MD5Importer vs MD5Reader 2 vs Collada

Hi, i need an advice

I'm pretty sure MD5Importer is as good as that bag of tricks gets.  While I've seen argument for Collada as well, there's no support for animation blending in JME at all, so far as I know.  MD5Importer at least has support for transitioning from one animation to another.

Without support for animation blending, you're going to be stuck with some awkward hacks, and won't be able to get the most out of any modern animation format.  I'd love to be proved wrong here, btw :D.

Collada importer is not complete and does not work in most cases (I could'nt make it work and in many threads, i have read ppl have not gotten it to work). MD5Importer works very well, but I am having trouble setting the orientation of rootbone properly in 3DsMax, so the animation would have the orientation I need in JME.

md5importer is fresher code and is actively developed. Md5reader is almost as old as doom3 itself and was still experimental when chaos left :frowning: , but had good ideas if finished it would have been great. that having been said, it still acheives marginally better framerates than md5importer(on my system) and still seems to deal with multple instances better, tried for a while to understand it but the lack of docs and my lack of skill made that dificult, I use it because  I am more comfortable with its layout and functions, but it should be considered dying or dead

I'm agree with Mcbeth because I'm comfortable with its layout and functions too,  and i think that MD5Reader 2 (Not the refactor) was working fine until i have to migrate to JME2.0, i tested it many times and the result is the same the animations loads just fine supplying all my needs, MD5Importer  is not so "tricky" like MD5Reader 2 in some way more comprehensible making easy the maintenance (I hope), so i will try the migration to MD5Importer to see what happens. Thanks to all for your advices.