All posts related to V2
By aqualis
#363501
Hello,

Is there a maximum of emitters that can be controlled in multilight?

I expect to have over 100...
User avatar
By Ernesto
#363502
It is a matter of how much Ram memory you have.
I couldn´t handle more than 15 channels with 8GB, but I guess it depends on the complexity of the scene. You can also can have all the emiters you want for each channel.
E
By hatts
#363504
Emitters are controlled in Multilight via materials. If all 100 emitters have the same material, you use just one slider. If you want to control each emitter individually, you need 100 separate materials.
User avatar
By dariolanza
#363538
Hi aqualis,

Yes each different material emitter will have an independent MultiLight slider, meaning if you have 1000 ceiling point lights using the same emitter material, they will only use one single ML slider to control them all, easing the way to adjust them all in one single step.

Said that, the maximum number of sliders in ML (the maximum number of independent emitters) is not defined by Maxwell, as Ernesto says, but you Ram memory.

Each MultiLight slider is like an image buffer, meaning that an a render with MultiLight is like a multi-layered file in Photoshop. The more layers (sliders and extra buffers like Zbuffer, materialID, objectID...) and the bigger the image resolution, the more Ram needed to host all that pixels information in memory.

If your render is too big in resolution, you have extra buffers enabled (sliders and extra buffers like Zbuffer, materialID, objectID...), and you have many MutiLight sliders, you can come with an image file too big for your Ram memory, and your system could crash due this.

But related to Maxwell, there is no upper limit defined in the code.

Greetings

Dario Lanza
By aqualis
#363540
Ok, thank for the info!

In my case, I will need the 100+ emitters. Even though all the lights are of the same color and intensity, the lights are based on a single "image emission texture" which has different coordinates when it is applied to different (100+) surfaces.

Also, the 3D geometry is in itself already very heavy and I won't have enough RAM to deal with it all.

I will try and render the emitters separately and put the .mxi image together in Photoshop.

So, maybe this could be a handy feature for future Maxwell versions: to be able to render just the emitter as an .mxi, based on the 3D geometry of the file. So all the reflections / refractions are taken into account for this particular emitter, but no RAM is spent on rendering any other emitter or sun / daylight. This would allow to break up the rendering process in smaller bits and composite everything in Photoshop. At the moment, when I use the multilight feature, I still have to render sunlight / daylight with the emitter I'm targeting. This takes up a lot of (wasted) time.
User avatar
By Mihai
#363547
aqualis wrote:At the moment, when I use the multilight feature, I still have to render sunlight / daylight with the emitter I'm targeting. This takes up a lot of (wasted) time.
Not sure I understand, you know you can turn off the environment lighting before rendering right? If you DO need also a render with sunlight, you won't save any time rendering first the sunlight, then the emitter. So you might as well leave both the emitter and environment and turn on Multilight. If you don't need a sunlight render, just turn off environment lighting before starting the render....
By aqualis
#363568
Thanks for mentioning that, Mihai, actually I don't know where to turn of the environment. I do know how to do that for the sunlight though. Can you point me to it?
(FYI, I use rhino plugin, in case that would make a difference)

Tx!
User avatar
By Ernesto
#363685
aqualis wrote:So, maybe this could be a handy feature for future Maxwell versions: to be able to render just the emitter as an .mxi, based on the 3D geometry of the file.
I really have no idea of what are you trying to do... but I thought that perhaps this is what you are looking for:
http://support.nextlimit.com/display/tu ... ght+Studio

E

Haha, thanks.

Hello, I'm still waiting for a solution to the pro[…]

Well.....they must have been proven wrong, as it's[…]

Hello dear customers, We have just released a new[…]