OctaneRender® 1.025 beta2.58b test release [Obsolete]

[matej]

The control points of the gradient node (the ones on the top bar) don't move anymore - you can move them by sliding the individual value, but not by clicking on them.

[roeland]

The control points of the gradient node (the ones on the top bar) don't move anymore - you can move them by sliding the individual value, but not by clicking on them.

The mouse interaction with some widgets in the node inspector (the gradient, and also the map in the daylight node) is broken in this release, we will fix this in the next release.

--
Roeland

[steveps3]

OH NO!!!

2.58b crashed whilst saving my project. Now every time I try to open that project it crashes again. I've tried other versions and they also crash. So it looks like all my work has been lost. 30+ materials all gone. The project parses all the textures and then displays the voxelizing dialog. This disappears and I get a white screen of death.

Anyone got any suggestions? I've been working on this project for about 2 months now and I would really hate to have to start all the materials from scratch.

[matej]

@steveps3, are those materials inside macro material nodes? If yes, PM me the .ocs (no .obj/.mtl, just the .ocs, if you can share it) and I'll see if I can help you get them out in .ocm files.

[PeterCGS]

Have you tried opening up the project witth 2.57?

[steveps3]

Sadly not. Most of them were just inside nodes.
I am looking at the .OCS to see if I can see what the problem could be. The .OBJ loads into Octane if I import it but obviously it has no materials.

I can send you the .OCS if you know about these things. For me I really don't know what I am looking for so I am hoping that something will just jump out at me.

[steveps3]

Yes Peter and it crashed.

[steveps3]

Here is the .OCS if anyone wants to have a dig about.

[matej]

The .ocs seems to be malformed. I used a script to extract top-level materials & macros, but it crashes because it doesn't find some nodes where it is supposed to find them. The .ocs is a very messy xml structure and searching what is wrong would take some time (manually looking at it is even worse). Maybe the developers have some debugging tools that could help.

EDIT:
Well, I got 6 of them out (only those that are not embeded in the mesh node). Though, I'm not sure of their state and if they will load.

ocms.zip

(10.69 KiB) Downloaded 126 times

Code: Select all

INFO: Parsing OCS version: OCS_1_0_23_Scene
INFO: 	-> glossy: "brick", (id=13) (pos=1.296 0.6925)
INFO: 	-> material macro: "frosted glass", (id=23) (pos=0.654 0.2375)
INFO: 	-> material macro: "ConcreteTileable", (id=24) (pos=1.003 0.58)
INFO: 	-> material macro: "Asphalt_1", (id=26) (pos=0.929 0.3025)
INFO: 	-> specular: "Specular Material", (id=27) (pos=0.71 0.4925)
INFO: 	-> material macro: "Rust", (id=28) (pos=0.428 0.1)

4 of them seem to be in a macro node, two are top-level material nodes, the others are "embedded" in the mesh node I suppose?

[steveps3]

I think that I have learnt my lesson. I am recreating the materials using material macros and saving these out. At least if it goes wrong again then I will be able to easily import the materials.

At least this is giving me the opportunity to improve upon the original textures. As you can tell, some of the originals were pulled from the liveDB. This time around I want to build my own textures from scratch. At least I still have to 60-70 hours of modelling. If I'd lost that then I think I might have cried.