Easier to import different file-formats

[gmfreaky]

Would it be possible to make it easier to import file formats like .3ds, .obj, .an8, .blend etc.? You have to convert it now, and it takes some time. It would be faster if you could just save and immediately see changes.

[calsmurf2904]

Don't think this will happen, they have chosen for the .mesh format because it has been optimized for use with the Ogre engine. Other file formats probably mean larger loading times etc.
You can write your own importer though. (An at runtime .obj/.3ds file -> .mesh converter. Especially obj isn't really hard since its fully ascii (nothing binary at all.)).

[_tommo_]

I agree with calsmurf, and this has been debated very often...

But, IMHO the "easier to import" issue is a very important one, and it is the weak spot in Ogre: it doesn't have an official content pipeline... especially today where the content expense for AAA games has vastly outrun the tech one.
My idea would be having the Ogre Team managing directly a toolset that comes with the Ogre SDK (mesh viewer, particle editor, material editor, importer for preferred editors) could greatly enhance productivity with Ogre...
and would make it much more artist-friendly: i had found often difficult to explain what to do with things to artists with no background in programming.

Anyway i can offer only my noob-ish opinion on the subject, and I know that the Ogre Team is always busy... so take this as vague idea

[reptor]

I see it as unrealistic to expect the Ogre3D project to handle a lot of 3D file formats.

Perhaps Sinbad could create a commercial tool set together with other people. I mean it would be paid work for everyone involved. That is of course his decision entirely. I am just pondering how the situation could be improved. Perhaps it could make sense for Sinbad if it was a commercial project. If it isn't then I have a hard time understanding what motivation there would be for him to do it.

[_tommo_]

Well, there are already a lot of good free tools, this hasn't to be started from scratch, and more importantly, it HAS to be free and multiplatform like the main engine as much as possible...
otherwise it would be just useless as a cohesive factor, the situation would just remain as it is now with another tool added.

One simple thing to do in the near term would be to choose the best existing free tools and call them "official", asking involvement from their authors.
This could look like a word play, but IMHO this would enforce on that tools many important features that they can't have when community-maintained:
-updated; they all work with the most recent Ogre (that is not always true)
-multiplatform;
-easily obtainable;
-uniformly documented;
-uniform interface;
-provide reviewed example source for custom editors;
-ship with the SDK for first time disoriented users
-integrated where possible
-continued development: the main line tools are never discontinued just because the original dev runs away.
And these are just some of the advantages i can see.

For example i always felt as a real lack an integrated Mesh + Material Viewer, there is a ton of existing solutions but neither really updated or complete (IMHO).

[sinbad]

Personally I think end-user tools only really work commercially. Open source is generally not great at maintaining user-friendly tools, and just from a time point of view, everyone contributing to Ogre is working for nothing, and therefore works on the things they want to. Unless someone has a need to develop such a tool, and is also willing to open source it (which basically means giving it away), it's not going to happen. I'd like it to occur, but my experience is that making end-user tools in open source is something of a once-in-a-blue-moon thing. Usually the large funded projects do it best, like Mozilla and Eclipse, because they have money to employ people to do it. Good tools take lots of time, and usually a central vision.

If you want to import other formats, just use a modeller. Blender, 3DS etc all import lots of formats, and you probably want to tweak things anyway (scale etc). I've never found it that useful to directly import random files from the internet into a project (unless that project is also an editor which can clean / alter / re-export them) because there's always something wrong with them. It might work for tests, but for serious projects you're going to have artists on staff anyway using your chosen pipeline, you're not going to be pulling together random stuff from different sources directly. Again, the only use for importing these random files is if you're already making an editor, which is therefore a circular argument.

And as for pipelines, check out OgreMax (3DS, Maya, XSI), and our Blender exporter. They work fine. Particle Universe will meet your particle editing needs too.

I'd love to say I'll write you a complete world editor (although I personally think there's no such thing as a universal world editor) and give it away with Ogre, but frankly I can't afford it! I'm pretty much tapped out just from working on Ogre.

[Praetor]

The other issue is Ogre is just too low level. It wears many hats in that it can be used as the rendering backend for a lot of things. File formats and editors are difficult to make universal when your universe is as large and Ogre's. What you'll end up doing is creating an editor specific to your problem, which won't work so well for someone else's problem.