I was doing some simulation analysis with these two methods and had some strange results. Could it be the climate file thing, because I keep getting the same results.
Being inexperienced in these things (but I’m getting there)) I have little idea of what could cause these weird results.
I’ve internalised all the necessary data, sliders are set to represent problem instances, definitions are safe to open.
Honeybee_Grid-based_Daylight.gh (498 KB)
Honeybee Image-based Daylight Analysis.gh (486 KB)
Your window polygon and walls appear to be coplanar and your walls do not have any holes in them for the window apperture.
I rendered three images in Radiance using your model.
The first one is the full model. I think the rays intersect the glass_material first so we get to see inside the room. The next one is without the glass. As you can see there is no recess inside the wall. The final one is without the walls.
In a physically based renderer like Radiance, for accurate results the geometry should mimic reality. The only way to actually punch into material using Radiance is to use Antimatter.
If you don’t want to manually subtract the geometries you can use addHBGlz component to add glazing to a surface.
I know, I tried your’s and Chris’s methods and they all give the same results, could you please test the file for me?
I did like Mostapha said, also checked the method with another analysis and it gave me expected results, still have no idea why image based analysis didn’t work.