Hi all!
As said Damien, the reposit is now open. You’re welcome to get it and compile deex shader yourself.
What we would like with this public reposit is that if anyone have a patch to submit, he is encouraged to do it and send it to us. 
The aim is to do a simple project to make compilation easy for anyone.
So use, learn, and share! 
Gracefully
nope… :sad:
If anyone know how to do it, he can use the reposit to make a copy of the project and send me a patch. We will add it!
http://bitbucket.org/narann/deex-shaders-pack/
Hey guys,
i’m having this problem at the moment which has stumped me. I’ve rendered out an EXR with raw and level reflection passes in it along with the normal reflection result pass. In nuke, i went to multiply the raw and level passes together where i expected it to match the reflection result pass exactly. However, it’s slightly different and doesn’t match correctly.
After spending some time trying lots of different combinations of render settings such as premultiply option,contrast etc… it always gives the same wrong result. I’m assuming that the raw multiplied by the level pass should be exactly the same as the reflection result after looking at the mental ray documentation to check. Looking closely at the level pass compared to the raw one, it looks like they’re slightly off each other, like an alpha issue but i’m not sure whether thats the case.
Has anyone had these issues before and know how to fix it? I’ve read about dividing between the passes and have tried this but had no luck
Cheers, Chris
Whenever you do any multiplication math in post you will always introduce issues like this, especially visible in high contrast areas. The same thing happens if you try multiply raw lighting against a colour pass. This works correctly during rendering because it is happening at a sample level, where in post you are dealing with final filtered pixels. The correct way is to only use additive math in post, and do all your multiplication in rendering. You can divide your passes in post to get the raw passes though, for example, you might divide colour / diffuse(which contains the colour premultiplied in) = diffuse raw
Thanks Mercuito, i was dividing the wrong passes when i was trying to use that technique. Just tried it again and it’s working now.
Cheers
This issue is very interesting. Can you point me to a place to deepen the subject? I generally use passes just to achive a “good” effect, but can be good to be aware of the theory.
The basic thing is clear but I’d like to know what do you exactly mean with “filtering” or sample level. And also (that could be weird): can’t we output a “filtering” pass separate from the “sample level” pass? Or can’t we do the filtering in post? Just wandering…
There is really no need when the divide solution gets you perfect results.
As for an explanation, I couldn’t tell you exactly how the Math works but maybe I can help with the theory behind it. The resulting RAW Pass from dividing will create inverted pixels at the filtering. You normally invert in Post when you need to remove the influence of something, in this case brightness at the filtering pixels causing the nasty artifacts. Inverting can come in handy for a lot of other things as well. For instance, you can take the quality of one shader and combine it with another without doubling the light intensity.
I don’t understand the inverting subject.
With “inverted” do you mean inverted RGB values? I don’t think you’re meaning this.
You normally invert in Post when you need to remove the influence of something
I can’t see how to use this advice. If yuo invert a b-w image the darkest pixels become white applying the effect in other areas; it does remove the influence, but only on one side…
With color images moreover this would invert the color that make even less sense to me…
Please can you explain this in other words?
(Or maybe do you mean inverting only a selection of pixels?)
edit: Or perhaps for “inverting” do you mean “dividing the result by the raw” (that off course is the main tecnique illustrated by Mercurito)?
Superior!
Could someone please explain me the benefit over the passes system which was introduced in maya 2009 ?
Hi all,
Feedback about auto gamma correct on shaders ?
There are sources since 1 month and no compilation for Linux or other ? :surprised
Thanks.
QFA.
I would love it if a 2011 mac version would show up. I can’t code, but I can buy beer!
I’m having some trouble with the install here, hoping someone can help me out. As you can see in the following screenshots the attribute editor I have for these shaders is looking quite different than it should. Here is first of all the Deex Mia shader, on the left is what it should be, and on the right is want I see:
Similarly, here is a side-to-side comparison of the Deex geometry buffer shader
I’m running Maya2011 on a PC with Windows7 (64bit).
What’s going on?
i just got this shader pack of deex and installed them on my maya 2011 x64. the shaders are all there but I have no idea on how to use them and test it in nuke. I have a box, sphere and a plane with the deex mia material x shader assigned to them and I have a deex buffer write node assigned to a random cube i created and i ticked the passes for the mia material x shaders of deex and also the same with the buffer write and i rendered it as an exr and tried to check in nuke and I don’t see anything in the channels or passes except rgba? what am i doing wrong? I am using only 1 render layer and no contribution maps. Sorry I am kinda confused on how to use this.
Anyone has a demo scene of basic geo for reference?