http://pasteall.org/blend/index.php?id=50535

Re-attached image. I'm on an iMac Pro. I'm glad it doesn't happen for you, but it happens consistently, 100% for me. The iMac Pro has some significant architectural and component differences from the other iMac models and (e.g. Xeon instead of Core I chips, Radeon Pro Vega instead of Radeon Pro 5xx).

There is no image
Also i'm on an iMac and i don't seem to have the same problem as you

AFAIK unless you start changing materials you don't break anything. It would be fine to have a warning but this issue kind of resolves itself with 2.8 materials being synchronized across renderers.

Not sure here but have you trying holding ALT before clicking into the number field and then typing it? There is sort of a hidden feature in Blender that if you hold down ALT the action gets performed for all selected. But that does not work completely consistently (probably why it's hidden) so it might not work here.

Text is batched in order to reduce drawcalls count. You might have to flush the text draw cache in order to draw things in a specific order. See usage of BLF_batch_draw_flush().

In T57292#544150, @Philipp Oeser (lichtwerk) wrote:

Can confirm in 2.79b (official release), but not in 2.8 nor 2.79 master.

@Alexander (deepforest) : havent tracked down a particular commit, but this might be solved already.
Could you try a new 2.79 build from https://builder.blender.org/download?

If you can confirm the new build works, then this could be closed as resolved...

SUMMARY: AddressSanitizer: heap-buffer-overflow //source/blender/draw/intern/draw_cache_impl_mesh.c:4430 in mesh_batch_cache_get_triangles_in_order_split_by_material
Full asan error backtrace: P804

Thanks for the answer Juan, that clears it up for me!

Hi Philipp Oeser (lichtwerk),

In T52063#544372, @Juan Pablo Bouza (jpbouza) wrote:

From my experience, you might want to use a copy loc + copy rot + copy scale + damped track + stretch to constraint on each of the deformation bones, instead of just a copy transforms constraint. This has given me better results on export.

Andy, well, yes, the thing is that you've gotta have a control armature and a deformation armature. So, in the control armature you set all the constraints up, spline IK and stuff. When you have everything working, you create the deformation armature that consists of bones that might be parented just to the root so that they don't have a child/parent hierarchy. You then add a copy transforms constraint to each of those bones, pointing to the spline IK bones of the control armature. That way the deformation armature will mimic exactly what the control armature does, but it won't have a hierarchy that ruins the export.
From my experience, you might want to use a copy loc + copy rot + copy scale + damped track + stretch to constraint on each of the deformation bones, instead of just a copy transforms constraint. This has given me better results on export.
Finally for exporting, you just select the mesh and the deformation armature, you don't export the control armature.

Juan but is it really possible to get spline IK to work without a standard bone to bone parent hierarchy?

Im guessing maybe so too, though rolling back with clean old drivers, does naught to alleviate the issue, it locks up a LOT when using the camera view and rendering with the border, i am resorting to using cpu for now though that would not be my preference.

Please try uninstalling the NVIDIA graphics drivers and installing the latest drivers from the NVIDIA website with the clean install option. The ones installed by Windows seems to sometimes get corrupted CUDA / OpenCL.

I guess something broken with a newer AMD driver.

This should be solved in the latest builds, the one you have is from before the fix.

Right, that console log has the same error messages as T57296: Blender 2.8 crash on startup on Linux, so it's the same bug.

System Information
Windows 10 Home 64-bit, GeForce 9600 GS 768MB (NVIDIA driver 342.01, which is the latest available for this card)
OS up to date.

I think this the same issue as T57296: Blender 2.8 crash on startup on Linux, assuming that earlier 2.8 builds did work.

Just an update on this:

I guess when converting the UVs to indexed format it should take into should group by identical (u, v, vertex_index) instead of only (u, v).

Andy, what I mean is that all bones should be parented to the root bone or some bone that doesn't scale. When you do that, the result is more or less reliable, although it's never 100% correct for some reason.

Let's close this as a duplicate..