Another user here Leaving aside the new collada exporter.. but talking more about exporting for game engines in general
I think that the collada spec may be ambiguous, but if blender developers and other game engine developers (ue4 and godot for example) work together, a more consistent spec can be made of it, maybe we could get to a open source official importer-exporter (like ogg opus, xiph made the open standard, the encoder and decoder are open source and maintained by them, so people are not interested in making their own encoder-decoder, but in getting the official one better). If collada is a hard beast to deal with, it may be a better idea to create a new more strict spec, maybe based in collada (or maybe not, if collada is a bad example to follow, i dont know), collaborating with game engine developers, and creating and maintaining an open importer-exporter that blender and game engines implement and contribute to... also, with a common official imp/exporter, the community would have something to test against when creating new importers/exporters (like in other languages) and it would be easier to achieve compatibility... Epic has announced his interest in making indie games easier to make with UE, and I feel they would be very receptive to talk about this and get the team to work in this with blender developers, and i'm pretty sure that many other game engine developers would be pleased to collaborate too. Sorry for my bad english if you see any mistakes ;) - BoteRock On Sun, Oct 12, 2014 at 5:25 PM, Sergey Sharybin <sergey....@gmail.com> wrote: > Juan, i don't expect a silver bullet and i don't even believe in this. But > 3 days ago you claimed "This plugin is very battle-tested, and files > written by it can be read by most DCCs and engine", now you say "You can't > write an exporter thinking that you will please every importer out there". > This two things i can't really link together. So the questions so far: > > - Is current blender's importer capable of importing the exported by the > addon scene? > - Which engines/software are supported with the current state of the addon? > - What are the plans about extending supported 3d party apps? > - What are the plans about importer? > - Who's gonna to maintain the code? > - Does someone from blender users tested the addon? > > But maybe the biggest one to be answered first: what are the benefits for > blender from the current state of the addon? > > Timur, speed is kind of separated topic. It exists for sure, but can it be > discussed as a separate topic perhaps? > > On Mon, Oct 13, 2014 at 2:27 AM, Timur Ariman <timur_ari...@yahoo.de> > wrote: > > > > > Hello Ton,Sergey,Juan, > > > > from a pure user standpoint ; > > > > one big hurdle regardless of file format(s) within Blender is that the > > python based import/exporter scripts are really slow, it takes in the > worst > > cases a couple minutes to do a I/O of a "huge" file. > > It would be good to optimize first the speed of loading and writing files > > when it comes to interchanging files between different 3d programs. > > for a start it would be great tom optimize obj.. > > this alone would tremendously help a lot of users and suercases while > the > > discussion over the game engine related fileformat(s) keeps ongoing. > > > > > > for game engine related stuff; > > > > the interest for > > the collada exporter is for UE4+Blender users at the moment is rather > > slim due to the fact that collada isnĀ“t inlcuded (yet?*) into Unreal... > > > > > > * Epic proposed to include the collada fileformat back a few month for > > Blender users if there is interest. > > > > it seems that there is a new maintainer - Juan available for an I/O > > Collada script in Blender available... > > give it a try and also inform Epic of a possible collada solution so the > > users can test it in their engine. > > > > > > > > maybe for "security" take the same route as the fbx fileformat got and > add > > it as an "experimental Addon " so interested Blender users can easily > > activate it and switch between the already existing nowadays unmaintained > > collada code inside Blender and the new code which is maintained by > Juan?! > > > > I think this is a first possible start for a solution because then the > > users can more easily start battle testing the new collada code. > > > > > > > > > > > > > > > > > > Juan Linietsky <redu...@gmail.com> schrieb am 21:48 Sonntag, 12.Oktober > > 2014: > > > > > > > > > > > > > > > > >Sergey: > > > > > > You seem to be somehow expecting a magic bullet. You can't write an > > >exporter thinking that you will please every importer out there, > specially > > >for a format as ambiguous as Collada. The best you can do is to export a > > >correct file and then work with those who have importers to make sure > they > > >are reading it properly. If users report an issue with an exported file > > not > > >working with a specific DCC, first you make sure that the file is being > > >exported correcty. If not, then the exporter is fixed (I take > > >responsibility for that), but if the file is OK, then you instruct the > > user > > >to talk with whoever is developing that software to properly support his > > >file, offering coopeation (I also take responsibility for that). This is > > >the only way something like this can work, and as far as I will get > > >involved. > > >_______________________________________________ > > >Bf-committers mailing list > > >Bf-committers@blender.org > > >http://lists.blender.org/mailman/listinfo/bf-committers > > > > > > > > > > > _______________________________________________ > > Bf-committers mailing list > > Bf-committers@blender.org > > http://lists.blender.org/mailman/listinfo/bf-committers > > > > > > -- > With best regards, Sergey Sharybin > _______________________________________________ > Bf-committers mailing list > Bf-committers@blender.org > http://lists.blender.org/mailman/listinfo/bf-committers > _______________________________________________ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers