how to apply a gamma correction to an image ?

[MikeeMax]

Hi, i'm +
trying to cross the gap to the LWF thing.
i'm reading everywhere that in some cases, whatever method you choose to use, you have at one point or another, to apply gamma correction, whether you need to actually correct the image, or uncorrrect it/
i was wondering what is the method to apply this correction
i have read that sRGB is a 2.2 profile. so, to correct an image (let's say i want my images to have a 2.2 gamma in them), do i simply need in photoshop to convert the image to sRGB profile ? or do i need to discard all profiles and apply a manual gamma/exposure correction ?
also, in phnotoshop, there are two options for profiles. in french it's called "convertir le profil" or "attribuer un profil". i think you can see the english correspondance. which one to choose ? (in case this is a method to gamma correct an image of course)
i really want this lwf to work, but this is such a mess with all those redundant options in all softwares, even in the same sotware, settings can override each other it seems(i'm thinking of gamma lut in 3ds max, crossing with color mapping in vray for example)....i'm a bit lost for the moment, but i have good hope.

Thanks
for info i have an lcd monitor, so i'll try to work in linear space, but i don't know yet how to calibrate my monitor to gamma 1.0... and most of the images i work with have an embedded sRGB profile... so i think i need to "uncorrect" them before using them.

sorry if the post is difficult to understand... it only reflects that all this is a bit hard to understand for me at the moment.

Thanks for reading

[plastic_]

hi,

first i'd suggest to stick with gamma 2.2 for everything.

if you store bitmaps from max with gamma 2.2, there is no problem with photoshop or other apps.

if you are using a color calibration device, as spyder, then it gets more complicated, because there will be a shift from monitor color to srgb colors (default for windows).

so if you open a max bitmap with photoshop, it will assume srgb, but this is wrong, because it should rather use the monitor profile.

i compensate this by using "attatch profile" ("attribuer un profil" in your case), to manually assign the monitor profile to the bitmap. that way it will look the same as it looked in max, and still will look the same way with color managed applications.

the "convert profile" option in photoshop tries to preserve the color apperiance between color spaces, it can be used to convert your monitor calibrated image to srgb.
(this isn't really needed, because srgb is the default for the web, and for windows, so a file with embedded monitor profile will look fine in srgb space, and even better in color managed apps)

all in one, color management is a pain in the ass

[dlparisi]

i compensate this by using "attatch profile" ("attribuer un profil" in your case), to manually assign the monitor profile to the bitmap. that way it will look the same as it looked in max, and still will look the same way with color managed applications.

the "convert profile" option in photoshop tries to preserve the color apperiance between color spaces, it can be used to convert your monitor calibrated image to srgb. (this isn't really needed, because srgb is the default for the web, and for windows, so a file with embedded monitor profile will look fine in srgb space, and even better in color managed apps)

When opening renders out of Max I too usually assign the monitor profile ("attribuer un profil") but I then convert it to sRGB just to avoid any problems with any other apps. This seems to get the photoshop image to match the vray fb exactly and also gets the image into a standard colorspace (although you're probalby right about it not strictly being necessary). If I have any multimatte or z depth channels I leave those alone and do not assign or convert any colorspace since they're just information channels and doing so would change the color values. I'm not 100% sure this workflow is correct but it's been working great for me for the past couple of years.

BTW: My monitor is calibrated to 2.2 with the Gretag Macbeth Eye-One.

all in one, color management is a pain in the ass

Agreed.

[MikeeMax]

Thanks very much for the input.
ok, I understood the "after render" workflow and the attach profile thing .
Now "pre render' workflow : what is it of the maps you use in your renders concerning gamma lut input/output in max(2.2 or 1 ?... or 1.8 ?), color mapping mode (gamma value), and pre correction of maps (pre-uncorrect images, pre-2.2 them?) ?
I have read from vlado post in 'tips forum' that with lcd, it is preferred to stick to the 1.0 linear workflow with no corrections at all.
ok...forget it. Now i'm really lost.

I'll come back in some days when I have digested all those infos..

edit: I agree about the pain in the ass.