OTOY Forums

I find that using poly lights much quicker to setup in a scene as I have geometry that is already placed and I just add a material and adjust the settings. However I find that I get a quicker noise free SOLVE by scaling actual octane lights to fit the area around a light (bulb if you will). Mainly use DISC and RECTANGLES.

Is there a reason for this, am I doing something wrong?

The Octane lights are only polygons with the emission materials added by the plugin at the extraction time. They should have the same performance and features that the regular polygon objects with emissive materials, but of course if they are physically the same, I mean, with the same scale and the same number of polygons.

Thanks,
-Juanjo

That is very interesting. I will do a test on my end to see if I can replicate what I'm seeing on my end and through that figure out what I'm doing wrong or point something new out for you to check out.

Thanks!

juanjgon wrote:The Octane lights are only polygons with the emission materials added by the plugin at the extraction time. They should have the same performance and features that the regular polygon objects with emissive materials, but of course if they are physically the same, I mean, with the same scale and the same number of polygons.

Thanks,
-Juanjo

Ok so in my initial test all I can see is that the power of a Geometry light set to the same values as a Light is giving off different values. This over time/scale may be causing some issues for me, not sure.

Not that it needs to be changed, as that would affect lot's of old scenes for many I'm sure. It's interesting that they should be reacting the same (as you say it's programmed) but are not.

I had to reduce the power of the GEOMETRY light from 20 to 5 to get similar look on bloom and specular highlight on floor compared to the LIGHT having 20 power. Both were sampled at 20 and I did not see much difference as it continued to refine in IPR. *shrug*

I know I have had scenes that react much better when using actual lights rather than Geometry, but might be some other settings somewhere....sample rates or other...

Thanks for any thoughts you have on the attached images and settings...

This is interesting; I, too, have noticed that Octane "lights" often cause much less noise than the corresponding luminous polygon - which in theory doesn't make sense, given the way lights work in the plug-in. I've always thought it's either coincidence or me being a complete nincompoop, but maybe there is something to it?

ADP.

Is your geometry light source a plane disk attached to the bottom of your fixture or did you apply the luminous surface settings to the inner surface of the fixture? If the latter, you're seeing expected results, in my experience. The surface area of the inner surface is roughly 3-4 times the surface area of a disk at the base of the cone of the fixture (the shape of your fixture, anyway). That would give you 3-4 times the light output.

What happens if you turn normalize off for both of them?

Scott