OTOY Forums

You know your scene very well, but no one else does. I cannot determine what the issue might be just by looking at your render, and that is why we need the actual scene file to review. Without your scene, it will be difficult for us to offer any cogent advice.

I've updated the builds to R2 with some fixes. And uploading....
Now material import will support most of new nodes and fixed some conflicts of material import. It was due to shift by new custom aov feature. Also fixed renderlayer selection. And more small fixes...

Hi Ahmet, thank you very much for the quick 2021 preview release.
Did I get it wrong
"Improved rendering of overlapping volumes and increased the number of volumes allowed to overlap in one location to 16"
Screenshot attached.

thanks
Yassin

y81r wrote:Hi Ahmet, thank you very much for the quick 2021 preview release.
Did I get it wrong
"Improved rendering of overlapping volumes and increased the number of volumes allowed to overlap in one location to 16"
Screenshot attached.

thanks
Yassin

As I'm a plugin dev that's not my part. Please send your test scene with assets to let us check that.

aoktar wrote:

y81r wrote:Hi Ahmet, thank you very much for the quick 2021 preview release.
Did I get it wrong
"Improved rendering of overlapping volumes and increased the number of volumes allowed to overlap in one location to 16"
Screenshot attached.

thanks
Yassin

As I'm a plugin dev that's not my part. Please send your test scene with assets to let us check that.

Ok, did it.
Thanks

Hi,

These are artefacts from a too large step length. It should fit inside the volume a few times. The clouds are about 1.5 units across so values up to 0.5 may kind of work. If you set it larger the ray marching is mostly skipping over the clouds making them almost invisible.

With the densities that are set up the clouds should appear much thicker than this.

In general, in 2021.1 the appearance of volumes will be less affected by large step lengths, but at some point you’ll get artefacts because large parts of the volume are simply skipped over:

There is another artefact on the boundary of the volumes because the phase function is set to a different shape in both volumes. This is not supported in either 2020.2 or 2021.1 XB 2.

So if we compare 2020.2 and 2021.1, with the phase function set to 0.7 for both clouds:

At step length 0.01 and 0.05 both look similar:
At 0.20 the quality goes down notably. 2020.2 is appearing a bit brighter:
At 1.00 we’re pretty much not able to render the clouds anymore. 2020.2 looks like it has much lower density, however this effect is achieved in a more controllable way using the density input.

Thank you so much Roeland for the detailed description.
The same values for Scattering Phase did the trick.
can you please explain the following
I have stacked the clouds several times, but the "Max. overlap volumes"
between 1-16 in the kernel settings has no effect on the VDBs, so what is the setting for?

Thanks
Y

There's some problem happening with RTX:
RTX off: RTX on:

y81r wrote:Thank you so much Roeland for the detailed description.
The same values for Scattering Phase did the trick.
can you please explain the following
I have stacked the clouds several times, but the "Max. overlap volumes"
between 1-16 in the kernel settings has no effect on the VDBs, so what is the setting for?

Thanks
Y

If the overlap value is lower then sometimes you can see the edges of the bounding box in the rendered image. You’re most likely to see these edges if you have a lot of volumes arranged in a row with a lot of overlap. At the moment the minimum used for this setting is 4, which is enough for most scenes.

marcus1070 wrote:There's some problem happening with RTX:
RTX off:

Screenshot_000104.jpg

RTX on:

Screenshot_000106.jpg

Try to disbale nested dielectrics in octane kernel settings