OctaneRender™ Standalone 1.22

[roeland]

Found a bug. Octane doesn't read the .MTL files specular (Ks) values. Setting the .MTL file to (Ks 1.0 1.0 1.0) or (Ks 0.01 0.01 0.01) doesn't changes the specular value at all.

Once an OBJ file is loaded into Octane, reloading will only reload the geometry, not the materials. So you don't lose your material setup in Octane if you want to reload the geometry.

--
Roeland

[Zay]

Once an OBJ file is loaded into Octane, reloading will only reload the geometry, not the materials. So you don't lose your material setup in Octane if you want to reload the geometry.

I'm not talking about reloading. But simply opening an OBJ the first time. Everything is looking like chrome now. In 1.20 this could be fixed by editing the MTL file before loading the OBJ. In 1.22 you can't which add 100 times more work to fix all materials. So a fix is needed.

[Rikk The Gaijin]

I want to report a problem I found. This is not related to 1.22 specifically, it's a problem that was ALWAYS present, even in previous versions.

Basically, the absorption node does not read the model's UVs.

Here's some example:

This is how the model looks when I use an RGB color in the absorption node, and it's correct.



This is how the model looks when I use a checker texture in the transmission node, and it's correct.



However, if I put the checker texture in the absorption node, this is what I see



I tried different textures, with colors, and it seems the the absorption node is reading the texture but it's ignoring the UVs. Also, the new 1.21 mapping features are ignored as well (and, for some reason, the preview sphere is different, although is not important).
Again, this problem was present also in previews versions, but I was hoping to see it fixed in the new one.

[Rikk The Gaijin]

I forgot to say, that the reason why I want to use a texture in the absorption node, is to add subtle variations in the sub-dermis layer of the skin, such as veins, capillaries, etc. Having a limitation of being able to put just one flat color is a bit frustrating.

[roeland]

The mapping features for absorption were disabled in version 1.20 so overlapping volumes could be implemented. It is quite hard to put this back in without degrading performance too much.

--
Roeland

[roeland]

Once an OBJ file is loaded into Octane, reloading will only reload the geometry, not the materials. So you don't lose your material setup in Octane if you want to reload the geometry.

I'm not talking about reloading. But simply opening an OBJ the first time. Everything is looking like chrome now. In 1.20 this could be fixed by editing the MTL file before loading the OBJ. In 1.22 you can't which add 100 times more work to fix all materials. So a fix is needed.

OK, I see the problem. It will be fixed in the next release.

--
Roeland

[Rikk The Gaijin]

The mapping features for absorption were disabled in version 1.20 so overlapping volumes could be implemented. It is quite hard to put this back in without degrading performance too much.

--
Roeland

This is a really bad news. In other words, this won't be fixed?

[abstrax]

Octane preferences -

I see now we have a 'preferences_v2' file. I guess we an delete the previous preferences file now? Hope so cos I did.

If you intent to use only versions >= 1.21, then yes, you can delete the old prefs file. Be aware that most of the preferences are not stored in OctaneLive anymore.

I note that v2 has a different order. The 'recently used' list is now in the middle whereas it used to be at the end.
This makes it even more inconvenient to edit in the event that the file location has changed or it has been deleted altogether.
Something I have asked for for several years now is to automatically remove a file from the list if it cant be found.
Its kind of annoying to have redundant file names in the list. Please can we tackle that sometime just as a user friendly little nicety.

We added a MRU file list cleanup to the Octane start up. It will be available with the next release.

[abstrax]

The program indicates always that memory is 100% occupied 3072/3072 MB but it renders. So memory monitor is not working correctly.

I confirm that too. MSI afterburner shows correct values of varm usage while Octane hits incorrect 100%.

Does it happen all the time? If yes, what exactly does the memory usage say (all 3 numbers) and what is reported in the CUDA device tab of the preferences dialog?

[pixelrush]

MRU cleanup... in the next release

Wow! Really? Gosh.... well that makes my year, plus last year and the year before.
What a Merry Little Octane Xmas this is becoming.
Thanks so much guys! Woot!!