or Connect
AVS › AVS Forum › Display Devices › Display Calibration › Question about gamma/greyscale
New Posts  All Forums:Forum Nav:

Question about gamma/greyscale

post #1 of 14
Thread Starter 
Hey guys,


I decided to join up after having a question about Gamma/Greyscale curve, a proper introduction will follow swiftly. A loooong time ago I had a training by an ISF technician and learned how to calibrate the earlier generations of Pioneer plasma's.
After the training I received Colorfacts Pro with a datacolorspyder2 to be able to perform these calibrations. in the shop we decided after some time that it took us too much time to perform it at clients and decided to refer to specialized businesses for calibrations.

However I do have the kit and was able to put my Panasonic TH42PZ800 at a nice Gamma of 1.79 but now I read that these days most screens are calibrated at 2.2. I know it has to do with the relation of putting a signal in your screen and the amount of output you get at the screen-end. What is the best way to try to get to 2.2? Do you put the output of all the primaries higher? I would like to create a daytimesettign with 2.2 and keep the nighttime-setting at 1.8.

Please advise, thank in advance for your help!smile.gif

Dan.
post #2 of 14
Greetings

I'm not sure that people that simply attend a class ... are suddenly qualified to be teachers and can then teach others.

The stuff you have said about gamma is scary wrong. It's backwards ...

In dark rooms we want higher gamma numbers like 2.4 ...
In rooms with lots of ambient light then something like 1.8 -2.0 might be better.

regards
post #3 of 14
Thread Starter 
I just tried to lay out a little history on how I came across the art of calibrating, we gave it a shot a couple of times in the showroom but quickly came to the conclusion that it wasn't for us to do commercially. The idea on taking that class was to get a better understanding of pioneer plasmascreens at the time, I would never claim to be an expert on this subject.

I'm trying to understand gamma better. I was wondering how one would be able to get the gamma value from 1.8 to 2.2 on a panasonic plasma, basically. I don't care if it takes me half a night, I just like to see what the difference is between the two so I can make a setting for both night and daytime viewing.

Thanks for your help,

Dan.
post #4 of 14
i am confused by your post??

higher output of all primaries is not used for setting Gamma

Setting Gamma is like view an 100% stimulus and adjust (all together) RGB (grayscale) from 90% to 0% with a 10 point (or 20 point) system AFTER you set gamma in user menu (if available)

Otherwise you set contrast and brightness and your stuck to your tv's standard gamma setting.

Nightime setting @ 1.8 ?
Daytime setting @ 2.2?

Night and Day gamma setting would be more logical to set this in reverse of what you suggesting.

Greetings
post #5 of 14
Thread Starter 
Quote:
Originally Posted by Michael TLV View Post

Greetings
The stuff you have said about gamma is scary wrong. It's backwards ...
In dark rooms we want higher gamma numbers like 2.4 ...
In rooms with lots of ambient light then something like 1.8 -2.0 might be better.
regards

@killerbeenl; if anything I am confused by Michael's post, I thought 1.8 was good for nighttime and for example 2.2 for daytime?

I just would like to know how to get the gamma to a higher level.

Later,

Dan
post #6 of 14
Short answer is that, yes, you can set RGB gains and offsets higher or lower to adjust gamma... maybe, sorta, kinda. You might do better there, or not. Pro Studios doing mastering for movies are calibrated to 2.2 or 2.3 at the most, never higher... I find setting gamma to 2.25 looks best the vast majority of the time.

The reality is that many displays have gamma problems and lack the controls to make gamma uniform for all steps. There's no such thing as ONE gamma value for a display... you can average the gamma for many grayscale steps and get an average, but that can be VERY misleading because every step could have a HUGE gamma error and it just might happen that they average-out to looking OK. You really need calibration software that shows you whether each grayscale step is as bright (or dark) as it is supposed to be to achieve the same gamma target (say 2.25) for every grayscale step, not just as an average over the entire grayscale. TVs with 10-point grayscale adjustments make that fairly easy to achieve. TVs with just 2 grayscale adjustments (high/low, cut/gain, offset/bias) are much more problematic. Even if the TV has a gamma control, the settings can be pretty bad and not be close to the labeled gamma (meaning a Gamma setting of 2.2 may not measure 2.2 at all... or it may only be 2.2 at one point on the grayscale while other steps are too bright or too dim).

Finally, ColorFacts Pro is woefully out of date as calibration software. It is essentially worthless now. And the Spyder2 was NEVER a decent meter to use for calibration of anything... it would be somewhat OK for CRT or Plasma displays, but it is TERRIBLE for LCD displays (produces very inaccurate results) and can't really be used for projection systems as it can't take readings off the screen. And colorimeters drift as the filters age... after 3-5 years the filters change enough that the measurement results will very likely produce visible errors.
post #7 of 14
Thread Starter 
Quote:
Originally Posted by Doug Blackburn View Post

Short answer is that, yes, you can set RGB gains and offsets higher or lower to adjust gamma... maybe, sorta, kinda. You might do better there, or not. Pro Studios doing mastering for movies are calibrated to 2.2 or 2.3 at the most, never higher... I find setting gamma to 2.25 looks best the vast majority of the time.
The reality is that many displays have gamma problems and lack the controls to make gamma uniform for all steps. There's no such thing as ONE gamma value for a display... you can average the gamma for many grayscale steps and get an average, but that can be VERY misleading because every step could have a HUGE gamma error and it just might happen that they average-out to looking OK. You really need calibration software that shows you whether each grayscale step is as bright (or dark) as it is supposed to be to achieve the same gamma target (say 2.25) for every grayscale step, not just as an average over the entire grayscale. TVs with 10-point grayscale adjustments make that fairly easy to achieve. TVs with just 2 grayscale adjustments (high/low, cut/gain, offset/bias) are much more problematic. Even if the TV has a gamma control, the settings can be pretty bad and not be close to the labeled gamma (meaning a Gamma setting of 2.2 may not measure 2.2 at all... or it may only be 2.2 at one point on the grayscale while other steps are too bright or too dim).
Finally, ColorFacts Pro is woefully out of date as calibration software. It is essentially worthless now. And the Spyder2 was NEVER a decent meter to use for calibration of anything... it would be somewhat OK for CRT or Plasma displays, but it is TERRIBLE for LCD displays (produces very inaccurate results) and can't really be used for projection systems as it can't take readings off the screen. And colorimeters drift as the filters age... after 3-5 years the filters change enough that the measurement results will very likely produce visible errors.

Thanks! That was very helpfull! At least it's good to know that I never used the colorspyder on anything else than plasmascreens, also good to know that it's probably more or less useless by this time.. The colorfacts software has a 10 point grayscale measurement, so on average I was able to see how the screen responded to it's poor two-point adjustmentsystem, that was actually how I was told to calibrate a screen, back in 2005 (Jeez time goes fast doesn't it?! LOL) a Pioneer plasma only had cuts and gains just like my '08 Panasonic.
In the end it means that I can try to use the cuts and gains of the primaries to get a higher gamma-curve.

Thanks for your response,

Dan
post #8 of 14
Greetings

Depending on the size of the test patterns used to determine the gamma, the smaller the boxes, the higher the gamma number goes. (Pick the box size to get whatever gamma you want ... it is also a form of lying to yourself that you are actually doing anything.)

If can't really change the gamma much by mucking with brightness and contrast. You muck with brightness and you end up crushing your blacks or washing them out. Better gamma but crappy blacks.

Muck with the contrast and you reduce the contrast ratio ... or start to clip or discolor whites ... Better gamma, but crappy contrast ratio ... or discolored image missing detail.

If you set your brightness and contrast correctly, then your gamma is pretty much what it is. If you want to change it, you need to get an external processor box for the TV.

regards
post #9 of 14
Quote:
Originally Posted by Michael TLV View Post

(Pick the box size to get whatever gamma you want ... it is also a form of lying to yourself that you are actually doing anything.)

Ain't plasmas fun? I suspect the same joy will come from OLED but please correct me if I'm wrong. rolleyes.gif
post #10 of 14
Thread Starter 
Quote:
Originally Posted by Michael TLV View Post

Greetings
Depending on the size of the test patterns used to determine the gamma, the smaller the boxes, the higher the gamma number goes. (Pick the box size to get whatever gamma you want ... it is also a form of lying to yourself that you are actually doing anything.)
If can't really change the gamma much by mucking with brightness and contrast. You muck with brightness and you end up crushing your blacks or washing them out. Better gamma but crappy blacks.
Muck with the contrast and you reduce the contrast ratio ... or start to clip or discolor whites ... Better gamma, but crappy contrast ratio ... or discolored image missing detail.
If you set your brightness and contrast correctly, then your gamma is pretty much what it is. If you want to change it, you need to get an external processor box for the TV.
regards

Yes, wouldn't want to do that, usually I stick with framed windows around 20% of the total screen area. So what you are saying is that you cannot change the gamma-performance on a plasma by going into the service menu and tweak the output of the primary colors without causing discoloration.
I was under the impression (and I could go waaaay off base here) that if you adjust the level of all the primaries upwards (servicemenu setting), the gamma would go up higher, simply because you put more energy into the screen.
Dan
post #11 of 14
Quote:
Originally Posted by buzzard767 View Post

Ain't plasmas fun? I suspect the same joy will come from OLED but please correct me if I'm wrong. rolleyes.gif

OLED are a bit different.

Since I've only had a chance to play around with the Sony BVM model, I don't know how this will translate to consumer models.

Every pixel is dynamic, which means if one is off, the pixel next to it can be full on, unlike plasma you can also have all of them on full bore with no loss in luminance.


The end result is that they are more similar to old fashion CCFL LCDs, where every pattern size measures exactly the same.
post #12 of 14
Thanks Joel. That should simplify calibrating and probably gives us the best pictures so far. wink.gif
post #13 of 14
Quote:
Originally Posted by buzzard767 View Post

Thanks Joel. That should simplify calibrating and probably gives us the best pictures so far. wink.gif

The Sony OLED BVM was amazing, out of the box dE of <0.5 as measured with the CS-2000 and a 300,000:1 contrast ratio measured with dim lighting.
post #14 of 14
Quote:
Originally Posted by sotti View Post

The Sony OLED BVM was amazing, out of the box dE of <0.5 as measured with the CS-2000 and a 300,000:1 contrast ratio measured with dim lighting.

I've read wonderful things. Now - get 70" into the $6000 range.... sigh
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration
AVS › AVS Forum › Display Devices › Display Calibration › Question about gamma/greyscale