11-14-2015, 09:21 PM
If you try to rip a linkset of mesh objects, attachment points nor transforms positions are preserved. A great example is if you try to rip any popular mesh body (try it with a demo), import it back into Blender, and you will see how you will end up with well almost a hundred "planes" of body and clothing objects, all mashed together as expected to their own bounding boxes.
Ripping something simpler like a single object clothing, still requires some guesswork in where it's supposed to be on the avatar, and how it should be scaled to size. The best one can do is eyeball it as best as possible in Blender, and compare the results to the original in-world one.
It would be nice if we can rip, and at least know exactly where it should go.
Ripping a mesh clothing that is made of too many linksets, might not even be worth it. There is much frustration in re-aligning each part with needed precision.
After using various viewers for years, I still can't figure out any way to work around this problem. It doesn't matter if I buy the item and it is in my Inventory (I prefer it this way, as opposed to ripping from other avatars in a busy sim). If only a viewer can dump a sl body shape along with a mesh, then at least we would have the avatar body exported in mesh along with the mesh clothing properly positioned on it, and we can just snap to that in Blender.
One solution (I have no tried yet) is to have your avatar in-world in the T-pose, and then rip the frame out of the GPU. A frame rip will litereally give you a WYSIWYG snapshot, because you can do that with a non-copybot viewer, as opposed to exporting from stored cached in a copybot viewer.
Ideas appreciated!
Ripping something simpler like a single object clothing, still requires some guesswork in where it's supposed to be on the avatar, and how it should be scaled to size. The best one can do is eyeball it as best as possible in Blender, and compare the results to the original in-world one.
It would be nice if we can rip, and at least know exactly where it should go.
Ripping a mesh clothing that is made of too many linksets, might not even be worth it. There is much frustration in re-aligning each part with needed precision.
After using various viewers for years, I still can't figure out any way to work around this problem. It doesn't matter if I buy the item and it is in my Inventory (I prefer it this way, as opposed to ripping from other avatars in a busy sim). If only a viewer can dump a sl body shape along with a mesh, then at least we would have the avatar body exported in mesh along with the mesh clothing properly positioned on it, and we can just snap to that in Blender.
One solution (I have no tried yet) is to have your avatar in-world in the T-pose, and then rip the frame out of the GPU. A frame rip will litereally give you a WYSIWYG snapshot, because you can do that with a non-copybot viewer, as opposed to exporting from stored cached in a copybot viewer.
Ideas appreciated!