Creative Communities of the World Forums

The peer to peer support community for media production professionals.

Activity Forums DaVinci Resolve Grading for P3 Color Space… on a budget

  • Grading for P3 Color Space… on a budget

    Posted by Chris Hall on March 14, 2011 at 10:30 pm

    I know this topic has come up in the forum a few times in the past, but I thought I’d rekindle the fires and see where everyone’s opinions lie. So I’ve been grading a number of projects over the past few years that have all been Rec709 bound (primarily broacast commercial spots, music videos, and Indie/Low Budget Films bound for HDCAM/SR mastering only), I haven’t had to deal with any DCP Package deliverables or P3 color space projects… yet…

    Recently a number of my clients have been asking me to grade P3 bound projects (specifically a few movie trailers and indie films). I’m trying to wrap my head around the most cost effective way to do this as a freelancer that works out of my own personal grading suite (Flanders Scientific 2460 Grading Monitor, and Panasonic G25 Plasma that has been calibrated to Rec709 for a client monitor).

    The rumor I’ve been hearing from my clients is that the larger post houses they have been recieving “low-bid” quotes from, is that the post houses are reccomending a REC709 color correction to save money, and then applying a LUT to the entire project to transfer it into the P3 color space for DCP deliverables. This sounded a bit fishy to me (why would you grade in Rec709 if you’re ultimately P3 bound… only answer is cost), but a number of clients have told me a number of LA based houses quoted them on this…. so that brings me to my question:

    I understand that the obvious/proper way to grade for P3 is to grade in a calibrated suite with a projector that can handle the entire color space and has been properly calibrated (and then doing a seperate Rec709 pass for HD deliverables). BUT, are people getting acceptable results out there by grading for Rec709 and then just using a translation LUT to transfer everything inside the P3 space afterwards(I realize that this wouldn’t take full advantage of the full gamma and gamut of P3, please don’t lecture me on the negative benefits of this, I’m well aware, I’m just asking is this a common practice… because it appears that it may be, at least for projects on a budget…)

    So how common is it for people to grade REC709 for P3 bound projects? Are the results acceptable (I realize the word “acceptable” is relative)? Who’s doing this? Are there any better alternatives for those of us without the physical space for projectors in our grading suites (can we generate LUTS to approximate P3 on a Flanders or equivalent high end LCD monitor that would do a better job).

    I’m trying to figure out if I need to sub-rent a P3 configured grading suite for these projects, or if I can just farm out the DCP creation at the end of a grade. What’s everyone else doing? Thanks for reading my novel of a post, I’m hoping this will generate some useful discussion points for everyone.

    Chris Hall
    Colorist – Basher Films
    Pasadena, CA

    Marc Wielage replied 10 years, 12 months ago 11 Members · 15 Replies
  • 15 Replies
  • Juan Salvo

    March 15, 2011 at 2:06 am

    Yes, people do grade for P3 in 709. Yes it works well. It looks when projected as good as things look in 709. Which when done well looks pretty damn good. I’ve venture to say most DCP work is done this way. Are you going to miss out on something in the extremes. Possibly. Will anyone notice? Doubtful. Since you don’t have any P3 gear it sounds like you would be far far better off working in the space you know and checking the converted material in a proper screening room.

  • Robert Houllahan

    March 15, 2011 at 2:19 am

    I think that some of the new 1080P DLP and LcOS projectors can be calibrated very close to P3 they won’t be on a huge screen and you will need a good probe (X-Rite Hubble, Klein K-10) and software like Lightspace or Cinespace but I think you can get close or all the way there. The Decklink card supports 3Gs SDI and a box like the HD-Link Display Port or the Gefen will transport the deep color data to such a projector.

    -Rob-

    Robert Houllahan
    Director / Colorist
    Cinelab Inc.
    http://www.cinelab.com

    MAHC-PRO 6-Core 3X GTX285 20Tb SAS Wave Panel Panny 11UK SDI Plasma.

  • Sascha Haber

    March 15, 2011 at 8:00 am

    Hi there,

    I graded a feature last fall on a proper P3 calibrated NEC NC2000C.
    The wonderful thing is that the lab we used also had a P3 starting point and could provide us with a LUT that emulated their printing process from LOG material.
    I agree that 709 is so close to P3 no one will see a difference.
    If you are handling eLin media you will be fine without a LUT.
    I just got test prints from ARRI back and they are look identical to what we saw in the grading cinema.

    A slice of color…

    DaVinci 7.1 OSX 10.6.6
    Dual Xeon 2,4 RAM 24 GB
    RAID0 8TB eSata 6TB
    GTX 285 / GT 120
    Extreme 3D+ WAVE

  • Ola Haldor voll

    March 15, 2011 at 8:04 am

    I have been grading a short film now and then in between other jobs for a while now. This was graded in my suite on a REC709 monitor. Then yesterday, we set up my Resolve workstation at the cinema across the hall and hooked it up to a Christie CP2220 projector.

    Using the RGB to XYZ LUT in Resolve, it was close to what it was on the monitor. Only needed to push things a tad more extreme to get the “cinema version” done.

    So yes, to save money.. REC709 -> LUT -> P3

  • Margus Voll

    March 15, 2011 at 10:31 am

    I have the same experience with NEC NC2000C.

    What i saw here in studio i got in cinema.
    No visible difference.

    Margus

    https://iconstudios.eu

  • Chris Hall

    March 15, 2011 at 3:41 pm

    Great feedback on this everyone, really appreciate your responses here. I’m going to contact another house and use them to generate some “test” DCP Packages from my Rec709 grades and give them a look in their P3 calibrated theater. Sounds like things should look pretty on par with this workflow judging from the feedback.

    So has anyone had any experience with monitors that claim to be able to display P3 inherently. Flanders 2470 (which they don’t sell anymore) claimed it could do 98% of P3, the Dreamcolor… well the dreamcolor claims alot of things of which I’m skeptical but it says it can emulate a P3 color space? I’ve seen the Sony BVM-L231 in its supposed “p3” mode (which actually looked fantastic) but that’s a $20,000 monitor. Anyone have any experience here? My guess is that monitors like the flanders and the dreamcolor can change their gamma response to that of P3, but cannot handle the gamut (range of color) of P3. I would assume the Sony BVM series could actually do this?

    Chris Hall
    Colorist – Basher Films
    Pasadena, CA

  • Craig Harris

    March 15, 2011 at 8:22 pm

    Awesome post!!! This is the kind of stuff I find very helpful, especially when so many post houses keep ‘tight-lipped’ in my area.

    Cheers,
    Craig

  • Joakim Ziegler

    March 18, 2011 at 1:04 am

    I’d recommend the DreamColor, if you can get it properly calibrated.

    Yes, P3 Gamma is trivial, gamut not so much, but the DC has a far larger gamut than most monitors.

    We made LUTs for our DC in full gamut mode using CineProfiler/CineSpace and a Hubbell probe, and that works excellently (also for emulating film output).


    Joakim Ziegler – Postproduction Supervisor

  • Chris Hall

    March 18, 2011 at 3:47 pm

    Joakim, might I ask what probe/lut generator combination you used to generate your luts (ie hubble/lightspace, cinetal davio, etc?), or did it gent sent to the lab for lut generation?

    Chris Hall
    Colorist – Basher Films
    Pasadena, CA

  • Joakim Ziegler

    March 20, 2011 at 11:32 am

    Hubble probe, CineProfiler for profiling. CineSpace Visual for LUT generation. The film profile was made by Cine-Tal. Currently applying the LUT in a Davio, although generating the same LUT for Resolve and applying it there yields practically identical results.


    – Joakim Ziegler

Page 1 of 2

We use anonymous cookies to give you the best experience we can.
Our Privacy policy | GDPR Policy