[mgroeteke]

hello, i have problems with emitters working. after a failed test in a real project, i constructed a small test project, assigned a 100W emitter to the l-shaped surface but don't see any light emitting. same result with physical sky or skydome, no sunlight. what could be the cause? the fz material assigned to the emitter is a simple white diffuse one. the normals are correct (see screenshot) and the polys are triangulated, too.



if interested, i can send you the project file.

edit: here the file: http://www.3dworks.com/extra/mxw_fz_emitterfail.zip

[juan]

Hello,

At now I am not in Madrid. Next week I will analyce your scene. Sorry for the inconveniences. Try to increase the emissivity a lot (10000 W) just to check if this is a scale issue.
Regards,
Juan

[big K]

hello,

i think i found the problem:
if you use Power [W] as emitter units, then it doesen´t work.
so change to one of the 3 other types and it should do it. (adjust your value though)

of corse, this has to be fixed

cheers michael

[big K]

perhaps it might be because you are using mac. i am on pc using the windows version.
what about you markus ?

[juan]

Hi,

I have done a quick analysis to your scene (thanks a lot for sharing )
I have checked that there is no bugs in the way MaxwellZ pass emitter values to Maxwell. It seems just a scale problem.
In your scene if you display the "query object" dialog for the emitter you can see the total surface area:



As you see its value is 71.8 square meters. If you set the emitter units to 100 W, the total object emits this amount of light. If you set the units to W/m2 the object emits 100*71,8 = 7180 W, so you can see the light in the scene. I have done a quick test decreasing the scale to 0.01 and increasing the emisivity to 5000 W and now you can see how the body emits. I also attach other object to show you the importance of setting properly the direction of the normals to see the emitters. (your normals were right Markus ) This is specially important with planes with no thickness, only one side will emit light.



I hope it helps, please post any doubt. Anyway we are working to make everything more usable.
Yours,

Juan

[big K]

hello juan,

yes it seems to be a scale problem, but the emitter values are not calculated correctly. i have done a quick test.
here are the results:







as you can see, the overall intensity should be equal in the first two images. because when i calculate the area of the light and multiply it with the excitance value (image1) you get a power[W] result that is inserted in the power input in the second image. the images should look similar. but only when i use a factor of times 10000 the results are the same (image3)

i hope it helps to find out what is the problem here.

michael

[juan]

i hope it helps to find out what is the problem here.

Michael,
Great analysis. It will very helpful for us. We are investigating this issue making comparisons between different platforms. We will show you results as soon as possible.

Juan