VA Calibration - which point to use? Center or Off-Center?

  • Thread starter Deleted member 273615
  • Start date
D

Deleted member 273615

Guest
Although I know a great deal about calibration, I still haven't been able to find answers to some questions, even on AVS Forums. I thought I would try here, just in case.

When calibrating a display, it mostly advised to use the center of the screen as the calibration point with exception for TN panels due to their gamma shift problem. However, I came across a somewhat interesting issue with my Eizo Foris FG2421, which is an MVA panel. There are 3 major blobs on the screen where contrast ratio is only about 3500:1 and blacks are slightly higher than on the rest of the screen, where contrast ratio is about 4500:1 and black levels are better/lower. These blobs only make up about 25-30% of the screen with 2 of them located near the corners of the screen and the last one right in the very middle! Now I am more than aware of why this happens - uniformity, but it makes calibration difficult.

If I use the center of the screen to calibrate it, then I end up with rather accurate RGB WB throughout the entire screen, but gamma (BT.1886) ends up being accurate in the middle, but low (a bit too bright) in the rest of the 70% of the screen. If I do it the other way around, calibrate using a point off the center of the screen to have accurate gamma for the 70% of the screen, then the 30% remaining end up with high (a bit too dark) gamma value.

VA screens are supposed to shift gamma from high to low (from dark to bright) as they move away from the center, but in this case uniformity screws that up. This is also why VA screens typically have black crush in the center, but in this case its the other way around! This creates a dilemma since using the center of the screen reduces black crush in that center, but leaves the rest of the screen with low gamma, while doing the opposite makes it hard to see low level gray/black details. Which way should I go and why?

Please don't turn this into a "Calibration is Overrated" thread and you could care less for it or how you think its impossible to calibrate a VA perfectly, a fact I am very aware of. I am only asking those who do care and do understand the dilemma. Thanks!

P.S. I use i1Display Pro colorimeter, i1Pro spectrophotometer, and the latest ArgyllCMS + dispcalGUI + HCFR software.
 
Umm, I don't think that is how VA gamma shift works. It is related to the viewing angle, not off-center position. As for uniformity issues, you calibrate to the area that you are more interested in getting right. That would be the biggest uniform coverage and/or the center of the screen.
 
display uniform dark gray image with gray of values like #101010 and see what happens when you move your head, try looking at screen from different positions :rolleyes:
 
Meeho said:
Umm, I don't think that is how VA gamma shift works. It is related to the viewing angle, not off-center position. As for uniformity issues, you calibrate to the area that you are more interested in getting right. That would be the biggest uniform coverage and/or the center of the screen.
Click to expand...

You're gonna laugh at yourself when you realize that it's physically impossible to see every part of the screen from the same angle.

OP, as for your issue, unfortunately, there's no objective truth. It will depend on your taste. I would personally go for what gives you overall deeper blacks and not sweat spots of black crush too much.
 
Last edited:
Siamese Almeida said:
You're gonna laugh at yourself when you realize that it's physically impossible to see every part of the screen from the same angle.
Click to expand...

What makes you think I've said any different? I was replying to OP regarding the following:
MonarchX said:
VA screens are supposed to shift gamma from high to low (from dark to bright) as they move away from the center, but in this case uniformity screws that up. This is also why VA screens typically have black crush in the center, but in this case its the other way around! This creates a dilemma since using the center of the screen reduces black crush in that center, but leaves the rest of the screen with low gamma, while doing the opposite makes it hard to see low level gray/black details. Which way should I go and why?
Click to expand...

Looking at the screen and the positioning of the contact meter are two different things. Putting the meter off-center won't change anything regarding calibration due to VA's gamma shift. What would make a difference is positioning the meter at an angle to the screen.
 
Meeho said:
What makes you think I've said any different? I was replying to OP regarding the following:


Looking at the screen and the positioning of the contact meter are two different things. Putting the meter off-center won't change anything regarding calibration due to VA's gamma shift. What would make a difference is positioning the meter at an angle to the screen.
Click to expand...

OK, that is partially true and I may have been wrong regarding the gamma shift for most VA panels, BUT the 3 blobs where contrast ratio is lower (black levels are higher) do exist on my screen and it doesn't matter which angle you use to look at them - the colorimeter readings don't lie.

If I use the center to calibrate the screen the resulting image is brighter than the image that results from calibrating the screen using an off-center point where contrast ratio is higher (black levels are lower). Brighter image would leave about 30% of the screen with a rather accurate gamma, but the 70% of the screen would have lower-than-needed gamma. The other way around seemed to make more sense as it would allow 70% of the screen to maintain accurate gamma with 30% (the 3 blobs) having higher-than-needed gamma.

In the, I settled for the darker image before selling my probe to buy a PC update. Right now http://www.lagom.nl/lcd-test/gamma_calibration.php shows that my gamma is at about 2.3, which is higher than I wanted, but again, the image is in the very center and such tests are not reliable...

I was wondering if there was a way to compromise between the center and off-center gamma discrepancies by manually adjusting gamma to be between the 2 possibilities, but how the hell would I even go about doing that?
 
MonarchX said:
OK, that is partially true and I may have been wrong regarding the gamma shift for most VA panels, BUT the 3 blobs where contrast ratio is lower (black levels are higher) do exist on my screen and it doesn't matter which angle you use to look at them - the colorimeter readings don't lie.

If I use the center to calibrate the screen the resulting image is brighter than the image that results from calibrating the screen using an off-center point where contrast ratio is higher (black levels are lower). Brighter image would leave about 30% of the screen with a rather accurate gamma, but the 70% of the screen would have lower-than-needed gamma. The other way around seemed to make more sense as it would allow 70% of the screen to maintain accurate gamma with 30% (the 3 blobs) having higher-than-needed gamma.
Click to expand...

Yes, unfortunately, with such uniformity defects you'll have to compromise and choose what makes more sense to you.

In the, I settled for the darker image before selling my probe to buy a PC update. Right now http://www.lagom.nl/lcd-test/gamma_calibration.php shows that my gamma is at about 2.3, which is higher than I wanted, but again, the image is in the very center and such tests are not reliable...

I was wondering if there was a way to compromise between the center and off-center gamma discrepancies by manually adjusting gamma to be between the 2 possibilities, but how the hell would I even go about doing that?
Click to expand...

That test is unreliable. Besides, you have a meter and proper software, why not use that? As for the other part, why don't you just measure the gamma in both spots and then calibrate to a middle target?
 
Hi MonarchX,

You should use the standard perpendicular viewing direction in the center of the screen. You can view the IDMS for further guidance. There is no mitigation for MVA issues aside from sitting further away.
 
You must log in or register to reply here.
Back
Top