Validation of LiVi results for artificial lighting

[Brett Horstketter]

Hello!

I want to use this amazing tool for a lighting design project, but I am having trouble getting lux values and w/m2 values that make sense to me. For reference, I also use Dialux Evo, so I am kind of comparing what I see in one vs the other and trying to understand why I seem to be a ways off using VI-Studio!

I verified I was doing everything in dialux in meters, and the same in blender, but for the same IES file at the same distance away from a calculation plane in empty space, I am getting 11x the lux using VI studio vs dialux, in what should be a reasonability identical set up.

If I can get set up properly, I would love to help craft a tutorial for indoor lighting simulation that includes some assets, perhaps an .ies file and a .blend that contains a basic room to light up. That way a new user can follow along very easily.

Thank you for creating this awesome vector of many things I enjoy, blender, python, and bright light!!

regards,

Brett

[Lars O Grobe]

Hi Brett,

just one potential source of error to check: IES files can describe absolute photometry (e.g. the intensity you would measure to a given direction) or a relative distribution, e.g. intensity normalised to some value. The latter allows to use a multiplier accounting for the lamp output, reflecting that you usually can combine one reflector with different lamps.

Best, Lars.

Am 04.10.2024 um 01:51 schrieb Brett Horstketter <bretthor...@gmail.com>:

Hello!

--
You received this message because you are subscribed to the Google Groups "VI-Suite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to vi-suite+u...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/vi-suite/44e63ae2-7041-4b36-b7d2-d00b9006c77cn%40googlegroups.com.

[Brett Horstketter]

Thank you Lars, I am working with absolute photometric files generated by UL Labs so shouldn't need a scaling factor (if I understand correctly!)

You give me a good idea to try though - I wonder what I would get if I normalized my IES file and then used the built in IES scale factor in VI... I will do some testing and share my results.

[Brett Horstketter]

I have put together a comparison of what I think are equivalent setups... A 1000W HPS Light 2 meters from a 4x4m plane. The calculation surface is 2x2m. The distribution looks very similar but the values are roughly 4 times different with this file. I have attached the IES file and the .blend!

Here is my node set up too:

Hoping to test a normalized ies file over the weekend!

Brett

[VI-Suite]

Hi Brett.

Just to confirm, you are using the same IES file in both programs?

[Brett Horstketter]

That is correct, same IES in both programs!

To complicate things - I opened up the .blend again to double check and for some reason couldn't recalculate. Started over and got MUCH more similar values - this time I left the IES settings alone. 

Previously I switched the IES dimension to feet, to match the unit set in the IES file for dimensions...

I must be misunderstanding what that unit drop down is for - I assumed it was supposed to match up with the units set in the IES file:

TILT=NONE
1 -1 1 91 13 1 1 0.833 1.625 0
1 1 1053

(pulled from the file in use, the bolded 1 signals feet for the next three values which are length width height)

Here are the updated results using Low Accuracy (which lines up very nicely overall with the Dialux output):

Light source settings:

Thank you!

[VI-Suite]

I too assumed the ies2rad option to pick the dimension unit was to cater for an undeclared dimension unit within the IES file, but as IES files contain the unit information I'm not sure what this option is really for, and I may remove it in future versions.

Maybe there are IES files out there that don't declare, or declare incorrectly, and this allows incorrect data to be overwritten. In short, use with care.

Cheers

Ryan

[Brett Horstketter]

Really appreciate your time, Ryan! 

I'm a pretty meager programmer, but if I can find an answer in the ies2rad source, I will report back with my findings.

Brett