Hey guys,
i’m trying to get the right passes ( Diffuse / Reflection / Shadow / Direct Irradiance etc )
out of my image, but it seems that the new ibl mode is not supporting the passes right.
Maybe someone can confirm that issue.
BTW i’m using the light mode without FG.
regards
naik
BTW guys,
i have recognized one certain issue when using the new ibl light mode.
Sometimes the light of the hdr is so bright that i have to set the light scale down.
But when i do this, i also get a much weaker reflection. Is that normal?
It would be very cool to have an idependant control for the light scale / power and the
reflection strength. Maybe i’m missing something.
regards
It acts like an area light so lowering it will make it less correct overall.
If using native IBL and a good HDR you will have some hot spots. Make sure you are using the string option to correct for non-bsdf shaders in Maya 2012:
light relative scale
0.318
scalar
It is considered direct lighting, just like an area light for passes.
Hi David,
thanks again for dropping in.
Can you perhaps elaborate a little bit more?
If i add the relative strength i can clearly see that the light strength is
going down, therefore less burned out areas.
I can remember the number in combination with the physical sun/sky as
a conversion number.
regards
When setting the rgb Unit Convertion values to 0.318, the final rendered pixels in the image are true photmetric luminance values in candela per square meter.
Never really understood why exactly 1/Pi but I guess it may have to do with the hemisphere and Pi beeing 180degrees ?
The lighting convention used in non-BSDF materials (or those with similar conventions) are scaled by pi. To correct for that when using a modern technique you should invert that.
Light relative scale does that for you and it ALSO allows you to mix and match old/new materials without them lighting differently.
This is going to become more important as time goes on. . . . 
If your HDR has a very hot source(s) of light I would be sure to add a mia_envblur and sample more (Unified Sampling) to improve the highlights.
What exactly is the new lighting mode on the IBL? Is this something that has to be enabled to be visible in the AE? Or are you talking about the Emit Light
good ‘old fashioned’ Light Emission has been on the IBL since Maya 6.5 I believe. Paulo Berto wrote a lengthy article on its use many years ago, I believe it was the first edition of trixr4*kids. It was just to slow to use in production back then… too low a quality (ie # of area lights) and low samples meant lots grain in the renders, but it was the only way to get a render out in a decent amount of time.
I will try to find time to do this. 
More informations here:
http://forum.mentalimages.com/showthread.php?5717-Builti-in-IBL-new-feature-in-3.7.50.11
:wavey:
Hi Narann,
I would make that string option a default and leave it on (under the hood) no need to have it off at all really.
Native IBL is a newer and [therefore] better implementation of the emit light function created by Nvidia for mental ray.
Is it me or the new ibl light is far slower than VRay’s dome light?, right now is useless in an animation project.
I’m rendering 3000 x 1688 frames at 15-20 minutes a frame with FG + IBL.
Seems perfectly acceptable to me. In fact, with Unified and motion blur some frames dip to 12 minutes a frame.
When I tried it in 2011 it was a bit disapointing for me, too slow. Do you use unified sampling in your 15 minutes renderings?.
No, regular AA because of a framebuffer bug. The bug has been fixed but not in this version of Maya. Hopefully 2013.
What is the dynamic range of your HDR?
I didn’t really try it very hard, I just put a 32 bit float image and render something…, it took quite long and stopped trying. I guess I need to give it a second try.
I love the IBL too, use it almost everytime. And it’s not slow, actually I think it’s quite fast for what it looks like. Don’t turn your quality knob to 1, unless you need a super smooth still. For animation I almost never go further than 0.6~
And good to know Bitter, that it’s a Maya bug - I think you’re meaning the extreme noise in passes like zdepth when using Unified Sampling? At least I had this noise with the Mitchell filter, didn’t try Gauss though.
Yeah, it’s a filtering bug (filter off works fine, Z should not be filtered anyway) but things like matte passes etc are bad.
It’s been fixed.
