Hi all,

I have a latent clipping of some black levels after applying any calibration LUT I create for any display.

probe: specbos 1211
room: black walls, completely light controlled

my workflow:
-15-30min warmup off display and probe (constant measuring)
-"display refresh rate detection"
-manual adjusting with calibration images for contrast and brightness, gamma is always spot on (all displays to 2.2) and setting white to 65k
-Quickprofile 709, 2.2 gamma

software: assimilate scratch
PEAK LUMA, PEAK CHROMA, OR FIT CHROMA luts show clipping
difgamma shows rather elevated low grey values, but precision is not high enough, the first measurement is too far from the black.

I am stuck and would appreciate any help.

cheers,
Nepo

I suspect the black readings from the probe are not accurate enough.

We never recommend using a Spectro for actual calibration - you should just use them to pre-calibrate a Tristimulus probe before using that for the actual calibration.

Spectro just can't read low light levels well enough.

Steve

Thanks, Steve for the quick respond.

I feared that. What would you suggest. Adding a X-RITE I1 DISPLAY PRO PROBE, or trying to sell the 1211 and buying a cr100 colorimeter, for example. I can't have both.

But also...Its probably more complicated than that, but shouldn't result half of the quickprofiles in LUTs with too high black levels, if its an accuracy problem?

Not sure I actually understand your last comment...
But, e-mail me your profiles ans the LUT generated from it, and I'll asses the data.

As for a Tristimulus probe - the normal approach is to by the best you can, and then add a Spectro later if needed.
The CR-100 or K10-A are both ideal.
The i1D3 is ok, but you do get what you pay for.

Steve

I meant, that bad accuracy will result in values that only go in one direction. It should be random. If not the integration time should be (even) longer?

Will send a LUT, thank you.

Is there a market for used probes? The cr100 seems better in any regard, even though its cheaper and not a spectro.

Ok - I have checked your profile data and the LUT you have generated from it.
There is no clipping in the LUT at all.
Actually, as the display has a min black of 0.05 nits (which is high for a plasma) the Jeti probe is reading well in the low-light levels.
There are no major issues with the profile, although as there are some RGB Separation variations a full 3D Cube profile would be preferable and generate better results.

But, the profile and the LUT look fine - no clipping at all.

It would seem the issue is with the LUT implementation, not with the LUT.

Hope that helps.

Steve

That helps a lot, much appreciated, Steve. That would indicate at least a cheaper solution.

Not a professional app, but I use mpc hc with madvr more for private BD watching and found the same clipping.

I can't recall, if I measured through scratch with the plugin for this LUT, so I will try that next.

@plasma: I always think the black was blacker once, but I can't find a problem in the signal path.