AVS Forum banner
1 - 10 of 10 Posts

·
Registered
Joined
·
3,248 Posts
Discussion Starter · #1 ·
All right, I am so desperately predictable.


I bought an eye-one display LT and I got a hold of HCFR...curiosity had the better of me and now I am doomed.



I spent 5 or so hours last night getting familiar with the procedure described in this guide: http://www.curtpalme.com/forum/viewtopic.php?t=10457


Prior to that, I made myself watch the entire DVE disc and I also did search these forums extensively to understand the role of the Backlight control and what range should light output should fall in for the IRE 100 window pattern.


And now, tada, here I come back...


My gray scale now seems much better, but I have some questions based on what I saw.


My TV is a Bravia KDL-40Z4100. In use for less than year.


1) After some tinkering, I found that using "Warm 1" for Color Temperature and "Medium" for Gamma would get me better results right off the gate. (by improving both the look of the Luminance graph and the look of the Gamma graph)


2) I also decided to go with "Standard" for Color Space. A PS3 was the source, RGB Full Range (HDMI) was set to Limited, Y Pb / Cb Pr / Cr Super-White (HDMI) was set to On, BD / DVD Video Output Format (HDMI) was set to Automatic.


3) When I started to measure my initial gray scale, I noticed something with the Video Black w/ new PLUGE pattern.

When I display it, my panel is a certain level of bright during the first 1-3 seconds, then appears to dim after that. I do have auto-dimming set to off and I do not see this behavior with the IRE 10 pattern or any of the other window pattern used for gray scale calibration.

Is this normal? Does it invalidate my measurements at IRE 0?

I also have to pause for that pattern because the DVE disc doesn't appear to loop it like other patterns.


4) Adjusting Gain and Bias for R, G, B proved to be easier than I thought. Except my IRE 0 and IRE 10 measurements are not so good with a deltaE of roughly 100 and 12. From IRE 30 to IRE 90, deltaE is mostly below 2.

After calibration, my Luminance graph is practically perfect and my RGB levels appear to be flat on 100% from IRE 30 to IRE 90. My average Gamma reading is also 2.23.


5) Adjusting primary and secondary colors was a lot more difficult.

Setting "100% Red" correctly was fairly easy, though.

Adjusting the secondaries was nearly impossible, "100% Cyan", "100% Yellow" and "100% Magenta" are still quite far apart, even though each ended up in the inner circle of the target on the "Selected color" panel.


6) My set doesn't have a proper CMS, so I had to stop there.


After calibration, I noticed that I was getting above 40 ftL for the IRE 100 window pattern, while I had started at nearly 31 ftL.

Is that normal?


After calibration, I also have the impression I have some color leaking into my gray bars. It was late and maybe my eyes were shot, but is this a common reaction after calibration?


I would appreciate your feedback.
 

·
Registered
Joined
·
3,248 Posts
Discussion Starter · #3 ·

Quote:
Originally Posted by mhdiab /forum/post/18155325


I am just starting down this path and a bit behind you so first congrats to getting going



Can't help, but may I ask why you didn't go with the AVS test patterns and then see if you have the same issue.

http://www.avsforum.com/avs-vb/showthread.php?t=852536

http://www.avsforum.com/avs-vb/showthread.php?t=948496

I used the DVE disc for setting the gray scale.

I used the AVSHD disc for setting Color and Hue.


My colors are still very unruly and I wonder if would not be better leaving that to default entirely.


Going to post my HCFR files in a bit.
 

·
Registered
Joined
·
3,248 Posts
Discussion Starter · #4 ·
HCFR files


File 1 is factory default after setting:


Color Temperature to Warm 1

Sharpness to 10

Gamma to Medium

Color Space to Standard


And turning off all the bells and whistles such Black Corrector, Advanced CE, various noise reduction controls, etc.


File 2 is after gray scale calibration.


File 3 is after primaries and secondaries were somewhat adjusted. Primaries and secondaries still have deltaE above 10, though, and I am not pleased with that.


I guess Color and Hue is not enough control to get accurate colors.

 

Color_after_CT_S_G_CS.zip 9.2099609375k . file
 

Attachments

·
Registered
Joined
·
1,011 Posts

Quote:
Originally Posted by Dralt /forum/post/18154934


All right, I am so desperately predictable.

Quote:
My gray scale now seems much better, but I have some questions based on what I saw.

I agree, it looks very nice.
Quote:
1) After some tinkering, I found that using "Warm 1" for Color Temperature and "Medium" for Gamma would get me better results right off the gate. (by improving both the look of the Luminance graph and the look of the Gamma graph)

For choosing which color temp to start with, you might do better looking at the RGB graph.
Quote:
2) I also decided to go with "Standard" for Color Space. A PS3 was the source, RGB Full Range (HDMI) was set to Limited, Y Pb / Cb Pr / Cr Super-White (HDMI) was set to On, BD / DVD Video Output Format (HDMI) was set to Automatic.

Since you're having trouble with the color decoding, you may try forcing Y'CbCr or RGB and see if it clears anything up, or makes anything worse.
Quote:
3) When I started to measure my initial gray scale, I noticed something with the Video Black w/ new PLUGE pattern.

When I display it, my panel is a certain level of bright during the first 1-3 seconds, then appears to dim after that. I do have auto-dimming set to off and I do not see this behavior with the IRE 10 pattern or any of the other window pattern used for gray scale calibration.

Is this normal? Does it invalidate my measurements at IRE 0?

I also have to pause for that pattern because the DVE disc doesn't appear to loop it like other patterns.

Hmmm. auto dimming off, is there anything else called "auto contrast" or "contrast enhancement" or the like? Yes, it looks like it may have messed with your 0% reading, which BTW, since I believe you're an engineer, the term IRE should not be used with digital video.
Quote:
4) Adjusting Gain and Bias for R, G, B proved to be easier than I thought. Except my IRE 0 and IRE 10 measurements are not so good with a deltaE of roughly 100 and 12. From IRE 30 to IRE 90, deltaE is mostly below 2.

After calibration, my Luminance graph is practically perfect and my RGB levels appear to be flat on 100% from IRE 30 to IRE 90. My average Gamma reading is also 2.23.

0% and 10% are the hardest parts for a display to do properly. You may notice that they never show you those points in magazine reviews. Did you just do balancing at 30% and 80%? You can try tweaking it at 20%-80%, take a full grayscale set, see if it improves... tweak at 10%-80%, take a grayscale set, see if it improves... Some meters are not to be trusted at these low light levels... in which case they sometimes suggest trusting your eyes, as long as you're looking at a steps pattern (your eyes are no good at looking at a single field and saying "it's right" but their very good at seeing "that bar is bluer than the one next to it...)... I think the I1D2 you bought is supposed to be ok at low light levels, but I may be mistaken... it may only be good down that low for absolute luminance, not color balance... sorry, but I don't know...
Quote:
5) Adjusting primary and secondary colors was a lot more difficult.

Setting "100% Red" correctly was fairly easy, though.

Adjusting the secondaries was nearly impossible, "100% Cyan", "100% Yellow" and "100% Magenta" are still quite far apart, even though each ended up in the inner circle of the target on the "Selected color" panel.

The I1D2 isn't (I believe) considered good for CMS work anyway, so no foul that you don't have one... it does look like your secondaries are off, I'm not sure how good the I1D2 is at doing that... how does it look to your eyes? You may be better off using the filters for that, especially if they're not leaking...
Quote:
After calibration, I noticed that I was getting above 40 ftL for the IRE 100 window pattern, while I had started at nearly 31 ftL.

Is that normal?

Well, let's say it may not be abnormal. Absolute fL don't really matter all that much. If you get color shift or clipping, you turn down contrast. If you're getting eyestrain, you turn down the backlight. Either of those could mess with your grayscale though.
Quote:
After calibration, I also have the impression I have some color leaking into my gray bars. It was late and maybe my eyes were shot, but is this a common reaction after calibration?

Color leaking? Not sure exactly what you mean. BUT... it's not uncommon for your first look at a calibrated pattern to seem, um, strange... some people complain of yellow, or red, or warmness... you've been looking at a blue picture for so long it takes a little time for your eyes to adjust...


Overall I think you did a fine job your first time out. Now, can you leave it alone and enjoy it, or are you going to spend days/weeks/forever tweaking a little bit here and there?
 

·
Registered
Joined
·
3,248 Posts
Discussion Starter · #6 ·

Quote:
Originally Posted by glaufman /forum/post/18156287




I agree, it looks very nice.


For choosing which color temp to start with, you might do better looking at the RGB graph.


Since you're having trouble with the color decoding, you may try forcing Y'CbCr or RGB and see if it clears anything up, or makes anything worse.


Hmmm. auto dimming off, is there anything else called "auto contrast" or "contrast enhancement" or the like? Yes, it looks like it may have messed with your 0% reading, which BTW, since I believe you're an engineer, the term IRE should not be used with digital video.


0% and 10% are the hardest parts for a display to do properly. You may notice that they never show you those points in magazine reviews. Did you just do balancing at 30% and 80%? You can try tweaking it at 20%-80%, take a full grayscale set, see if it improves... tweak at 10%-80%, take a grayscale set, see if it improves... Some meters are not to be trusted at these low light levels... in which case they sometimes suggest trusting your eyes, as long as you're looking at a steps pattern (your eyes are no good at looking at a single field and saying "it's right" but their very good at seeing "that bar is bluer than the one next to it...)... I think the I1D2 you bought is supposed to be ok at low light levels, but I may be mistaken... it may only be good down that low for absolute luminance, not color balance... sorry, but I don't know...


The I1D2 isn't (I believe) considered good for CMS work anyway, so no foul that you don't have one... it does look like your secondaries are off, I'm not sure how good the I1D2 is at doing that... how does it look to your eyes? You may be better off using the filters for that, especially if they're not leaking...


Well, let's say it may not be abnormal. Absolute fL don't really matter all that much. If you get color shift or clipping, you turn down contrast. If you're getting eyestrain, you turn down the backlight. Either of those could mess with your grayscale though.


Color leaking? Not sure exactly what you mean. BUT... it's not uncommon for your first look at a calibrated pattern to seem, um, strange... some people complain of yellow, or red, or warmness... you've been looking at a blue picture for so long it takes a little time for your eyes to adjust...


Overall I think you did a fine job your first time out. Now, can you leave it alone and enjoy it, or are you going to spend days/weeks/forever tweaking a little bit here and there?

Thanks for the feedback, but please don't make me do this.


I want to stop as soon as practical and possible.


The color situation does bug me, though. I had to move the Color control from 50 to nearly 70 to get 100% Red to give 21% of 100% White. Adjusting any control by such a large amount doesn't sound too good to me.

On the other hand, I am not sure I have a better alternative.
 

·
Registered
Joined
·
1,011 Posts
You can stop any time you like. It matters not what the color control is set at, if it measures/reads correctly. Since I'm not sure your meter is good for this purpose, How do reds look?
 

·
Registered
Joined
·
591 Posts
Yes, the i1 is okay but dont put total emphasis on getting it 100% perfect. That meter may or may not be good enough to balance rgb good enough. I compared one to chroma 5 and the grey scale is way more balanced than the i1 could ever be. Its not bad but just know it has limitations. Go for close AND looks good. If you need to error a little to fix something then its probably right.
 

·
Registered
Joined
·
5,884 Posts

Quote:
Originally Posted by Dralt /forum/post/18156538


Thanks for the feedback, but please don't make me do this.


I want to stop as soon as practical and possible.


The color situation does bug me, though. I had to move the Color control from 50 to nearly 70 to get 100% Red to give 21% of 100% White. Adjusting any control by such a large amount doesn't sound too good to me.

On the other hand, I am not sure I have a better alternative.

Without CMS what I did was adjust color and hue that yielded the least avg delta E. If you make adjustments to make Red match 100%, your other colors will be way off. By the way you're getting poor delta E on the 0 to 20 IRE because the meter isn't capable of reading accurately when material is that dark.
 

·
Registered
Joined
·
3,248 Posts
Discussion Starter · #10 ·

Quote:
Originally Posted by jkcheng122 /forum/post/18166703


Without CMS what I did was adjust color and hue that yielded the least avg delta E. If you make adjustments to make Red match 100%, your other colors will be way off. By the way you're getting poor delta E on the 0 to 20 IRE because the meter isn't capable of reading accurately when material is that dark.

Thanks.


Very true. If I make Red match 100% with the standard, Green and Blue deviate more significantly.


Is it more important to get Red right? Why does the CurtPalme.com method recommends to start with Red?


I think I'll do what you said. I will reset Color and Hue to their default values. Then, I will tune Color until I get the lowest average deltaE for Red, Green and Blue.


After that, I will tune the Hue control until I get the lowest average deltaE for secondaries.


Although, what I want is probably not the lowest average, but the lowest difference between max and min deltaE.


As for my deltaE reading at IRE 0, it's almost certainly that nasty auto-dimming feature, which dims the backlight when I display the video black pattern. It is apparently a lame trick Sony introduced when lots of people started to complain about the flash-lighting problem.
 
1 - 10 of 10 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top