• New Patreon Tier and Early Access Content available. If you would like to support AVForums, we now have a new Patreon Tier which gives you access to selected news, reviews and articles before they are available to the public. Read more.

Kuro 5090 - luminance reduction over time?

E

engineering

Guest
So I was doing a test with one of our 5090 panels and making gamma adjustments with an ISF calibration yesterday, and noticed as I was taking readings making adjustments the luminance was dropping off over time. Odd, so I put up a 100 IRE box and took a luminance reading, around 190 nits. Continued to take luminance readings over time and the luminance dropped from 190.......all the way to 145. The screen then visibly refreshed and suddenly the luminance was back up to 190 again?!? Similar behaviour is also shown below 100IRE so it doesnt appear to be related to peak white, and we are talking about a time period of somewhere between 5 and 10 minutes, didn't note the time exactly.

This is the case on two panels I have tested so far, so what the heck is going on, is this normal, another function of ABL perhaps?!?
 

JonStatt

Well-known Member
Sounds like its connected with ABL to me. I just discovered this recently.

That means if you have a black frame with a small white 100IRE box on it...it will be MUCH brighter than the whole screen being 100IRE. THe more white on the screen, the dimmer the maximum output becomes.

Use digital video essentials and play a 100IRE frame...then go to the brightness setting screen with a 100 IRE rectangle in the middle of the frame only. You will see a significant difference in output (I think somewhere around double) of white.

The way to avoid this is to aim for a lower maximum output (by reducing contrast) until that difference is minimised. This effect is due to the power output/heat and other limiting factors of the plasma technology. I suspect that the effect you talk about...tailing off of brightness over time is related to the heat control aspect.

I believe I read somewhere that below 50% APL, all of these effects stop happening....
 
Last edited:
E

engineering

Guest
Ok thanks Andrew. Yes, 190 nits is very high, but is the standard out of the box contrast setting believe it or not. We aim for 25ftL here, so around 85-86 nits.

Hi JonStatt, sounds like you are more referring to the luminance reduction according to APL, which is what initially I thought was all ABL did, hence I have been calibrating with as small a greyscale windows as the probe allows, but what I am referring to here is a luminance reduction over time with a static image displayed. EDIT: just reread your post, heat control huh? I had hardly noticed the near face burning radiator like characteristics of the panel when up close :)

Calibrating these panels is to say the least 'challenging'. Luminance change according to picture content raises many questions about what APL to calibrate on its own, but now with luminance adjusting over time as well, how exactly is a proper gamma adjustment meant to be done? ABL, get rid of it!!! I think I will be looking into how ABL can be disabled or at least reduced in the service menu as I understand is possible with ControlCAL (but understandably risky).
 
Last edited by a moderator:

andrewfee

Well-known Member
Well here's how the Pioneers respond with APL:
sg36ev.png


Piers measured an LX5090 for me, and I did the KRP-500M.

I would either use a 1% size window as the EBU recommends (this is what I use) or around 15% when calibrating. Much more than that and the ABL is dimming the picture quite a lot.

The results are smoothed out a bit, but you really only see a drop from 1–5%, 5–20/25% are fairly linear and then after 25% APL brightness really starts to drop on the 5090. With the KRP-500M, this is extended out to 50%.

Measurements were taken with a peak white of 80cd/m² on both displays. That said, I have a feeling that it probably doesn't matter what your contrast setting is, it's probably just "hard-coded" into the processing of the display, rather than monitoring power usage.


When calibrating the KRP-500M, I found I had to keep switching patterns back and forth to stop it from activating the dimming "feature", leaving it to settle for 5–10s before taking a reading.


From the KRP-500M manual:
“Why does the picture seem to dim when I leave a picture on the screen?
When still images (such as photos and computer images) stay on the screen for an extended period of time, the screen slightly dims. This function protects the panel from damage by automatically adjusting the screen brightness. The screen dims when a still image is detected for approximately three minutes.”​
I've had a look at the LX5090 manual and it says it's three minutes there as well.

Using ControlCAL did not appear to stop this dimming function of the display.
 
Last edited:

JonStatt

Well-known Member
Energy Saving.

I have a graph of the 5090 with both energy saving modes, but it starts at 5% rather than 1%.

Thanks...should have thought of that! :)

Very interesting to see the difference between the 5090 and the KRP versions! I wonder if that is because of the removal of the media connections/tuner from the panel allowing more ventilation for the screen itself..and thus less heat build-up.
 
E

engineering

Guest
When calibrating the KRP-500M, I found I had to keep switching patterns back and forth to stop it from activating the dimming "feature", leaving it to settle for 5–10s before taking a reading.

Thats just painful. I like the pictures these panels produce, but I hate all this stuff. I can sort of understand this in the user presets, but in ISFccc mode?!? The people who get into this mode know what they are doing, just let us use the damn things without all this nonsense. At the very least allow this behavior to be disabled during the calibration!

Grrrrrrrr.
 

The latest video from AVForums

Guardians of the Galaxy Xmas Special, Strange World, Bones and All, and Cabinet of Dr Caligari in 4K
Subscribe to our YouTube channel
Support AVForums with Patreon

Top Bottom