and the Irrlicht tool all produce different results and none seem to have export animation.

The best exports seem to be Wavefront obj and the .X file.