How to introudce phisical values on Corona Color picker

[n2graf]

Hey comunity

What is the correct PBR way to introduce 4% of diffuse reflectivity of a real material, with corona color picker?


In lineal space, If I put 4% on Value bar, the Red Green and Blue bars turns to 0,1 beacuse the gamma. So i have to put 4% at RGB bars? Wich of two is the correct?

thanks!

[romullus]

I think you're confusing things slightly. If you want to introduce 4% reflectivity (at 0° angle), you just need to set material's reflection level to 1,0 and its fresnel IOR to about 1,5 (default). Corona will take care of the proper balance between diffuse and specular reflection. There's no need for any guesswork with colour picker.

[n2graf]

its not like this. The darker diffuse reflection in real world is 4% and the brighter is 86%, as you can see here

https://en.wikipedia.org/wiki/Albedo

So we have to put this value at the color of diffuse.

[romullus]

Are you saying that dark materials reflects 4% light as diffuse component and 96% as specular and bright materials reflects 86% diffuse and 14% specular? Hmm, doesn't sound right, don't you think? As i said before, Corona, like any other physically based renderer, does proper energy conservation and you don't need to subtract or add anything. How much light will be reflected as diffuse and how much as specular, is controlled by reflection level and by fresnel IOR.

[n2graf]

No im not saying that. I'm saying what WIki says. Forget the specular. Focus on difuse reflexions. The darker material reflects 4% of difuse. How can i intorduce this value of difusse in corona?

[niljut]

Just keep diffuse value between 41-230 for non-linear, like Corona Color Picker, and 5-203 for linear, like the standard 3ds Max color dialog. That's pretty much it. If you want 0.8, like for snow, you would have to input 0.8 for the RGB channels with sRGB unchecked, which gives 0.904 in value, since value does not currently respond to unchecking sRGB. Alternatively you can also use a CoronaColor node and put the value in Solid HDR Color values, with "Input values are in linear space" checked.

[n2graf]

Yes man I was thinkink that is the correct way, but 41 it's a very bright gray that it not seems to be a black material on a render exposed correctly. In a sunny day on exterior  with ISO 100, f8, and 1/200s i see so much darker materials than with the same values in a render with corona cam and corona sky /corona sun. 

its for this that i t seems that something is wrong understanding the sRGB gamma input color value

[Jpjapers]

Yes man I was thinkink that is the correct way, but 41 it's a very bright gray that it not seems to be a black material on a render exposed correctly. In a sunny day on exterior  with ISO 100, f8, and 1/200s i see so much darker materials than with the same values in a render with corona cam and corona sky /corona sun. 

its for this that i t seems that something is wrong understanding the sRGB gamma input color value

Can you show an example render of what the problem is?

[n2graf]

Sorry i read so fast. Wy 41-230 dor non linear and 5-203 for linear in the value? Chenking or uncheking sRGB the number of the value dont change it. So what do you refer?
4% of 256 its 10, and if I put 10 on the RGB bars on linear, the value goes to 59, that it seems to be much birghter to be the darkest diffuse posible on the reality, don't you think?

(in a while i post the image)

[n2graf]

the image

[n2graf]

this is the render with f8 ISO100 and 1/200s and as you can see the 4% black its so bright. I've seen much darkers materials on real life at the same conditions of daylight. So what is wrong?

[Jpjapers]

this is the render with f8 ISO100 and 1/200s and as you can see the 4% black its so bright. I've seen much darkers materials on real life at the same conditions of daylight. So what is wrong?

Can you show an example of what you're expecting to see?
A black colour in direct full intensity sunlight will absolutely be brighter than what it shows in the colour picker.

Theres some old discussion with regards to colour in this thread if youre interested
https://forum.corona-renderer.com/index.php?topic=13398.0

[Juraj]

but 41 it's a very bright gray

Linear view is not supposed to represent how things look. To see that the same happens in reality, you only need to linearize raw photo. In CapturePro you select "linear profile", in Lightroom (or ACR in PS/Bridge) you need to select older process and zero-out all the values (also in curves!), because by default Adobe already beautifies the image (similar to profile Camera applies if you save directly to jpeg)

Here is example on my photo. So the rough guide of PBR to use 50-235 +/- for most material fits. That's their physical albedo, not how you see them with your eyes or how they would end up on photo with any profile applied.
Of course exemptions exist, I am pretty sure Vantam Black is around sRGB <20.

So if your black materials look grey, you just don't have enough contrast applied ;- ).

[n2graf]

Juraj Talcik: But 41 its so brighter for the render, too, as you can see at the render i have post. And we all see the render in sRGB not linear.

Jpjapers: Here I leave an image with a much darker black than the render. There are a lot on real live, its full of darker objects, my black pullover looks darker on a sunny day.

[Jpjapers]

Juraj Talcik: But 41 its so brighter for the render, too, as you can see at the render i have post. And we all see the render in sRGB not linear.

Jpjapers: Here I leave an image with a much darker black than the render. There are a lot on real live, its full of darker objects, my black pullover looks darker on a sunny day.

I see what you mean now. As Juraj said cameras do their own in-camera tonemapping and what your eye sees vs what a photo shows, are two completely different things.
The colour is correct for the light setup and material settings you've used.