Collada/Mixamo Importing Bug. #39099

New Issue

Harry Boltz III · 2014-03-10T22:03:18+01:00

Harry Boltz III commented

2014-03-10 22:03:18 +01:00

System Information
Win7, Mac OSX, ATI Radeon HD 5700 Series

Blender Version
Broken: 2.65, 2.70
Worked: 2.49

Short description of error

Users have experienced a problem with importing certain Mixamo characters into blender while using the Collada format. Its been noted that these characters are usually made of multiple meshes, and these submeshes have multiple joints deforming the mesh. This means that submeshes with a singular joint skinning information (swords, shields, eyes) remain unaffected. These problems are also apparent while trying to bring in a file with animation as opposed to one with just a skeletal hierarchy.

Attached is an example .dae(demo_error_1.dae) with the problem shown. The walk cycle was created via Maya using the Mixamo workflow, and it successfully imports back into Maya without any problem. However, upon trying to bring the file into Blender, blender receives an error and will not bring neither the mesh nor the skeleton into the application.

Having talked with /Nathan, he mentioned that creating a bug log would be helpful. demo_error_1 is the scene as is straight from export. This scene created demo_error_1.txt log, which made the error sound like a text/material problem.

With this knowledge, I modified the mesh of demo_error_1's character to share one brand new diffuse texture, and I saved the new collada as demo_error_2.txt. Importing this new file into blender gave the log demo_error_2.txt.

Exact steps for others to reproduce the error

(I received the log.txt's by opening Blender via cmd and exporting log file to my c:\tmp folder. Not a necessary{F80576} step to get the bugs, but I guess it is to see the exact logs.)

Open Blender 2.7
Import either Mesh.

Files
ColladaBug.zip

**System Information** Win7, Mac OSX, ATI Radeon HD 5700 Series **Blender Version** Broken: 2.65, 2.70 Worked: 2.49 **Short description of error** Users have experienced a problem with importing certain Mixamo characters into blender while using the Collada format. Its been noted that these characters are usually made of multiple meshes, and these submeshes have multiple joints deforming the mesh. This means that submeshes with a singular joint skinning information (swords, shields, eyes) remain unaffected. These problems are also apparent while trying to bring in a file with animation as opposed to one with just a skeletal hierarchy. Attached is an example .dae(demo_error_1.dae) with the problem shown. The walk cycle was created via Maya using the Mixamo workflow, and it successfully imports back into Maya without any problem. However, upon trying to bring the file into Blender, blender receives an error and will not bring neither the mesh nor the skeleton into the application. Having talked with /Nathan, he mentioned that creating a bug log would be helpful. demo_error_1 is the scene as is straight from export. This scene created demo_error_1.txt log, which made the error sound like a text/material problem. With this knowledge, I modified the mesh of demo_error_1's character to share one brand new diffuse texture, and I saved the new collada as demo_error_2.txt. Importing this new file into blender gave the log demo_error_2.txt. **Exact steps for others to reproduce the error** (I received the log.txt's by opening Blender via cmd and exporting log file to my c:\\tmp folder. Not a necessary{[F80576](https://archive.blender.org/developer/F80576/ColladaBug.zip)} step to get the bugs, but I guess it is to see the exact logs.) 1. Open Blender 2.7 2. Import either Mesh. **Files** [ColladaBug.zip](https://archive.blender.org/developer/F80577/ColladaBug.zip)

Harry Boltz III commented

2014-03-10 22:03:18 +01:00

Changed status to: 'Open'

Harry Boltz III commented

2014-03-10 22:03:18 +01:00

Added subscriber: @heboltz3

Harry Boltz III commented

2014-03-10 22:12:15 +01:00

Removed subscriber: @heboltz3

Gaia Clary was assigned by Bastien Montagne

2014-03-10 23:40:27 +01:00

Bastien Montagne commented

2014-03-10 23:40:27 +01:00

Added subscriber: @mont29

Bastien Montagne commented

2014-03-10 23:40:27 +01:00

Gaia, care to check this one? Thanks! :)

Harry Boltz III commented

2014-04-08 02:25:08 +02:00

Added subscriber: @heboltz3

Harry Boltz III commented

2014-04-08 02:25:08 +02:00

I was wondering about the status of this bug? Since posting we have been flooded with reports of this breaking workflow. We are desperately seeking a fix.