Nuke needs to unpremultiply the alpha before applying the sRGB lut and the premultiply it again. Did you check the "premultiplied" box in the read node properties next to the lut selection?
-deke On Tuesday, April 29, 2014, Florian Einfalt < [email protected]<javascript:_e(%7B%7D,'cvml','[email protected]');>> wrote: > I have tried to use: > png 8 and 16 bit > exr 16bit float > tiff 8 and 16bit > > Same behavior with all of them. > > > On 29 April 2014 12:20, Martin Constable <[email protected]> wrote: > > An obvios suggestion maybe, but have you thought about moving your > workflow to tiffs? Or exrs? > > PNGs do not store info as floating point. They use integers instead. This > might be at the root of yr problem. > > > On 29 Apr, 2014, at 6:56 PM, Florian Einfalt < > [email protected]> wrote: > > > I tried both using 8bit per pixel and 16bit per pixel. No difference > unfortunately. > > > > Flo > > > > > > On 29 April 2014 11:51, Martin Constable <[email protected]> wrote: > > Are yr pngs in 24 bit? That might help. > > > > > > > > On 29 Apr, 2014, at 6:45 PM, Florian Einfalt < > [email protected]> wrote: > > > > > Hello, > > > > > > I am having this problem with PNGs from Nuke that I need to deliver in > a Photoshop psd-file. This specifically concerns the shadow pass of my comp > but generally there is a discrepancy on everything that is not solidly > opaque. > > > > > > My workflow is as follows: > > > - I write my 8-bit PNG out of Nuke using the sRGB colorspace. > > > - When I open the rendered PNG in Photoshop I assign the sRGB color > profile and over a 'transparent' background the shadow looks to have > similar density to the one in Nuke. > > > - When I comp that layer over a white background though, the shadow is > suddenly more dense than in Nuke. Also, there is a much harsher fall-off in > the shadow. > > > > > > Is this because Photoshop treats alpha information differently from > Nuke? > > > > > > I thought I can maybe counteract this by additional grading in Nuke > but even when I grade the alpha gamma to 0.4545 in Nuke and then write the > PNG I get closer but not exactly to the same result. > > > > > > Does anyone have a suggestion? > > > > > > Thanks. > > > > > > Florian > > > > > > -- > > > > > > > > > Florian Einfalt Senior Digital Artist > > > T +44 (0)20 7833 3032 M +44 (0)7808669580 > > > [email protected] > > > > > > Saddington Baynes Ltd | Studio 3, 21 Wren Street. London WC1X 0HF. > United Kingdom | Map > > > Disclaimer | Printing? Consider the environment. > > > > > > _______________________________________________ > > > Nuke-users mailing list > > > [email protected], http://forums.thefoundry.co.uk/ > > > http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users > > > > _______________________________________________ > > Nuke-users mailing list > > [email protected], http://forums.thefoundry.co.uk/ > > http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users > > > > > > > > -- > > > > > > Florian Einfalt Senior Digital Artist > > T +44 (0)20 7833 3032 M +44 (0)7808669580 > > [email protected] > > > > Saddington Baynes Ltd | Studio 3, 21 Wren Street. London WC1X 0HF. > United Kingdom | Map > > Disclaimer | Printing? Co > > > > Florian Einfalt Senior Digital Artist > T +44 (0)20 7833 3032 M +44 (0)7808669580 > [email protected] > [image: Saddington & Baynes] <http://www.saddingtonbaynes.com/> > Saddington Baynes Ltd | Studio 3, 21 Wren Street. London WC1X 0HF. United > Kingdom | Map <http://signature.saddingtonbaynes.com/map.php> > Disclaimer <http://signature.saddingtonbaynes.com/disclaimer.php> | > Printing? Consider the environment. > [image: Twitter] <http://www.twitter.com/saddbaynes> [image: > Facebook]<http://www.facebook.com/SaddingtonBaynes> > [image: linkedin] <http://www.linkedin.com/company/saddington-baynes> [image: > pinterest] <http://pinterest.com/saddbaynes/> [image: > vimeo]<http://vimeo.com/saddingtonbaynes> > [image: behance] <http://www.behance.net/saddington_baynes> [image: > instagram] <http://instagram.com/saddingtonbaynes> > -- -- Deke Kincaid Creative Specialist The Foundry Skype: dekekincaid Tel: (310) 399 4555 - Mobile: (310) 883 4313 Web: www.thefoundry.co.uk Email: [email protected]
_______________________________________________ Nuke-users mailing list [email protected], http://forums.thefoundry.co.uk/ http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
