place a polygon, facing down into a room, with a lightmtl on. it lights the room. if its close to the cieling, it casts a gi shadow on the cieling.
this still happens if the object is set to “invisible to camera” or “affect reflections and refractions” is disabled (obviously)
turnng off “recieve gi” doesnt help
and obviously, turning off “generate gi” or “visible to gi” will stop it emitting light.
thats all the options we have to play with.
i understand why it happens, and you can kinda reduce it near zero by putting a transparent material on the back face of your light emitting plane or whatever, but thats not a proper fix, and doesnt work completely for some reason.
- it would be good if the lightmtl shader had some way to kill this effect if required.
Hello,
Can you post a sample scene?
Visible to GI look to work.
but in your test scene, is the plane illuminating the space? its hard to tell. when i turn off “visible to gi” the shadow goes, but so does the illumination contribution from the lightmtl. (i dont have “direct illumination” ticked, as i may as well use a vraylight in that case.)
im on a deadline at the moment, so i cant do a test.
Here is the scene:
https://ftp.chaosgroup.com/support/visible\_to\_gi.max
Yes, the plane is illuminating the scene, check the file and let us know if you have more questions.
ok i could not open your scene as im on max 2014, but i created a similar one in a couple of minutes.
im not sure what result you are seeing, but if you open my file, the plane is clearly not lighting the scene. its set to “invisible to camera” and unticked “visible to gi”
if you select it and tick " visible to gi" it lights the scene as desired, but you have the shadow above it on the cieling.
this is also the case when it is visible to the camera.
to be honest if its set to “invisible to gi” i cant see how it can possibly light the scene, since the gi rays are passing straight through it, and lightmtl works by adding light to the gi rays.
visible to gi.zip (21.7 KB)
Here is the file for max 2014.
Please check if this is the result you want and let me know if you have more questions.
visible to gi_2014.zip (22.1 KB)
but, you have enabled “direct illumination” to get it to emit light. my point was that it doesnt work *without* using direct illumination. in this case id use the more efficient vraylight.
id like the result you have, but without “direct illumination” enabled.
that sounds like just the solution i was after! many thanks! i always assumed that areas of the light with black in the opacity slot would not emit light, obviously not 
