Show Us Your Iray Renders. Part II
This discussion has been closed.
Daz 3D is part of 
Connect
DAZ Productions, Inc.
224 S 200 W, Suite #250
Salt Lake City, UT 84101
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
In sRGB space, a 50% mid-gray is not 0.5 or 127 but rather 0.5 raised by the inverse of gamma 2.2 which equals 187 in Photoshop
lol, same in the northeast.I have to admit, the past several pages got me lost a few times, lol. I thought the sRGB scale in the computer was based similar to 'Perceived brightness'. A few other places seam to indicate as much, that 127.5 (127 or 128 RGB) is 0.5, is mid gray???
http://en.wikipedia.org/wiki/Gamma_correction#Power_law_for_video_display
So if I under stand the chat, 187 values of darker gray, and 68 values of lighter gray??? (When picking colors in that sRGB window thing)
I have to admit, the past several pages got me lost a few times, lol. I thought the sRGB scale in the computer was based similar to 'Perceived brightness'. A few other places seam to indicate as much, that 127.5 (127 or 128 RGB) is 0.5, is mid gray???
http://en.wikipedia.org/wiki/Gamma_correction#Power_law_for_video_display
So if I under stand the chat, 187 values of darker gray, and 68 values of lighter gray??? (When picking colors in that sRGB window thing)??? then my monitors should be set to gamma 1.0 for working in PS????
(I quote myself from another thread)
With mid gray some where other then 127.5 in a 24bit sRGB file, there are less numbers at one end or the other to represent that part of the scale. (Not everything can use 384bit floating-point color value image file, JPG, gif, bmp, etc. lol).
Monitors with gamma 2.2 cost 30.000 if you use this one then yes ;) lol
??? then my monitors should be set to gamma 1.0 for working in PS????If you not working with sRGB profile on textures you will get wrong result while rendering them later in Iray
a lot of people haven issues with displacement and normal as they created them in Photoshop not in sRGB space and later run into problems . sRGB color profile is much darker and less saturated than other color profile like for media and printing , it is recommended to have the same sRGB profile for your monitor and Photoshop
and if you edit pictures videos or other media it is another story , as we talking about texturing for PBR
(I quote myself from another thread)
With mid gray some where other then 127.5 in a 24bit sRGB file, there are less numbers at one end or the other to represent that part of the scale. (Not everything can use 384bit floating-point color value image file, JPG, gif, bmp, etc. lol).
still just playing around with the Pixar Campus HDRI, really love how it makes things turn out
You know, after experimenting with a bunch of stuff, I think I'm just going to use regular diffuse map and take a fairly basic approach to gloss and top coat and SSS. Works well enough for my purposes, and with good bump/normal/displacement can actually look rather nice.
It's not perfect, but should tide me over until 'proper' diffuse maps start coming out.
In the end, I find it basically next to impossible to tease out shadows from darker pigments, so whatever. ;)
My Iray render. Hair needs work. Any ideas on how to make it better will be appreciated!
oh, forgot the top-first post, sorry. :red:
I think I just thought of exactly why I get this whole linear vs 'logarithmic' so confusing. Tho it is also for that other gamma thread as well.
All the other lines of hobbies I have, Geology, Amateur radio, Music, etc. The real world values are somewhat logarithmic in nature. Richter scale, VEI, dbs, dbv, etc. The numbers used to represent them in a computer is linear to an extent. like, mm/year, Ejecta volume, psi, volts, sRGB files, etc.
What your calling Linear, is in fact not exactly linear.
??? then my monitors should be set to gamma 1.0 for working in PS????(I quote myself from another thread)
With mid gray some where other then 127.5 in a 24bit sRGB file, there are less numbers at one end or the other to represent that part of the scale. (Not everything can use 384bit floating-point color value image file, JPG, gif, bmp, etc. lol).
IMVVVHO We were used to the 3DL way, but in Iray the rules have changed. The main actor is light, then materials that have properties that determine how they both interact. A material is very simple, it has a Surface and a Volume and for both you have to describe what happens with light.
In order to design a material like skin you have to know certain basic properties, thickness, refractive index, but only the necessary to simulate the light interaction.
For skin, you have three main players, the base color(skin tone), traslucency(pigment/melanin) and transmitted interior color(blood concentration).
With this aproach I have been able to simulate almost all skin colors according to von Luschan's chromatic table.
I know this was discussed pages back but I am just catching up and can't quite seem to get the Finite sphere scaling right.
:(
I am using the Pixar Campus but can't seem to get it to scale down so I have ended up scaling my figures up, (Gen2), to 500%.
Also. How do you set the HDRI Environment map to coincide with what the figures consider the floor, (x, y, z 0 coordinates)?
Confused...
This image was made with original Olympia textures.
...hoping a lot of this work on skin shaders will somehow manifest itself in Iray skin resource products.
Let's go SSSurfin' - Latest study added to the gallery
Of course likes & comments there are always welcome :coolsmile:
Rendered at full size in under 2 minutes :bug:
Did a normal diffuse map, 187 Red translucence at about .5 weight, and so on.
No eyebrows because I just didn't feel like it, really, but you can pretend I was trying for Ninja Visser from Die Antwoord...
For the hair use this shader.
Uncompress on \My Library\Presets\Shaders.
HTH
Love the hair.
I know this was discussed pages back but I am just catching up and can't quite seem to get the Finite sphere scaling right.
:(
I am using the Pixar Campus but can't seem to get it to scale down so I have ended up scaling my figures up, (Gen2), to 500%.
Also. How do you set the HDRI Environment map to coincide with what the figures consider the floor, (x, y, z 0 coordinates)?
Confused...
...I found when I reduced the scaling of the sphere, I was getting a "Fisheye" like effect. Really wanted the character with the 'Steve Jobs Building" in the background but it looked like I was using DOF on the entire set even with the 8,000 x 4,000 resolution sphere.
I was joking yesterday with Alex saying , where the ladies and swords in Campus lol well here we go.. looks nice btw
the Campus sunlight is magical lol
I agree with you totally however with the one difference , THE SKIN TONE , as we all my know, pictures used for texturing have already baked in the SSS and Lighting , so when using just regular textures you double the effect ending in fire burned faces , some textures like the one from the Base G2F are actually good enough to work with without changes but mostly any others are or in wrong tones or over saturated, I found that removing the Vibrancy by -60 value set it where it should even with so yellowish skin tones like M6 base textures , after the color cast is removed from the diffuse map it is ready ..easy bread
When you build your shader you create layers of skin and the base color is the grayish surface bellow the top coat for that reason the color need to go for the maximum effect , the SSS not only create a soft look but it need o form also in wrinkles and for example eyelids edges , if his happening in front of you mean you are on the right path , all you need are base color , translucent shadow catcher and normal maps.. the rest are the right values , this is nothing new as this is the same for over a decade for a simple SSS that works
Top coat can affect the rest of the setup very easy , yesterday I made my final setup with the allowed values for skin in PBR and it seems working the best so far doing it for YOUR EYES only will works but not in all light condition , the most of the SSS values you find was created for Specular/Glossiness base mixer but half of you do it under the other , also the value sliders are working sometimes funny like for example to coat glossiness at value 1 do not produce anything, slide it to 0.98 and you get full glossy at 100% it is like in the shader in DS 3Delight value 1 is actually zero of glossy so this can be confused .
I did not tried Olympia maps yet , from your render looks like it may working also
and at the end to give some of you an idea about base color , it looks like death person without blood running in the veins lol but that is good as you can create from it all skin tones you wish for
IMVVVHO We were used to the 3DL way, but in Iray the rules have changed. The main actor is light, then materials that have properties that determine how they both interact. A material is very simple, it has a Surface and a Volume and for both you have to describe what happens with light.
In order to design a material like skin you have to know certain basic properties, thickness, refractive index, but only the necessary to simulate the light interaction.
For skin, you have three main players, the base color(skin tone), traslucency(pigment/melanin) and transmitted interior color(blood concentration).
With this aproach I have been able to simulate almost all skin colors according to von Luschan's chromatic table.
...those of us without Photoshop don't have quite the same toolset (or things are not labelled the the same between different applications). I recognise the Saturation channel but Gimp or PSP do not have anything called "Vibrance".
I use the Finite Sphere with ground for full figures and without for portrait , I scale the Finite Sphere to 9% from 100%
that is a magic number , since mostly all models I import from Zbrush or other applications that use standard scales I have to scale them down to 9% to get the same scale as genesis , so some way I was lucky to think about and it does works and produce more light than scale on 100%
With HDRI maps you get things like shadow issues, only models that are in the original center will get the same shadows direction as the rest of the HDRI map but shifting the figure on the side will already change the shadows as the sun is now just a Distance light and not more infinite light source so if you have a lot of things in the scene it may looks like each figure have different shadow directions and the angle of specular . But for one figure it is ok and nobody notice if you don't tell lol
when using the Finite with ground, you need to scale the ground textures to the level you think is ok , so higher level will bend the corner and distort the edges of the HDRI map , smaller works better, this way you get the model standing exactly on the floor you see in preview render and if you put objects like mirror it will reflect correctly the figure standing in the area of choice .
If you not use Sphere with ground all you see is floating figure with shadow catcher bellow but the rest is just illusion , now for animation it will not works as moving the camera will slide the background to somewhere else , reflective objects will not reflect accurate, you will not get accurate lighting reflecting from the floor and the figure will not blend with the backdrop as much...
I know this was discussed pages back but I am just catching up and can't quite seem to get the Finite sphere scaling right.
:(
I am using the Pixar Campus but can't seem to get it to scale down so I have ended up scaling my figures up, (Gen2), to 500%.
Also. How do you set the HDRI Environment map to coincide with what the figures consider the floor, (x, y, z 0 coordinates)?
Confused...
Gimp have plugin called ColorVibrance to do the stuff ... http://registry.gimp.org/node/107
In PSP I believe you need to use photo filter to adjust the Vibrance but not sure I am not using any of them , just searched google for the Grimp plugin ;) as it is most used tool
The Vibrance works different than saturation
Saturation
Saturation is a uniform bumping up the intensity of all colors in your shot, regardless of the starting point of the colors. This can result in clipping (over saturation of certain colors which results in loss of detail in those areas) and over saturation of skin tones leaving them looking too orange and unnatural.
Vibrance
Vibrance is a smart-tool which cleverly increases the intensity of the more muted colors and leaves the already well-saturated colors alone. It’s sort of like fill light, but for colors. Vibrance also prevents skin tones from becoming overly saturated and unnatural.
Nice but with your skin setting you used scatter without transmit on your glossy or top coat what create the hard contour edges like in cartoon shader style , changing the settings to scatter and transmit will remove the dark edges on the corners what ends with nice blending not only in skin area but also better blending with the backdrop
...thanks, I'll have to DL and install that tomorrow. almost 02:00 here.
So what controls do you use the scale the dome? All I see are Dome Scale Multiplier and Dome Radius. When I adjusted either of these sliders downward, I would a slight "fisheye" or "barrel" effect to the backdrop scene.
Only Dome Scale Multiplier to 9% then change the camera focal length it will restore the fish effect .. for landscape 27-35 for portrait 85-135
Camera has also control for automatic reduce Distortions you can set it ON OF under camera tab
...thanks, I'll have to DL and install that tomorrow. almost 02:00 here.
So what controls do you use the scale the dome? All I see are Dome Scale Multiplier and Dome Radius. When I adjusted either of these sliders downward, I would a slight "fisheye" or "barrel" effect to the backdrop scene.
How has this gone from PBR albedo textures to, basically, desaturated photos with little colour information in the highlights and too much colour info in the darks?
...thanks, I'll have to DL and install that tomorrow. almost 02:00 here.
So what controls do you use the scale the dome? All I see are Dome Scale Multiplier and Dome Radius. When I adjusted either of these sliders downward, I would a slight "fisheye" or "barrel" effect to the backdrop scene.
...thank you again. Will try that tomorrow.
Currently rendering in layers for compositing. Rendered the background first then used the Barrel Correction tool in PSP to fix the distortion and the "Wisest Sharpening" script in Gimp.to correct the blurriness. Currently rendering the character.with the same framing and size resolution. Discovered turning "Draw Dome" off still left the shadow catch plane.
It was just a test and the result was pleasant however I noticed the highlights later in some light conditions but as you saw better than the base material look
the textures are bad for PBR anyway but with simple solution it works better, the Vibrance does not desaturate it balance the darker and lighter colors to the same level, now if I used reverse layer and removed the light highlights it would be perfect skin albedo for the layer in SSS, the idea here is to remove the red channel and over cast color from the light the pictures used for texturing was shot in , the skin should have grayish yellowish tone with magenta and not red at all as that is what Translucency and SSS will produce under this layer ending in a composition of all layers into final skin tone so called "BEAUTY" and my SSS shader is made with the same technique as in Mental Ray
I try to fine easy way for people to edit the old textures as easy as possible finding compromises .. or they will have to trash them all and start over again the proper way
bellow my Eyegore character with painted albedo skin maps
I am rendering test animation for the lighting , with the proper shaders I am getting rendering one frame in 6 sec 640x360 16:9 ratio
in 2 hours and 10 min 246 frames already done with GTX 760 2GB vram , it seems like how closer I stick with the PBR values how faster it calculate , before it took 4 min to render one frame as I left some surface unconverted so it wasted time recalculating .
I wanna see in animation how the light movements affect the new SSS shader on the skin ... almost done I will upload link asap
Year it does left the shadow catcher plane on in the alpha image so easy for composition
...thank you again. Will try that tomorrow.
Currently rendering in layers for compositing. Rendered the background first then used the Barrel Correction tool in PSP to fix the distortion and the "Wisest Sharpening" script in Gimp.to correct the blurriness. Currently rendering the character.with the same framing and size resolution. Discovered turning "Draw Dome" off still left the shadow catch plane.
It definitely desaturates in a Halfway House kind of way. You can't have much colour info in the brights in a low dynamic range image without selectively targetting those areas and adding saturation, in which case they're no longer brights ;)
Your results are really good, by the way, and obviously work in a generalist way using non-PBR textures. I'd certainly never detract from that and well done.