Brightness Calibration

[Neil]

Hi Guys!

I'm toying with an idea, but I need to start with something basic.

Is there a graphic I could use that would ensure multiple displays are at the same brightness and constrast setting?

I don't think color is important, I just want something simple that would ensure multiple displays can easily be adjusted to have the same brightness and contrast.

Regards,
Neil

[Likay]

You need a Lux-meter which can be obtained for cheap (~30 Euros) if you're not after the best precision. Make a tubeshield (even a toiletroll tube would be ok for as long as it shield light well) around the sensor so you can measure directly on screens to get the same conditions for all screens without surrounding light causing issues either..
Make a testimage with a small white square (needs to be big enough to cover the entire area of the tube you put in front of the meter, even on small displays) and the rest black.
Adjust the brightness until you have a certain amount of lux on the white square (make sure even low lumen monitors can match this). Then measure the black field and adjust contrast until you have a certain value. Measure the white square again etc until values matches.
Practically when measuring the black you should adjust contrast until the luxmeter just starts to give an indication. Projectors would be difficult to measure though (but if necessary you can make a box with the lumenmeter which contains a magnifying lense to measure a small spot on the screen).
Gamma and other interfering parameters needs to be default.

I don't know the purpose but remember that some devices have really bad blacklevels which forces you to calibrate better monitors into worse levels as well. You need to consider if this could interfere with the results you want.

[Neil]

Let me ask the question a different way. You know the ghosting test images we do on MTBS? How do we know that the images are being tested under fair and reasonable situations? What simple calibration can we do so that when these tests are done, we are reasonable sure everyone ran it under the same circumstances no what what display was tested?

Here are the rules:

1. No special meters
2. Simple graphics test
3. Simple display adjustments only - something anyone can do.

Regards,
Neil

[Likay]

Lol! I had the impression that you meant for internal tests of monitors where you control all the parameters.
Well, the best thing i can think of is then the ghosting tests that we have now. Launch the testimage. Calibrate brightness and contrast until all calibrationfields are visible.
However: With a low lumen solution it will become very hard to estimate ghosting anyway since the nuances in the lower fields of the brightness calibration will be very close to each other.
There was a thread where Fredz and others discussed solutions. We quite agreed that the image should be captured using a camera and then analysed by a software for a somewhat fair judgement. In any means it still means that the whole calibrationscale needs to be visible before measuring.

[ShawmK]

I suppose you could always use something like the THX optimizer:

http://www.thx.com/consumer/home-entert ... -overview/

It's what I always use when I'm setting up a Head-Mounted Display.

[Neil]

Ok...Likay - which ghost testing image are you thinking of?

Can you share a link?

Regards,
Neil

[Likay]

http://www.mtbs3d.com/phpBB/viewtopic.p ... 66&start=9" onclick="window.open(this.href);return false;
Werner Bloos image is imo the best for the purpose.

[AntiCatalyst]

Let me ask the question a different way. You know the ghosting test images we do on MTBS? How do we know that the images are being tested under fair and reasonable situations?

We simply don't.

you really do need light meters to do this. And even then, you'd need to calibrate each screen to a specified "MTBS" standard, to make sure that they all display the same contrast ratio.

A specified contrast ratio is needed because not all monitors can make their blacks say 1/2000th as bright as their whites(1:2000 contrast), so on "blackier" screens, you have to lower the contrast ratio intentionally to a value that any monitor can display, to make an unbiased comparison.



This is a giant task and i'm not expecting anyone to do it.. but without meters, it's all guesswork

[Likay]

Let me ask the question a different way. You know the ghosting test images we do on MTBS? How do we know that the images are being tested under fair and reasonable situations?

We simply don't.

you really do need light meters to do this. And even then, you'd need to calibrate each screen to a specified "MTBS" standard, to make sure that they all display the same contrast ratio.

A specified contrast ratio is needed because not all monitors can make their blacks say 1/2000th as bright as their whites(1:2000 contrast), so on "blackier" screens, you have to lower the contrast ratio intentionally to a value that any monitor can display, to make an unbiased comparison.



This is a giant task and i'm not expecting anyone to do it.. but without meters, it's all guesswork

I tend to agree.

[Fredz]

Unfortunately I think you can't have meaningful values in regard to contrast and brightness with a colorimeter using 3D glasses. Due to the way active or passive displays do work, colorimeters don't seem to work well in this case, people at DigitalVersus and other websites have tried that without much success.

Still, using the standard DDD ghosting test may help but you need the input of the guy who took the pictures to be sure what he saw was really what is displayed. But I think you won't get unequivoqual contrast or brightness values using that.

Good luck though, that would be the holy grail in stereo 3D if a solution to measure these values could be found.

[cybereality]

Even if the monitor is calibrated perfectly, people will be using different cameras to take the photos (even the same make and model of the same camera can produce varying results). We would have to insure that the cameras were also calibrated, though this may be difficult to impossible (for example, with point-and-shoot cameras that have automatic settings). This is probably more of an issue for active glasses, as I assume passive technologies can be filmed easier. That said, I think we can still take pictures, and that they still hold value. But its very difficult to look at pictures from two different people under all sorts of different conditions and say for sure what it means. At the same time, monitors with really bad ghosting will show that in a picture and monitors with almost no ghosting should show that too. But when its close it may be hard to say whats better.

[Neil]

I'm not interested in having pictures taken at this point. I'm more interested in just having a simple test that everyone can do to ensure that their brightness and contrast settings (or to be more precise, the brightness and contrast that is visually displayed) is reasonably closely matched. Also, this brightness calibration isn't done while wearing 3D glasses.

Regards,
Neil