How to get better VRay renders (aka working in linear space)

[gilicom]

Hello throb , what your basicly saying is that this methood can only be used to see how your final render will look like.
since you can't render vray vfb through backburner 'you will render it in the old way (no gamma correction) and then fix it in ae or df.
am i right?

[throb]

Hello throb , what your basicly saying is that this methood can only be used to see how your final render will look like.
since you can't render vray vfb through backburner 'you will render it in the old way (no gamma correction) and then fix it in ae or df.
am i right?

the fix is to adjust your viewing system (and that's in the compositing package) so that the viewer does an inverse sRGB curve to your image data.

the fix is not to put a 2.2 gamma once the image is loaded.

why?

because then you are not working in linear space. you are back to correcting an image in a non-linear method. your viewing system should be correcting for this. correcting the data before you color correct, etc will cause things to not work as well as they should.

[gilicom]

the fix is to adjust your viewing system (and that's in the compositing package) so that the viewer does an inverse sRGB curve to your image data.

Lets say i use ae , i've searched over ther and couldn't find any srgb related topics , can you direct me how do i do that?
does premire also support this methood?

[throb]

the fix is to adjust your viewing system (and that's in the compositing package) so that the viewer does an inverse sRGB curve to your image data.

Lets say i use ae , i've searched over ther and couldn't find any srgb related topics , can you direct me how do i do that?
does premire also support this methood?

don't use AE so i have no idea.
also don't know about premiere.
sorry

[gilicom]

Ok , so basicly i need to get a software the does support srgb before i start working like that ?

does combustion support srgb?

you are saying that i should work in a linear space in max and vray, then i should render it through backburner (non linear) and the i should import it to a package that will fix the render footage in a way that i will see what i've worked on max and vray (did i understand it this time?)
gili

[throb]

Ok , so basicly i need to get a software the does support srgb before i start working like that ?

does combustion support srgb?

you are saying that i should work in a linear space in max and vray, then i should render it through backburner (non linear) and the i should import it to a package that will fix the render footage in a way that i will see what i've worked on max and vray (did i understand it this time?)
gili

you are 100% correct!!!
combustion does support viewer correction so you can work this way.
other apps that do are
nuke (my package of choice)
shake
digital fusion

[gilicom]

thank yuo for all the help
who makes nuke?

[flipside]

Cool, I finally understand it too now!

[throb]

thank yuo for all the help
who makes nuke?

http://www.d2software.com

[themaxxer]

I've made some tests in linear space. But I noticed some strange things. When you open the image you made in photoshop, then the black point is only moved a bit to the white point, so you lost contrast.



but that's not what it should be, best would be a linear curve, or am i wrong?



so you have to correct this again in photoshop and you get the same result as you get in non linear workspace...

best regards
themaxxer

[sbravo]

This is great info - I went through the process and got the results as expected - actually a bit brighter than expected but nice. Anyway while in the v-ray image buffer I clicked the Copy to Max FB button and the copy that came up came out very bright, too bright. Now if I turn off the "Use color curve correction" in the v-ray FB (goes back to darker image) and click the Duplicate to Max FB the image that comes up in the max FB is brighter than the Vray FB but lighter than the Vray w/ curves adjustement on. It looks good - nice amount of contrast but still the clarity of the brightened image - whats going on here?

[throb]

This is great info - I went through the process and got the results as expected - actually a bit brighter than expected but nice. Anyway while in the v-ray image buffer I clicked the Copy to Max FB button and the copy that came up came out very bright, too bright. Now if I turn off the "Use color curve correction" in the v-ray FB (goes back to darker image) and click the Duplicate to Max FB the image that comes up in the max FB is brighter than the Vray FB but lighter than the Vray w/ curves adjustement on. It looks good - nice amount of contrast but still the clarity of the brightened image - whats going on here?

Well, vray's vfb does not listen to the max gamma setting, which is why i change the curve in the Vvfb.
my curves are not exactly the same as a 2.2 gamma which is why you're seeing a difference.

[rpc212]

So when you save the image from the VFB it must be taken into Combustion or another compositing program as either a HDRI or RPF image. Then when you take that image and save it as a tiff or whatever and then into Photoshop, Photoshop asks if you want to apply a color profile because the incoming file does not have a color profile embedded. What should I answer to this? If I assign sRGB am I applying a color curve twice to the image?

Thanks,

rpc212

[throb]

So when you save the image from the VFB it must be taken into Combustion or another compositing program as either a HDRI or RPF image. Then when you take that image and save it as a tiff or whatever and then into Photoshop, Photoshop asks if you want to apply a color profile because the incoming file does not have a color profile embedded. What should I answer to this? If I assign sRGB am I applying a color curve twice to the image?

Thanks,

rpc212

well, your software should apply the 2.2 at some point.
I know that in nuke, if i write out a .jpg or .png or .rgb it bakes the 2.2 into the image so that i know i have a gamma corrected 8 bit image.
if i use a floating point format (and i use .exr for my float) it puts the 2.2 in the viewer so that i am working in linear float.
the 2.2 is applied at the very end by you (the user) so that your end result (the very end) looks correct.
the viewing system puts a 2.2 on there so that when you do a multiply, you get a correct multiplication.

--however--
if you take your stuff into photoshop and put a 2.2 gamma on it and then do some fixes and save it out, that is totally cool.

[Trapezium]

rpc212:

You should look into Photoshop CS2. You can use either the Exposure Adjustment or you'll get a dialogue when you convert it to 8bit for print.

--Jon