Ah thank you, jumping from instance to instance of Nuke I must have left it off in one of them.
Thanks again frank ! Date: Tue, 9 Oct 2012 12:26:08 +1300 From: [email protected] To: [email protected] Subject: Re: [Nuke-users] OCIO logLin and Middle Grey I'm guessing Tahl doesn't have the spi-vfx config set in the project's lut settings?! If that is set to nuke-default you get the values he posted. On 10/9/12 12:11 PM, Schneider, Abraham wrote: Doing exactly what you propose, I'll get a perfect 0.18 for the 445 swatch when using the OCIO LogConvert node and 0.17965 when using OCIO colorspace with in set to 'lg10' and out set to 'lnf'. Same behaviour as the normal (not OCIO) Ploglin node. Maybe you have something setup wrong? Here my testscript. The constant is the picked color of the Marcie 445 swatch: #! /Applications/Nuke6.3v8/NukeX6.3v8.app/../Nuke6.3v8.app/Contents/MacOS/Nuke6.3v8 -nx Root { inputs 0 name /Users/abraham/Desktop/ocio_test.nk first_frame 0 lock_range true format "2048 1556 0 0 2048 1556 1 2K_Super_35(full-ap)" proxy_type scale proxy_format "1024 778 0 0 1024 778 1 1K_Super_35(full-ap)" luts {linear {} sRGB {} rec709 {} Cineon {} Gamma1.8 {} Gamma2.2 {} Panalog {} REDLog {} ViperLog {} AlexaV3LogC {} PLogLin {} SLog {} REDSpace {} AlexaV2LogC {"((t > 0.1312920 ? (pow(10.0, (t - 0.3910068) / 0.2471896) + 0.6229143) / 182.20240 : (t + 0.5651548) / 178.2469) - 0.003907204) * 36 "}} defaultViewerLUT "OCIO LUTs" OCIO_config spi-vfx } Constant { inputs 0 channels rgb color {0.4349952638 0.4349952638 0.4349952638 0} name Constant1 xpos 173 ypos -142 } set N16427a60 [stack 0] OCIOColorSpace { in_colorspace {{4} ln/lnf ln/lnh ln/ln16 lg/lg16 lg/lg10 lg/lgf gn/gn10 vd/vd16 vd/vd10 vd/vd8 hd/hd10 dt/dt16 cp/cpf nc/nc8 nc/nc10 nc/nc16 nc/ncf srgb/srgb8 p3dci/p3dci8} name OCIOColorSpace1 xpos 173 ypos -11 } push $N16427a60 PLogLin { name PLogLin1 xpos 293 ypos -9 } push $N16427a60 OCIOLogConvert { name OCIOLogConvert1 xpos 55 ypos -8 } Viewer { frame_increment 8 viewerProcess "Raw (sRGB)" input_process false name Viewer1 xpos 179 ypos 58 } Read { inputs 0 file "/Users/abraham/Dropbox/Vortrag Nuernberg/marcie_log.dpx" format "1024 778 0 0 1024 778 1 1K_Super_35(full-ap)" origset true colorspace linear raw true timecode 00:00:00:00 name Read1 xpos -20 ypos -152 } Am 09.10.2012 um 00:47 schrieb Tahl Niran: Hoping someone who has some experience with OCIO, espeically at Sony, can clarify this for me. I have been taking a look at the Sony OCIO docs and it all seems pretty clear, I notice that in spi-vfx they are equating 445 (lg10 value) with 0.18 (inf, inh value) which is great and I like. What I want to know is, are Sony pre-treating any of the the Film Log files (for example lg10) or using some specific scanning correction not common to the rest of us ? If I just take a nominal dpx log file for example, Marcie. Read that in as Raw Data, then apply an OCIO Log Convert. Then her 445 swatch is not returning a..... 18 more like .1493. Same if I use an OCIO to Colorspace or have I missed something ? Abraham Schneider Head of VFX pipeline / VFX Supervisor ARRI Film & TV Services GmbH Tuerkenstr. 89 D-80799 Muenchen / Germany Phone +49 89 3809-1269 EMail [email protected] www.arri.de/filmtv Click here to visit us on Facebook! ----------------------- Tahl Niran www.tahlniran.com _______________________________________________ Nuke-users mailing list [email protected], http://forums.thefoundry.co.uk/ http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users ARRI Film & TV Services GmbH Sitz: München Registergericht: Amtsgericht München Handelsregisternummer: HRB 69396 Geschäftsführer: Franz Kraus, Dr. Martin Prillmann, Josef Reidinger _______________________________________________ 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
_______________________________________________ Nuke-users mailing list [email protected], http://forums.thefoundry.co.uk/ http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
