I’ve been working on a workflow to get models from production quality and resolution down to something that will run in Papervision. It’s been a tough process — Papervision can’t really handle more than 10K polys, and probably 3K is about all you can really use.
Take a look at the sample Mustang 2008 DAE (Collada) Model:
This is built of:
- mustang.dae (model data)
- car_uv.jpg (car uv map)
- tire_uv.jpg (tire uv map)
Taking production detailed models into Papervision isn’t really possible at this time — it’s just too limited. You should consider redoing the model as a simplified shape or using a polygon reduction tool to degrade the model (very hard to accomplish as polygon reduction will often destroy the model during the process). So, sometimes it’s best to rebuild the model trying to save some of the form of the model and putting it into a simple enough package that can be used. Rely on the texturing to provide the necessary suspension of disbelief required to ‘sell’ the concept.
In the case of the above model, I rebuilt the model from scratch and kept the polygon count as low as I could.
I haven’t seen a lot of information regarding best formats etc so I thought I’d mention how I’m currently doing it. I’ve given up on ASE format for now — I can’t seem to find a working workflow; lets focus on Collada. There’s no Collada exporter for Cinema4D (my personal favourite 3D app for now) and I’ve been trying all different ways of getting a file into Blender, and have settled on 3DStudio.3ds format (which transfers the UV and texture information into Blender effectively from C4D). Blender can export to Collada 1.4.x format, which is required for Papervision, so that’s what we’ll be using Blender for (all referenced product links are at the end).
Break out the parts that require animation, and simplify all models to the most basic structures possible.
Using Bodypaint, wrap texture onto uv’s using your favourite mapping projection technique.
Export as 3ds, and import into Blender.
Export as Collada 1.4 and then you can import into papervision previewer swf to test the model.
Here’s the settings I used from Blender:
Referenced tools:
Pingback: dabu's blog » Blog Archive » papervision3d and cinema4d - tutorial
I found the Collada file from Blender wasn’t quite ready for use in Papervision3D without some manual text editing. I wrote a tutorial at http://www.alteredegg.com/cms/index.php?option=com_content&task=view&id=42&Itemid=1
that details how I handled it.
Alteredegg,
You’re right that the collada file references to image maps need to be relative paths instead of absolute as Blender generates them. I remember doing that in my tests but forgot to mention it here.
There’s additional steps (like deleting that silly default cube in Blender that it adds by default to a new scene) that I also forgot to add in here. I’ll update the post with more detailed steps.
Thanks
You can now import 3ds directly into papervision 3d 2.0 !!!
Check the papervision documentation for details..:-)