There’s nothing to say. Kelvin temperature colors, as well as vray sky and sun should be gamma correctable…
In this thread i posted two pics that use the ‘LWF’ = Lele’s Work Flow 
http://www.chaosgroup.com/forums/vbulletin/showthread.php?t=38783
I didn’t have to correct anything in post, not even the background, these are straight renders with Sun and Sky multiplier on 1.0.
To me it looks like the sky isn’t washed out and is properly contributing to the scene-lighting and reflections.
Can you post an example of a picture where you end up with a washed out sky, together with your settings?
Do the same, but with a vray sky this time.
I’ll run a test when i have some more time, working on a tight deadline atm.
But the link i posted to the thread with two images i did earlier shows clearly there is no mismatch between sky and lighting, nor a washed out look of either the model or the background, which was the complaint in your first post.
There’s a difference between your workflow and mine though. You’re gamma-correcting your textures with 2.2 (or a multiplier of .454), while i am darkening them even more, rgb-multiplier of around .255 for most textures, as per Lele’s measurements with VraySky and VraySun. And, when using a VraySky, your background should be colormapped just like the rest of your scene, so you need to check ‘Affect background’.
No corrections, straight out of the box, this is what you payed for, this is what you normally would get. If you had never came across that gamma tutorial, or Lele’s method, or Chris Nichols dvds, you would have been under the impression that this was correct:
However, you started playing with gamma, lut, .255 rgb multipliers (this is something that i totally don’t understand), and now you believe that this is what your skies are supposed to look like.
Hehe, good thing those tuts are around
I get your point though. But if with using a certain workflow and settings you still need to change the background for every render, i’d rethink my strategy. In the end it’s all about what looks right to the artist/clients, it’s no absolute science. It doesn’t need to be physically accurate if it looks photoreal.
Baseline: for me it gives good and quick results, for almost every type of scene. Here’s an quick and early WIP straight from the framebuffer with same workflow, i don’t think the sky would look better with even more color in it.
Greetings, Ruud
Linear Converter
pailhead, if you simply want an easy way to “correct” the color of those lights, try the free application floating (no pun) around called “linear converter” I think from Westlund Studios?) It doesn’t get down to decimals, and for colors very close to black or pure white, it might not help, but it’s really handy I’ve found for correction colors with no plug-ins (i.e. CC or Vraycolor.) By the way, i forgot who uploaded it, but thanks a million! If I can, I’ve been trying to use it for things like paint colors that get washed out. These are just my opinions and I know this “LWF” can be a very touchy subject, so please don’t flame me guys 'n gals, just trying to add a helpful suggestion.
But this is very very subjective. I only took a glance at the .255 theory (can’t open the original thread) but i for one, never had a problem with my skies being too dark. If anything, they always seem overexposed IMO.
The bottom line is, it seem sto me that i am given an alternative by being able to choose any gamma while loading an image, plus having vray color with a gamma parameter… not to mention the color correct plugin.
The only things that i can’t color correct are vray sky, vray sun, and now IES colors (be it in kelvins, or actual RGB color).
@voltron7,
Can you upload the app again?
link
Here’s the original post, sorry, I don’t know if I may re-upload it without permission, etc.
@voltron7: Good to know people are still using my little app. Making it output floating point values and decimals is still on my list of things to do, but it’s not a high priority as we don’t have a use for that feature internally. I’ve got to do the work we get paid for first ;).
Vlado? …
Yes…?
Best regards,
Vlado
Well, can you read my first post and give an opinion? When working with kelvins, the results tend to differ greatly.
In general, light colors should not be gamma corrected. This is because they are derived from physical observations or formulas - they do not start as user-input colors. The IES Kelvin color is derived from Plank’s formula for the spectrum of black body radiation. The color of the sun is based on a table of the observed and measured illuminance spectrum of the sun. There is no gamma involved anywhere in this process.
Note however, that when using the IES lights or the sun/sky, it is recommended to use the VRayPhysicalCamera with the white balance set to D65 standard CIE illuminant - otherwise, the colors might tend to be too much on the blueish side.
Best regards,
Vlado
But how am i supposed to use them in a gamma corrected workflow?
These are obviously different temperatures, one is much colder than the other, or one has more kelvins than the other.
But they are the same light, with the same amount of kelvins. This is where i get confused.
Note: The issue would be obvious with warm colors as well…
Hmm, maybe I don’t understand something correctly here. The Kelvin temperature defines what light spectrum is emitted by a light source, not how warm or cold this light looks to you. It is an absolute measure rather than a subjective one. On the other hand, the “warmth” or “coldness” of colors is purely a subjective term without a particular physical meaning. Green is a warm color, but there is no blackbody spectrum for it (there is no Kelvin value that will produce green).
If you have a bulb that emits light at approximately 3200 Kelvin spectrum, you can take many different photos of this light, with different white balance settings - each of them will produce an image with different color tint, but that does not mean that the light source has somehow changed.
Best regards,
Vlado
I was under the impression that all of the professionals who have anything to do with Kelvins refer to light as either warm or cold. Lower temperatures are referred to as warm, (red/orange/yellow) while higher temperatures are referred to as cold (blue).
Let me ask you this:
I notice that the color switcher changes depending on how many kelvins i choose. The color seen in the switcher corresponds to what the IES emits in a gamma uncorrected render. So, when i choose 8000K i get a blue, cold color. If i tick of the temperature parameter, i can actually read the color in RGB.
The bottom line is, gamma/LUT settings screw up everything, but while i can bring everything back by applying inverse gamma, vray sky and IES are out of reach.
How does this relate to the blackbody theory?
Take a look at the first to images in my post above (ones without GI), if i were to put a yellow filter on my lense in real life, the deep blue would shift towards neutral, but the neutral one would actually shift to a warm color.
But i am not using any white balance settings. Even if i did, the problem would remain. While a white balance setting would shift my tint, gamma settings would still wash it out.
Yep, washing out is what gamma does - under the presumption that your monitor is calibrated and will revert this back to the original image when it is displayed.
Best regards,
Vlado
From my understanding most photograph are taken with setting that is asthetically pleasing
to the eye. Setting that emulate moods or add to the atmosphere. Every little tweeking and
even darkroom post processing affects an image output. In science, lights color such as
8000K may not look right when rendered off a machine becos the human eye is a very advance
equipment it corrects color. For all you know, every image looks slidely different to everyone,
every sunlite environment as well. So there is an absolute number in science - in some cases
non, or at least we can draw graphs to read behaviors -, but in art there is non.
What looks right is right afirms this believe. In art we break some rules of nature making things
looks better than they really are. Or to emulate a mood.
I would say the blueish 8000k is the correct scientific render of the actual scientific colors.
However, a human standing in that same space using his eyes (not with a camera) would
see a less blueish environment. Hope it explains that nature of our biological part are really
advance.
More to colors and how it appear to be there are many factors, your monitor calibration,
your immediate environment (seating near a window) affects colors. Viewing angle of the
screen … . Some graphics studio have their environment made to view images with D50
controlled light environment, then you are set to talk about correct color!