LG C8 Lut

Home Forums General Discussion LG C8 Lut

Viewing 15 posts - 31 through 45 (of 139 total)
  • Author
    Posts
  • #17926

    tonybcc
    Participant
    • Offline

    Hi Guys,

    but license is free so it is easy do get it.

    I use rasppbery pi as pattern generator and i had to stop trying to get LUT unles Florian will add it to the program . Meanwhile i bought Calman and will test it. Once PI will be added then i can donate and come back to LUT.

    My expierience with DisplayCal was good just not sure how to calibrate my C8 since i cant change anything in TV or like in HCFR change settings in program. Maybe I dont understand how to calibrate it  using DisplayCal. As for LUT it is easy but cant get good results before i ger automated generator thats why i asked for PI support.

    Regards,

    #17936

    stama
    Participant
    • Offline

    @tonybcc, do you mean you can get access to the LG OLED template for DeviceControl with a LightSpace ZRO license (the version that does not cost money)?

    #17967

    tonybcc
    Participant
    • Offline

    No i meant i can get acces to tv using DeviceControl and DispalyCAL.  LightSpace ZRO is free and doesnt have 3dlut .

    #19179

    residentjack
    Participant
    • Offline

    Hi Everyone,

    I’ve been struggling to understand the process of calibrating a C8 for use with Davinci Resolve. Could someone please let me know if these would be the correct steps to take:

    1. Connect Davinci Resolve to DisplayCal.
    2. Run the typical 3D Lut creation process with settings tweaked for an OLED display.
    3. Save the 3D Lut created by DisplayCal.
    4. Upload the 3D Lut to the LG TV via DeviceControl.

    Is all of this correct? Thank you so much in advance for your help!

    #19180

    residentjack
    Participant
    • Offline

    Hi Everyone,

    I’ve been struggling to understand the process of calibrating a C8 for use with Davinci Resolve. Could someone please let me know if these would be the correct steps to take:

    1. Connect Davinci Resolve to DisplayCal.
    2. Run the typical 3D Lut creation process with settings tweaked for an OLED display.
    3. Save the 3D Lut created by DisplayCal.
    4. Upload the 3D Lut to the LG TV via DeviceControl.

    Is all of this correct? Thank you so much in advance for your help!

    Ok, Never mind… I realize that I would need a paid license of LightSpace in order to get the 3D lut uploaded into the TV.

    So I guess the only way to go is the traditional Davinci Resolve 3DLut Monitor route right?

    #20318

    János Tóth F.
    Participant
    • Offline

    Alright, the channel order (of most rapidly increasing to least rapidly increasing channel) in the LG LUT files is different from Autodesk/Kodak format (R – G – B vs the more common B – G – R). DisplayCAL 3.8.1 offers the LG format as a separate choice.

    Do you plan to add an option to generate 1DLUT for LG (either from static measurement data collected with a neutral [post-DDC-reset] 1DUT or by running the regular iterative dispcal process on the VGA LUT and then rounding the 16 values to 12bit and putting them to the correct format) and 3DLUT for HDR10/HLG modes (I think that’s a 3×3 matrix)?

    #20328

    János Tóth F.
    Participant
    • Offline

    Note though that this EDR file has been created with the TV in Rec. 709 gamut emulation mode. DisplayCAL comes with a CCSS (LG OLED 6 series) that was created with the TV in native (roughly P3) gamut mode.

    If this would still be considered useful, I can upload a CCSS for LG OLED 2018 (C8) in native panel mode (full DDC reset = neutral internal 1D/3D LUTs) with an i1Pro2. However, to be honest, I wouldn’t personally trust this anymore than the i1d3 with any similar OLED EDR/CCSS created with a significantly better spectrometer (of higher native resolution, more tight manufacturing tolerances, more heath stability, etc). I actually chose not to use my i1Pro2 to create a matrix correction in CalMAN for my i1d3 the last time I calibrated this C8. The factory OLED EDR (yes, I know that’s for RGB, not WOLED) gave me de~1 for the factory D65 preset, so I assumed it’s probably best to go with it. Moreover, I didn’t even calibrate this C8 (I used it in full factory Cinema mode with the “enhancers” off) with the old firmwares. I only started experimenting after the “macro block” fix screwed up the factory LUTs (effectively caused some black-crush like tone response drifts) because I don’t really trust it’s low-light handling either (well, not as long as it looked fine — and it did, until the firmware updates).

    i1Basic Pro 2 on Amazon  
    Disclosure: As an Amazon Associate I earn from qualifying purchases.

    #20330

    Florian Höch
    Administrator
    • Offline

    Do you plan to add an option to generate 1DLUT for LG

    I might look into it in the future, but it’s not a priority.

    If this would still be considered useful, I can upload a CCSS for LG OLED 2018 (C8) […]

    It’s always nice to have some more data.

    […] I wouldn’t personally trust this anymore than the i1d3 with any similar OLED EDR/CCSS created with a significantly better spectrometer (of higher native resolution, more tight manufacturing tolerances, more heath stability, etc)

    In the case of WOLED that seems surprising, as the spectra are very smooth with no harsh peaks. I would expect the variation between different WOLED TVs (possibly even of the same model) to result in higher variation in the measurements than the choice of prepress grade vs lab grade spectrometer.

    #20331

    János Tóth F.
    Participant
    • Offline

    I might look into it in the future, but it’s not a priority.

    Understandable.

    I  think the easiest and possibly the best solution would be running dispcal in it’s regular iterative mode using the VGA LUT as usual (after a DDC reset of the LG hardware LUTs) and then converting the resulting 1DLUT to LG’s 1DLUT format for upload (with external software). So just rounding and formatting, not much extra work (CalMAN is very dumb when compared to dispcal, especially if there is some back-crush on some units).

    I am not sure about the HDR10/HLG mode matrix formats (that probably needs a little extra work with a new 3DLUT preset). But CalMAN seems to be able to handle matrix based 3LDUTs, so a neat 1DLUT would be a bigger help, actually.

    In the case of WOLED that seems surprising, as the spectra are very smooth with no harsh peaks. I would expect the variation between different WOLED TVs (possibly even of the same model) to result in higher variation in the measurements than the choice of prepress grade vs lab grade spectrometer.

    Well, in the absence of better data, I am still going by this: https://www.drycreekphoto.com/Learn/Calibration/MonitorCalibrationHardware.html

    #20396

    János Tóth F.
    Participant
    • Offline

    I think I uploaded the LG C8 .ccss but I can’t find it in the database.

    And I just remembered I was wary about using a ccmx because it’s a WRGB panel, so the pure R,G,B readings (captured for the gamut mapping) will be drifted by including W in the ccmx calculation. CalMAN didn’t show me by how much but DisplayCAL gave me an idea (probably de~0.4 on which is not much).

    Oh, and that 1DLUT thing would be extremely helpful with DolbyVision modes which currently requires expensive pattern generators (or semi-expensive metadata injectors) for CalMAN when in actual reality DV probably needs the exact same 1DLUT as the HDR10 mode does (but CalMAN won’t allow to save and re-upload a 1DLUT, not even between HDR10 presets – I think for this very reason, to sabotage DolbyVision workarounds).

    Attachments:
    You must be logged in to view attached files.
    #20399

    j82k
    Participant
    • Offline

    +1 for adding LG 1DLUT and HDR LUT generation with DisplayCal.

    My LG C8 has one of these black crush panels and a really messed up gamma curve. I’ve already done various 3DLUTs and it certainly is better than what I can archive with a manual calibration but I think a proper 1DLUT would be extremely helpful for messed up panels like mine…

    • This reply was modified 2 months ago by j82k.
    #20683

    Anonymous
    Inactive
    • Offline

    Hi,

    I’ve just calibrated my B8 and uploaded the 3D LUT (LG OLED format) with DeviceControl but what I get is an orange -ish image.
    What went wrong?

    #20742

    chros
    Participant
    • Offline

    I’ve just calibrated my B8 and uploaded the 3D LUT (LG OLED format) with DeviceControl but what I get is an orange -ish image.
    What went wrong?

    LG B8 only supports lower resolution 17 points cube instead of 33 that DisplayCal doesn’t support yet (?).

    Florian, can you add support for 17 points as well?

    Thanks

    #21078

    János Tóth F.
    Participant
    • Offline

    I ran into some unexpected issues during my first attempt with a C8 and CalMAN (as hardware controller) and DispplayCAL 3.8.8:

    I had to  rename the .dcl file to .3dl and replace the # lines from an old CalMAN generated .3dl file because CalMAN complained about the size. But this wasn’t fully straightforward because the CalMAN file also contained some strange data (with no hashtags before that line) between the metadata header (software version, date, etc) and the obviously human-identifiable 3DLUT entries. It looks like this (see below) and seems to be 10bit R=G=B levels (but, by the looks of it, not the ones from the “26 point SDR 16-255” test patch series recommenced by CalMAN for LG AutoCAL) and it’s not a 3×3 matrix either…

    #title: 33x33x33 3D LUT
    64 94 124 154 184 214 244 274 304 334 364 394 424 454 484 514 544 573 603 633 663 693 723 753 783 813 843 873 903 933 963 993 1023
    0 0 0
    176 32 31

    And once I managed to load the file, the R and B channels seem to be swapped (similar to using  an “XYZ + swapped matrix” ICM file, except I used a normal XYZ profile with ‘inverse d-t-p’). But image is otherwise decent (no other crazy color shifts or banding, etc).

    Edit: Ah, the original Autodesk / Kodak .3dl format seems to work fine, except for a “points mismatch” error: the DisplayCAL file now comes with that leading “points” series (0-1023) but I still have to replace that with one form a CalMAN file (64-1024) to satisfy CalMAN’s LUT loader. I am not sure if this might cause inaccuracies because I used a small amount of measurement points for the profile (512 iterative patches), so a little black crush and general inaccuracy is sort-of expected (especially since I used a neutral 1DLUT).

    #21086

    stama
    Participant
    • Offline

    It’s possible DeviceControl is the one which requires a slightly different format of the 3dl file than the original Autodesk format, if that’s what Calman uses (I don’t know where to find the original 3dl specification to check myself). That’s why the DisplayCal generated 3dl file is likely adequate to use only with DeviceControl.

    Regarding the range of values in the file, I remember to have seen Ted (on avsforum) post that recent versions of Calman also generate the whole 1024 series of points.

    In any case, profiling using full RGB works fine with DisplayCal while having the LG Black Level control set to High, there’s no need to calibrate using 16-235 or 16-255 and the LG Black Level control set to Low, which is the only way Calman is able to. DisplayCal is only able to generate the 3D LUT, but leaving the 1D LUT untouched seems to be the recommended way to minimize banding and posterization anyway, according to posts in the LightSpace threads on avsforum.

    In my case I did a really simple profile using just the fast LUT profile in DisplayCal, which is a cube of 4^3 with 52 grayscale points I think, and have no black crush and dEs are unexpectedly low for such a fast job, all below 1. But I did not touch the 1D LUT at all, not even the white balance controls, I only set OLED light and Brightness before profiling (had to set brightness to 49, otherwise there is a light glow). My only concern is that the white point was made worse by the DisplayCal/argyll generated 3D LUT for some reason (it was originally quite good already).

    Btw, access to DeviceControl no longer requires having a LightSpace license. Just contact Ted for access to the LG templates in DeviceControl using the “Request LG Templates for DeviceControl” form in the Contact section on his website: https://www.displaycalibrations.com

    • This reply was modified 3 weeks, 1 day ago by stama. Reason: additional info on profiling
Viewing 15 posts - 31 through 45 (of 139 total)

You must be logged in to reply to this topic.

Log in or Register

Display Calibration and Characterization powered by ArgyllCMS