Wert, I’ll use your posts as tags on the next dvd I’ll release, lol :lol:
I have this all working now and the results are excellent - thanks Lele, much appreciated that you have taken the time to inform the rest of us.
N
i don’t get it ![]()
…all these tried and tested workarounds by users should be collected together in one place…easy reference for peeps… ![]()
Where can i find some info on the advantages of working in actual linear space i.e. applying the correction after you render?
Now with the don’t affect color thing in the color mapping it gets all the more confusing. If vray is seeing everything with 2.2, if you are inputing everything with 2.2, if you correct your colors by 2.2, why do you output as 1?
So far I only see the gamma as another method of exposure. The only thing that i have learned is that i can make my maps really intense (by applying the inverse gamma) and thats how i can compensate for the washed out look that you would normally get if you fooled around with gamma…
Well, linear workflow doesn’t really have that much to do with exposure. It’s actually a way to counter the way computer monitors work. Most monitors use a gamma of 2.2 (although it varies wildly) and so most pictures you use as textures will have been gamma corrected for the monitor. in other words every single texture you have (unless you intentionally made it linear gamma in the first place) will give the appearance of correctness on your monitor while being wrong. It can create some funky results since Max assumes pictures are properly made by default. When you set the Max gamma correction to input 2.2 you undo the gamma correction that were done to the images to make them look good on the monitor. It tells max “these images were gamma corrected so treat them as such”. The display gamma setting tells Max to display all the images with 2.2 gamma applied (even though it still calculates them without) so that they still look the same to you. The gamma correction in VRay is used to make the final render calibrated for the monitor again by changing the gamma back to 2.2.
The advantages of working with linear gamma in Max is that you get less artifacts (as you generally do across the board when using it) and you get more realistic results with GI and so on. whether you render out to gamma 2.2 or not usually depends on how you intend to use the image. if you intend to edit the render in another application that supports linear workflow then that is preferable. If you intend to just upload it directly to the net, then you have to use gamma 2.2.
You’re probably even more confused now… I confuse myself sometimes. ![]()
you output to 1 because you’ve already corrected everything on the way in to 2.2, so correcting on the way out would just double the correction.
Actually the exact opposite. Everything up until the render has been to remove the 2.2 gamma correction from the textures used. If you were to output as linear gamma 1.0, save it as a jpg, and send it to someone as is they’d get a picture that’s much darker then it’s supposed to be. Although you won’t notice if you use the Max frame buffer because it applies a 2.2 gamma correction to the display for preview purposes. The VRay frame buffer displays the images how they actually will be saved (although you can apply a temporary gamma correction to them by clicking the sRGB button, but that won’t affect the saved images)
Most “final” images that won’t be used by other professionals (who knows how to use color profiles) will usually need to be gamma corrected to 2.2 so they are bright enough. but this should be the last step i the process. after having done all the photoshopping and so on.
If I’m wrong about any of this please speak up Lele! ![]()
so the LWF method is just a way of temporarily correcting the gamma on textures while setting up materials in the mat editor?. Guess i still dont get it because everything looks too washed out in the material editor which actually makes it harder for me to set up materials as I would like. Surely if vray is calculating using the washed out 2.2 gamma corrected textures, the end result will be less accurate?. The majority of LWF images i’ve seen have this look to them…flat and washed out, so I cant see the advantage. I think i’ll just wait till this method is integrated as an option in vray (if that would happen?)… click the LWF button and all confusion is magically ended ![]()
at the moment it just seems non-linear to change gamma, render, and then change it back… if thats the idea? (dont mean to be rude, i’m just confused)
In a sense that’s true, but the reason to do it is because the images you usually deal with on the computer are wrong even though they look right on your monitor (this means pretty much all jpgs you find on the net among others).
VRay doesn’t calculate using the “washed out” images in the material editor. That’s the whole point. The previews there has temporary gamma correction added to them by Max without altering the actual materials. Only the thumbnails are corrected for your viewing pleasure.
Anyway it’s important to note that to use LWF you HAVE to calibrate your monitor beforehand. If your monitor isn’t set to gamma 2.2 as well, then things won’t look right. You can go here to do that: Accurate image manipulation-Top Wien Hotels buchen.
Note that LCD monitors often don’t follow the sRGB standard so they might be hard, or even impossible, to calibrate to 2.2. in that case it might be easier to calibrate them to 1.0 (But then you couldn’t use this LWF method as is.)
It does not seem likely this will ever be built into VRay in such a way since you still have to use the built in Max features to preview everything in your monitors gamma space. (Remember, What you see is not the same as what VRay/Max sees.) But who knows…
Not really, your explanation is pretty good, i wish it was around a while back when this whole mess started. I’m still confused about certain things, maybe it’s the terminology. If i were to look at an image in linear space, without any correction on it, it would most likely show as very dark on my screen. If i apply correction to it, for example the 2.2 gamma, i’m bringing it from linear space to sRGB space… correct?
Now what i don’t understand, what is the advantage of working in linear space, as opposed to the gamma corrected space? In the end, we’ll be looking at it in sRGB space anyway, what use is the dark linear, uncorrected one?
Am I confusing some terms here?
Would that mean that they’re already in linear space?
Correct.
The monitors, yes. Although you always have to calibrate them a little because the lighting in the room you’re in affects your perception of it as well.
It sounds silly, I know. But the problem is the pictures are still calculated as if they were linear space. That can create some big problems. Think of it this way: Make a wall with a brick texture without working in LWF, set up a Physical camera and Vray sunlight and render it with GI. What happens? The brick texture is more then twice as bright as it is supposed to be. You’d then change the exposure of the physical camera to compensate and it’d look fine, but VRay would still see the texture the same way. Any Global Illumination reflected off the wall would be twice as bright as they were supposed to. In some cases making almost a caustic effect. It would create bad effects in other ways as well. And it’s not limited to renderers. you can go here to see lots of examples of artifacts.