odf posted at 1:37 AM Mon, 17 October 2022 - #4446840

Very cool! Primorgetonia is taking shape.

Can I have that unsuccessful injection file, if you still have it? I am somewhat reminded of trying to inject weight maps. The maps inject fine, but apparently the necessary joint transformation channels can't be created by an injection pose, so either they have to be in the CR2 or the figure has to be manually prepared for injection. I'd like to see if there might be something similar happening for animated origin injections.

Incidentally, I've had some problems loading materials that use animated material node parameters. The trick there was to load the material twice. Could be a similar issue. Anyway, I'd like to play around a bit and see if this can't be solved. If you don't have the injection file, anymore, no problem. I'll just cook up something myself.

I haven't tried creating the injection for this particular successful iteration. The injection I created and tested was for only the eyes scaling up. In that instance as soon as I loaded the morph the eye centers would automatically pop into the scaled up position and be stuck there. In this version the centers are animated to scale up and down in the same dependency. I'm hoping that this causes the centers to behave as they should in the animated origins portion of the inject. When I get home from work today I will test.

Yes, I can give you any unsuccessful injection files. I'll zip it up and link you via PM later this evening.

Your last suggestion about deleting the eyeball scaling and leaving the head deformations and using a scale to match dial is a very viable alternative. Scaling dials pass well to injection. Matching the exact scales could be a bit fiddly, and because I created the scalings in several steps it's not possible to record the exact scales I used from Blender. Also because I foolishly expected everything to work in injection without any set backs. It's worth an attempt though. In any case it all works as it should embedded in the cr2 at the moment, which for me is a very good step forward. The working version I have now will need a slight blink difference morph adjustment baked or attached as a sub dependency (more than likely at this point) into the head deformations to fix a small discrepancy at full blink.

I'll update you this evening.

I'm not sure why ADP is so excited, in it's current state it isn't injection passable. Only through a live cr2.