shaders_p 3.3 released


#1

OK, enough for beta versions… let say it’s release 3.3 :slight_smile:
Full changes list after 3.2 version:

- Added p_shader_replacer shader (mostly for override shader on Fur)
- p_MegaTK: Can be used as photonic shader (diffuse, opacity, reflect and refract)
- p_MegaTK: Changed default value for 'Opacity Mode' from 'Alpha' to 'Average'(need for Maya's swatch)
- p_MegaTK and p_HairTK: Added SpecialC5-SpecialC10 passes
- p_scatter_wrap: Enabled the support for instance light lists (Mode 4)
- p_scatter_wrap: Fixed bug with wrong light direction if 'Spread' == 0
- p_scatter_wrap: Fixed NaN value if 'Max Scatter Distance' == 0
- p_scatter_wrap: Improved sampling pattern for 'Shadow Blur'
- p_HairTK: Added 'Color Variation' parameter
- p_scatter_wrap: Added 'Thin Wall Translucence' effect
- p_HairTK: Added 'Visible for FG' parameter
- p_MegaTK: Fixed bug in 'Fur Diffuse' part. It may produce a little bit different result
- Added p_dielectric_interface shader
- p_motion_to_rgb: Added 'SmoothKit' and 'ReelSmart' mode
- p_MegaTK: Fixed bug with NaN value in reflect/refract part. Fixed reflect/refract amount
- p_HairTK: Added 'Coefficient Back Diffuse' parameter. Diffuse may be look different
- p_HairTK: Added 'Saturation Variation' and 'Gain Variation' parameters
- p_HairTK: Fixed possible bug with incorrect specular normal. Result may be different. Default deviation value was changed (to -0.25)
- Added p_eye texture shader
- p_MegaTK_pass: Fixed bug with resaving all frame images when incremental mi and p_OpenEXR_saver used
- p_OpenEXR_saver: Added meta data to file (like camera matrix, focal, aperture...)
- p_MegaTK and p_HairTK: Fixed crash with detail shadow in rare cases. Actually it's mental ray bug
- p_HairTK: Fixed rare crash bug with rasterizer and raytracing enabled. Actually it's mental ray bug, state->bary[0] return NaN value
- p_MegaTK_pass: Fixed possible crash if 'half' buffer type used with 'frame buffer gamma'. Now frame buffer is always 'float'. It's mental ray bug that was fixed in 3.8.1.20 (Autodesk branch) and 3.8.2.3
- p_math_output: Fixed possible 'inf' value
- p_MegaTK: Added some optimization for final gather with bump if 'Ignore Bump' enabled
   And changes list after last 3.3 beta 11 version:
- p_MegaTK_pass: Fixed bug with resaving all frame images when incremental mi and p_OpenEXR_saver used
- p_OpenEXR_saver: Added meta data to file (like camera matrix, focal, aperture...)
- p_MegaTK and p_HairTK: Fixed crash with detail shadow in rare cases. Actually it's mental ray bug
- p_HairTK: Fixed rare crash bug with rasterizer and raytracing enabled. Actually it's mental ray bug, state->bary[0] return NaN value
- p_MegaTK_pass: Fixed possible crash if 'half' buffer type used with 'frame buffer gamma'. Now frame buffer is always 'float'. It's mental ray bug that was fixed in 3.8.1.20 (Autodesk branch) and 3.8.2.3
- p_math_output: Fixed possible 'inf' value
- p_MegaTK: Added some optimization for final gather with bump if 'Ignore Bump' enabled
   As ussual you may download shaders_p here: [http://www.puppet.tfdv.com/download/shaders_p_e.shtml](http://www.puppet.tfdv.com/download/shaders_p_e.shtml)
   There are only these versions that I could compile on my machine are available on my site in current time.

Possible I’ll add more versions later…


#2
  • p_MegaTK and p_HairTK: Added SpecialC5-SpecialC10 passes

been waiting for this so long :applause:

wow thank you very much Puppet|!


#3

Thank you for all your effort Puppet, really appreciated!
/Rachid


#4

Thanks Pavel for the 3.3 release! :applause:

I posted this in the older 3.2 thread as well:

Here is a couple side-by-side comparison renders of the issue that I am having using the p_z shader. Notice the difference in the refraction results. The shader works but is giving me different outcomes for refraction even though the IOR settings are the same (IOR = 1.52). I am using a ‘dielectric_material_photon’ in combination with the p_megaTK shader for the caustics. Anyone know what I am doing wrong here? I need the p_z shader to give me the same result that I am getting on the left image for me to make any use of the zdepth pass. Any help is greatly appreciated.


#5

I just ran a test this time without the dielectric_material_photon controlling the IOR in the glass material. I thought that perhaps the ‘shaders p’ materials handled the IOR results differently from mental ray’s materials. Turns out they are handled exactly the same. So, this confirms that the ‘p_z’ shader’s IOR setting is broken? Or am I just not using this shader correctly? Any ideas?


#6

p_MegaTK have “more glass” reflection/refraction.
If you need exactly same z you should use build-in “Z” pass in p_MegaTK.

Why you are using dielectric_material_photon? Why no p_MegaTK? In 3.3 version it can be used as photonic shader.


#7

Thanks for the input Pavel. I didn’t realize the new 3.3 version allowed this capability. Now I know. :slight_smile:

So, what if I only need the “Z” pass? Can I have it calculate and render the “Z” pass only using the p_MegaTK shader? I already have the beauty pass and I don’t want to have to re-render it again. Print resolution = long render time.


#8

No, sorry. You should rerender all passes.


#9

:bowdown: As Rachid did already, I want to thank you one more time, for all your efforts Puppet.
We all appreciated!

Alex


#10

That’s what I was afraid of. :cry:


#11

Thank you for all Pavel. Your shaders are great and you give source, it is excellent to learn.

:bowdown:


#12

I’m using p_MegaTK_pass to enable passes. I can’t seem to get the z pass to come out correctly. The problem is that it is not obeying transparency/refraction. Anyone know how to get this to work properly? All other passes come out fine.


#13

COOOL!
It does deserve a huge applause … :applause:
great work Pavel !


#14

Hello Pavel,

I have found a NaN bug in p_HairTK. When I turn on the transparency in the diffuse occlusion and color passes the R channel has some NaN pixels.

The previous release (shaders_p v.3.3 beta 11 for Maya 2009 (win32 win64) ~4.4mb 23.06.09) is okay, but the new release (shaders_p v.3.3 for Maya 2008-2010 (win64) ~4.8mb 13.01.10) has this issue.

This is the link where are the beauty, diff occlusion and color passes in 2K merged next to each other. On the first frame the transparency is on and the transparency color is dark grey. On the second frame the transparency is still on but the trans color is black. On the third frame the transparency is off.

I would like to emphasize that this is different than the previous bug in the beta 11, because when we re-rendered a wrong frame with it (btw. it was on the spec shadow and shadow passes) it became good. However this bug stays on the aforementioned passes constantly.

I used maya 2009 64bit on vista and shave and a haircut with hair primitives and various voxel and buffer sizes.
In the scene is a sphere with the default shave preset and one spotlight (i turned off the shadows for quick renders).
If you need the scene I will PM it.

I would be very happy if you could find some time to track this bug down. Thank you in advance.

0G


#15

Thank you for images.
Example scene will be great, but I have no Shave. Can you reproduce this issue with regular Maya Fur or Hair?


#16

Pavel, can you say when shaders pack for 2011 will be released?


#17

I suppose on next week. I just need download and install Maya 2011.
I have tried it only with Maya 2011 RC2.

Possible it will be compiled only for win64 and linux64 (for the first time).


#18

Added builds for Maya2011_win64 and Maya2010_win32.


#19

THANK YOU ! :bowdown:… really great.


#20

Thank you, Pavel