crushed black levels

Home Forums Help and Support crushed black levels

Viewing 8 posts - 1 through 8 (of 8 total)
  • Author
    Posts
  • #1226

    tony359 SourceForge
    Member
    • Offline

    Hi
    I have done again a madVR 3D LUT profile tonight, this time with calibration included.
    At the end I verified the system and the report was good.

    Next, I checked black and white levels using AVS-HD. I loaded the MP4 folder in MPC-HC.
    White is fine, but black is crushed. I can barely see 19. 18 is basically black.

    Black level of my projector is poor, I know. But I would expect not to see crushed levels.

    What have I done wrong?

    I’m attaching the calibration file if it’s needed.

    https://www.dropbox.com/s/2gqm6w87tqlfvdl/madVR%202015-10-24%2019-41%20D6500%20Rec.%201886%20M-S%20XYZLUT%2BMTX.rar?dl=0

    • This topic was modified on 2015-10-24 19:57:35 by tony359.
    #1227

    Florian Höch
    Administrator
    • Offline

    Hi,

    White is fine, but black is crushed. I can barely see 19. 18 is basically dark.

    The 3D LUT seems to be good though. In fact, the near black levels are slightly lifted which is what I would expect from BT.1886 on a display with not-so-deep black level like this one (level 17 becomes 3-3-2, level 18 becomes 6-6-5 on a 0..255 scale).

    #1228

    tony359 SourceForge
    Member
    • Offline

    Thanks.
    So what’s the next step? The room was completely dark (besides walls are white), there was nothing else on screen which could have spoiled the reading. Dynamic iris disabled.
    What would you suggest?

    I’ve done a more extensive verification and it looks fine.

    • This reply was modified on 2015-10-24 20:08:48 by tony359.
    Attachments:
    You must be logged in to view attached files.
    #1230

    Florian Höch
    Administrator
    • Offline

    As I said, I see nothing wrong with the LUT.

    #1231

    tony359 SourceForge
    Member
    • Offline

    Thanks.
    I have done a few more tests.
    I’ve created a verification test chart from 0% to 10% in 0.5% steps. All checks out.
    I’ve also checked with HCFR. 1% is reported as basically the same level of 0% and the 1% window on screen is really barely visible (took a while to notice it!). That said HCFR has a linear gamma as a reference.

    Now, after all we’re talking about a minor discrepancy even though I feel that proper black levels are important (ok, I know my projector is crap. Used to have a JVC, now this is what I can afford in this space).

    Any suggestions? Shall I ask Graeme’s opinion? Do you think I’m chasing ghosts here?

    I know I tend to be persistent and perfectionist! 🙂 But hey, half of the fun is setting up!

    Thanks again
    Tony

    Attachments:
    You must be logged in to view attached files.
    #1233

    Florian Höch
    Administrator
    • Offline

    That said HCFR has a linear gamma as a reference.

    I am pretty sure you can set that to Bt. 1886.

    Do you think I’m chasing ghosts here?

    I have a feeling this may be the case, to be honest 🙂

    #1234

    tony359 SourceForge
    Member
    • Offline

    You’re right and it is already set to 1886.

    I appreciate your opinion, no worries! In fact when testing I realised that that missing level is really imperceptible, I thought it was a more vast error.

    I guess I could adjust the brightness on the projector to compensate but the projector has the bad habit to massively change colour temperature when brightness and contrast are tweaked!

    Well, anyway, I believe I can be happy with the outcome for now. Sorry for being a pain and thank you for your help! 🙂

    #1235

    tony359 SourceForge
    Member
    • Offline

    Just FIY and out of curiosity I did a further test yesterday. I tweaked the levels from the projector before doing a calibration. It ended up with good black levels but with wrong white levels (not crushed, just too low, I had 237-238 flashing too).
    That said, when watching content it was clear that there were more details in the low level areas. I know it was not a major difference but to me that seemed important.

    Don’t get me wrong, you and Graeme have done a hell of a job putting together these projects and I simply love Argyll and dispcalGUI, but I feel this should be improved? I also realise that my current setup is not great, so maybe that is affecting the readings? Is there anything I can do to help?

    Thanks!

Viewing 8 posts - 1 through 8 (of 8 total)

You must be logged in to reply to this topic.

Log in or Register

Display Calibration and Characterization powered by ArgyllCMS