Like the title says I think Octane is actually wrong with its IOR.
The way I understand it is if your going for accurate specular is 1.0 and you set your index accordingly.
Take a simple sphere make it black, add any HDR for lighting. Now make the index 1.52, this is 4.2% reflection at a 0 degree viewing angle. render it in any app, keyshot, modo, maxwell. Now do the same in octane.... it is way more reflective than it should be. Polarized and non polarized all start from the same position. at 589nm thats 4.2%. Octane is off a black plastic in octane looks very chromatic in a black env. Your only choice is to either lower the Index or the spec. but now you have no clue if its still accurate, of course do what looks right. But thats not the point.
Then take roughness into account. Set that same black plastic to .4 roughness.... you pretty much get a gray material. No other engine does this even physically accurate ones like Maxwell. So why is Octane look so far off?
Is this a issue with me and not understanding how to use Octane? I could be totally wrong, I am fairly new to octane. But my plastics no longer look plastic they are way shiney for their Index
Octane Render does not give correct reflection values
Forum rules
Please add your OS and Hardware Configuration in your signature, it makes it easier for us to help you analyze problems. Example: Win 7 64 | Geforce GTX680 | i7 3770 | 16GB
Please add your OS and Hardware Configuration in your signature, it makes it easier for us to help you analyze problems. Example: Win 7 64 | Geforce GTX680 | i7 3770 | 16GB
- ristoraven
- Posts: 390
- Joined: Wed Jan 08, 2014 5:47 am
What I have understood is that there shouldn´t be such thing called "specular" in the first place, if the idea is to make accurate, real world, materials. Only diffuse and reflection. Logic should go in surface: 100 diffuse - 0& reflection, 95% diffuse - 5% reflection. So that if the surface has reflection value of 5, then diffuse value should be 95%, or if reflection value is 35% then diffuse should be 65%.. so that the sum of these two is always 100. There should be no "specular" thing at all. Specular it self is an old and quick cheat to mimic something, but, if I have understood correct, it has nothing to do with "reality". There are no "specular" objects in real world, only reflective surfaces, or non reflective..
I'm confused by "The way I understand it is if your going for accurate specular is 1.0 and you set your index accordingly"
If you set the IOR to 1, you get a mirror/metallic reflection with no fresnel effect.
Off the top of my head, the exposure settings may make reflections look too strong.
Regarding the roughness, I'm pretty sure the settings in Octane are exponential, so .4 is pretty rough and the reflection is spreading over a large area, hence the grey. A roughness of 1 is not the equivalent of 100% roughness in Lightwave, for example.
If you set the IOR to 1, you get a mirror/metallic reflection with no fresnel effect.
Off the top of my head, the exposure settings may make reflections look too strong.
Regarding the roughness, I'm pretty sure the settings in Octane are exponential, so .4 is pretty rough and the reflection is spreading over a large area, hence the grey. A roughness of 1 is not the equivalent of 100% roughness in Lightwave, for example.
Glad to see you on board Seth.
If you're using the MODO plugin, the material should convert pretty closely. I know Funk helped get the roughness conversion more accurate than it was. But my understanding is the same as yours: set the glossy material spec to 1.0 and set IOR accordingly. For those materials that should have a defined reflection color, plug that color into the specular channel. The MODO conversion multiplies the reflection color and the reflection amount, I believe, and uses that for spec value.
The roughness in Octane is definitely using a different response curve. When converting a MODO material, one multiplier works at low roughness values, but does not match at higher values.
If you're using the MODO plugin, the material should convert pretty closely. I know Funk helped get the roughness conversion more accurate than it was. But my understanding is the same as yours: set the glossy material spec to 1.0 and set IOR accordingly. For those materials that should have a defined reflection color, plug that color into the specular channel. The MODO conversion multiplies the reflection color and the reflection amount, I believe, and uses that for spec value.
The roughness in Octane is definitely using a different response curve. When converting a MODO material, one multiplier works at low roughness values, but does not match at higher values.
I'm assuming you set diffuse to 0.0, specular to 1.0, and change the index?
Double-check you're not changing the film index instead, you can use this one to simulate a thin film effect on your material.
I did a quick test, and the reflection will be just above 4% at normal viewing direction, as expected. Can you post an image of what you expect vs what Octane is rendering?
--
Roeland
Double-check you're not changing the film index instead, you can use this one to simulate a thin film effect on your material.
I did a quick test, and the reflection will be just above 4% at normal viewing direction, as expected. Can you post an image of what you expect vs what Octane is rendering?
--
Roeland
- sethRichardson
- Posts: 95
- Joined: Thu Feb 27, 2014 2:16 am
Seems this was an error on my part. My camera exposure got cranked to 3 for whatever reason making highlights on a black plastic very bright.
One thing I am noticing is there is not any relation to specular and roughness. As roughness increases the specular component should decrease, it appears octane doesn't account for this, is this just something the user is expected to do?
One thing I am noticing is there is not any relation to specular and roughness. As roughness increases the specular component should decrease, it appears octane doesn't account for this, is this just something the user is expected to do?