Doing manually what mkillum does...

Dear Radiance users,

I hope you all had a nice christmas and are happy playing with your brand-new radiance number crunchers already (time to have a look at the christmas peak at Mark's benchmark site again :wink: )...

I am having a question which may be basic for those here doing the calculation-and-formula stuff on a regular basis, but appears confusing to me. To simplify a model, where direct sunlight is not important and daylight can be considered as a hemispherical light source (uniform sky), I would like to do manually what mkillum does when producing a light modifier (not the default illum modifier). The surface to be modified is a horizontal flat rectangular face of, say, 1sqm. If I would normalize the sky source in a way that I would get a horizontal illuminance of, lets say, 1000 lux, ignoring obstructions, how would I calculate the 1 sqm light source's modifier that would replace the sky in my scene? Is mkillum basically calcultating an iradiance image here?

Lars.

Hi Lars,

Merry Christmas -- I hope you didn't get the virus I caught that brought my whole system down for two days on the 24th. I'm still feeling a bit weak, so apologies if none of this makes sense...

Setting d=0 in one of the #@mkillum comments in your scene file will create a diffuse source, which is what you want in this case. (And yes, it will be computing irradiance in effect.) Since it takes almost no time to compute such a source, you may as well use mkillum if you're comfortable with it. If you want to create the source by hand, then simply divide your external irradiance by PI to get the radiance of the source, whatever size it may be (doesn't matter), multiplying of course by the transmission as well.

Cheers,
-Greg

···

From: "Lars O. Grobe" <[email protected]>
Date: December 29, 2008 12:58:02 AM PST

Dear Radiance users,

I hope you all had a nice christmas and are happy playing with your brand-new radiance number crunchers already (time to have a look at the christmas peak at Mark's benchmark site again :wink: )...

I am having a question which may be basic for those here doing the calculation-and-formula stuff on a regular basis, but appears confusing to me. To simplify a model, where direct sunlight is not important and daylight can be considered as a hemispherical light source (uniform sky), I would like to do manually what mkillum does when producing a light modifier (not the default illum modifier). The surface to be modified is a horizontal flat rectangular face of, say, 1sqm. If I would normalize the sky source in a way that I would get a horizontal illuminance of, lets say, 1000 lux, ignoring obstructions, how would I calculate the 1 sqm light source's modifier that would replace the sky in my scene? Is mkillum basically calcultating an iradiance image here?

Lars.

Hi Greg,

get better befor new year! I am pretty safe from flue and cold, sitting just some meters from the equator in nice and warm Singapore...

Thank you for your reply! It made me think once more about what I am asking for, and essentially the funny thing is that in my case, having a uniform sky, this led me to the (late) insight that I can simply use the luminance of the sky source (in this case 1000/pi) to modify a surface on the entrance of the system - nothing else I do when applying the gensky-created description to a window pane. So the luminance of my replacement-source is the same as the luminance of my uniform sky source (as long as I ignore the transmission properties). It just took me a while to get it that, when looking through a perfectly transparent window, I see the sky luminance and not the window glass. Maybe a sign that I have been working in a window-less office space for too much time now...

I wish you all a happy new year and a nice celebration, those in cold climates beware of the flue,

Lars.