IES profile for light omni?

[Lales.Charles]

Hi,

We are trying to use an IES profile with a light omni, and the result is not necessary the expected one.

Simply with Excel, we can draw the profile:

If you open attached project, you'll see inside the sphere:

NB: after reconfigured the project to get back all missing files -> reimport ies file...

Would you see it as normal?

Kind regards,
Charles

ies-omni-light_export.zip

[silent]

Hi Charles,

Thanks for the example IES file.

Looks like a bug to me, need some additional time to verify and fix.

[silent]

Should be fixed right now. Please copy attached file into <Project>/data/core/materials/shaders/render/light/omni.h. This bugfix would be also included into the upcoming SDK update.

Btw, if you have more similar IES files - could you please share them with us? :) We would check if all of them are working fine after fixes.

Thanks!

omni.h

[Lales.Charles]

Way better, thanks!

Submit more IES profiles would be good idea, will check what I can share.

[Lales.Charles]

Hi,

Would have another questions related to light meter rendering mode:

• Is it normal that texture / albedo (here default grid_for_world_creation.mat) impact the visualization?
  Isn't it related only to received lights, not reflected one?
• Color scale is linear?
  Would it be possible to add some values under the axis?
  Would it be possible to directly specify color ranges somehow?

Regards,
Charles

[silent]

Hi Charles,

Albedo should not affect the final results, indeed. This behavior is introduced with 2.15 with new BRDF (which now takes in account albedo texture). Right now we don't have any plans to change this, maybe in the future releases, since there would need some additional changes to made that are not trivial.

 

Quote

Color scale is linear?

What do you mean by that? Could you please elaborate?

 

Quote

Would it be possible to add some values under the axis?
Would it be possible to directly specify color ranges somehow?

I'm afraid this list is hard-coded inside the Editor sources, so you can't really change anything here. I'm also not quite sure what is the use-case of that? This tool mostly used to calibrate lightning inside the interiors, so the legend is marked with typical interior lightning values.

Thanks!

[Amerio.Stephane]

Hi Silent, I'll complete Charles here. We are using LightMeter to calibrate lighting in and out of an helicopter (lighting from light projectors, position lights, beacons, screens, etc). We are comparing Unigine to more specialized industrial solutions, and we find that for a large part of 1st order estimates, Unigine can serve us right and save us a lot of time/money if we can somewhat "calibrate" it.

Now, in order to make a valid calibration, we absolutely need to measure the light intensity received by a surface and NOT the one reflected by it, so albedo/specular must be ignored. So please reconsider this point, or maybe add an option for it.

About the color scale: Here we need to be able to have both a linear scale (just like the one you have now) and a log scale (as the eye sensitivity is a log scale).

Also, having a direct reading of a color to a light intensity would be a must, as it is needed to check the actual intensity for a specific point in space. Maybe a "color picker" would help here.

If we could specify our own colormap, that would also help to highlight specific ranges.

We understand we may try to use this tool out of your intended usage, but we think it can really reach much more usage than the "simple"  interior lighting. :)

Thanks! 

 

[silent]

Yes, understandable. Current BRDF changes indeed affects this behavior, but the average error shouldn't be that big (maybe around 5%), so you still can use it keeping mind this numbers.

Right now we don't have any plans to fix this in the upcoming 2.15.1 update, but in the future for sure we would find a way to do this.