AVS Forum banner
1 - 20 of 50 Posts

·
Registered
Joined
·
199 Posts
Discussion Starter · #1 ·
I am beginning to better grasp the objective processes of calibrating gamma, grayscale, and color. What I find interesting is that now what used to be the much simpler process of adjusting the basic settings such as contrast, may not be so simple after all. There are a number of calibration screens which I am just now beginning to understand. These include things like clipping patterns and ramps and so on. These screens rely on subjective interpretation of the patterns. For example, if I have calibrated grayscale and afterwords appreciate a red tint to the contrast/clipping pattern, then the contrast is likely set too high.


I have seen some suggest using the 109 IRE field and turning up contrast to see when a color runs out. I just don't seem to see much of a difference in doing this. I am concerned that my subjective interpretation of what I am supposed to be looking for may be flawed.


I do see some red tinting of the white clipping pattern bars when the contrast is turned up. If I start at a lower contrast setting and turn the contrast up, I will arrive at the correct setting. If that one setting is correct and turning up the contrast one tick gives a very slight reddish tint, I may not be able to appreciate it at this point even though this is an incorrect setting. Turning it up further until it is apparent may be several ticks above the correct setting. So what is the correct way to arrive at the proper setting? Any other related tips would be appreciated.


I am still having a good time learning from all of you more experienced calibrators. I appreciate your helping us home enthusiasts to continue to improve.


-Mark
 

·
Registered
Joined
·
6,723 Posts
Greetings


One more time ...



Three rules for contrast ...


1. No clipping

2. No Discoloration

3. Eye fatigue factors


That's how contrast is set.


regards
 

·
Registered
Joined
·
7,028 Posts
Quote:
Originally Posted by Michael TLV
Greetings


One more time ...



Three rules for contrast ...


1. No clipping

2. No Discoloration

3. Eye fatigue factors


That's how contrast is set.


regards


That's easy for you to say.



1) When you say "no clipping", do you mean while using a clipping test, or the contrast test? I have the S&M disc and the AVS-709 test patterns. When I do the clipping test with S&M, it shows it's clipping red. To get any boxes in the red to appear, I have to reduce contrast to (-8). Does that mean (-8) is my correct contrast setting?


2) No discoloration when using the contrast pattern? Do I just look by eye for color changes?


Thanks.
 

·
Registered
Joined
·
6,723 Posts
Greetings


The spectracal site has a write up on Contrast 101 in their Calibration Basics section. Go read that.


And then watch the training Video that was shot for the topic of contrast.


regards
 

·
Registered
Joined
·
6,335 Posts

Quote:
Originally Posted by Michael TLV /forum/post/19607514


Greetings


The spectracal site has a write up on Contrast 101 in their Calibration Basics section. Go read that.


And then watch the training Video that was shot for the topic of contrast.


regards

How about a link..not finding this info at http://www.spectracal.com/
 

·
Registered
Joined
·
3,435 Posts

Quote:
Originally Posted by Razz1 /forum/post/19606337


I am concerned that my subjective interpretation of what I am supposed to be looking for may be flawed.

Since it sounds like you're taking measurements, you can use a meter to look at discoloration. One item I like about ColorHCFR is the option to display RGB. Using the RGB values you might find that one of the colors in the grayscale may tend to stop increasing before the other colors. For example if you begin with a low contrast setting and do free measures on a 100% (or higher) pattern as you increase contrast, you may find that R tends to reach a point where it goes no higher, while G and B can continue to increase as contrast is turned up. Grayscale balance would generally show the same thing as the color would drop, but the ColorHCFR RGB representation makes it very clear when a color runs out because the displayed value basically stops increasing.


Note: I'm not sure what ChromaPure does, but the default Red, Green, and Blue values from CalMAN appear to be normalized to 100%, so they don't work the same as the RGB values from ColorHCFR. If you measured superwhite where 100% and 109% had a similar Red value that would tend to indicate Red is running out, but it's not as clear-cut as the ColorHCFR representation in my opinion.
 

·
Registered
Joined
·
3,435 Posts

Quote:
Originally Posted by Jim McC /forum/post/19606939


1) When you say "no clipping", do you mean while using a clipping test, or the contrast test? I have the S&M disc and the AVS-709 test patterns. When I do the clipping test with S&M, it shows it's clipping red. To get any boxes in the red to appear, I have to reduce contrast to (-8). Does that mean (-8) is my correct contrast setting?

It's a decent idea to double-check color along with grayscale. Like I mentioned before, I think the Spears & Munsil color clipping item you are referencing may be very conservative. While I do not have a copy of the disc to look at, the following quote supports my guess that I think their patterns may go higher than reference white (235). Based on prior looks at commercial video I can't come up with any clear-cut reason to calibrate for material that exceeds reference white (235). On many displays if you calibrate for maximum white (254) then reference white (235) will not be as bright.

Quote:
Originally Posted by sspears /forum/post/19278193


When people talk about 16 and 235, they are talking about the YCbCr space. You don't actually view in that space, you view in RGB. Values between 16 and 235, in YCbCr, can produce values outside of 16 and 235 when properly converted to RGB. I think this is the number one confusion today. I feel this is something we need to work on explaining better. All content has values above 235 in RGB. They are usually found in specular highlights. By clipping them, you are shifting the hue of those pixels. It may or may not be noticable. My display is calibrated to display up to 255 in RGB and 254 in YCbCr.
 

·
Banned
Joined
·
3,587 Posts
If you simply set the contrast to the highest setting with no linearity problems, you are very likely to have the Contrast set way too high (too bright) for dark room viewing.


What you really want to know is:


- How bright can the TV get without any one color "running out of gas" - on some, but not all, TVs, as you increase contrast, at some point, one color is likely to fall behind allowing the other 2 colors to get brighter. But there are a lot of new TVs where this never happens. This "brightest possible" setting is WRONG for dark room viewing - it will cause headaches. But it is worth knowing if you need a "bright room" setting.


- What Contrast setting produces comfortable viewing in a dark room (no eyestrain)? This is impossible to "eyeball" with a test pattern. If you don't have a meter, all you can do is watch a 2 hour movie and see if your eyes feel fatigued at any time. If you do have a meter, setting the 100% white level to 30-40 foot-Lamberts is where you want to be for dark room viewing. I generally will use 35 fL unless there's a reason to use a lower or higher setting. Without a meter, you never know what level is 35 fL - you can't guess at it. You're only recourse is to watch TV or a movie for a while and see if your eyes are "tired" afterwards. In severe cases, you might get a headache (100% white much too bright), but as you get closer to a reasonable level for dark room viewing, it will get harder to tell if you have eyestrain or not. Using a meter is really the only way to remove the guesswork.


It doesn't help that the instructions on most setup discs lead you to believe that the best Contrast setting is the highest setting without an off-color tint. That's just wrong.


Also... if the video display is not calibrated, any "tint" you see in steps near or above 100% white could be calibration problems and not clipping (or 'out of gas') problems. So the whole process can be very misleading without instrumentation if you are interested in having the best images your TV can produce. Your eyes can also EASILY be deceived... for example, a yellow tint to the brightest step in a Contrast pattern may not really be yellow... it could EASILY be caused by the next lower shade of white being too blue. You never notice that the next lower step is too blue - it's an optical illusion situation that you can't beat even if you KNOW the next lower step is too blue.
 

·
Registered
Joined
·
3,871 Posts

Quote:
Originally Posted by Doug Blackburn /forum/post/19608548


If you simply set the contrast to the highest setting with no linearity problems, you are very likely to have the Contrast set way too high (too bright) for dark room viewing.


What you really want to know is:


- How bright can the TV get without any one color "running out of gas" - on some, but not all, TVs, as you increase contrast, at some point, one color is likely to fall behind allowing the other 2 colors to get brighter. But there are a lot of new TVs where this never happens. This "brightest possible" setting is WRONG for dark room viewing - it will cause headaches. But it is worth knowing if you need a "bright room" setting.


- What Contrast setting produces comfortable viewing in a dark room (no eyestrain)? This is impossible to "eyeball" with a test pattern. If you don't have a meter, all you can do is watch a 2 hour movie and see if your eyes feel fatigued at any time. If you do have a meter, setting the 100% white level to 30-40 foot-Lamberts is where you want to be for dark room viewing. I generally will use 35 fL unless there's a reason to use a lower or higher setting. Without a meter, you never know what level is 35 fL - you can't guess at it. You're only recourse is to watch TV or a movie for a while and see if your eyes are "tired" afterwards. In severe cases, you might get a headache (100% white much too bright), but as you get closer to a reasonable level for dark room viewing, it will get harder to tell if you have eyestrain or not. Using a meter is really the only way to remove the guesswork.


It doesn't help that the instructions on most setup discs lead you to believe that the best Contrast setting is the highest setting without an off-color tint. That's just wrong.


Also... if the video display is not calibrated, any "tint" you see in steps near or above 100% white could be calibration problems and not clipping (or 'out of gas') problems. So the whole process can be very misleading without instrumentation if you are interested in having the best images your TV can produce. Your eyes can also EASILY be deceived... for example, a yellow tint to the brightest step in a Contrast pattern may not really be yellow... it could EASILY be caused by the next lower shade of white being too blue. You never notice that the next lower step is too blue - it's an optical illusion situation that you can't beat even if you KNOW the next lower step is too blue.

Excellent points! It's important to keep in mind that "reference" calibration theory and recommended practice must be understood in the context of human factors. "Dark room" viewing conditions are defined to include the proper ambient light level and type in the standards bodies literature. Video system alignment must always include viewing environment and human factors considerations. Both objective and subjective elements will be required. A display device cannot be properly calibrated independent of the accompanying system elements, if reference viewing is the goal.


Best regards and beautiful pictures,

G. Alan Brown, President

CinemaQuest, Inc.

A Lion AV Consultants Affiliate


"Advancing the art and science of electronic imaging"
 

·
Registered
Joined
·
199 Posts
Discussion Starter · #10 ·

Quote:
Originally Posted by alluringreality /forum/post/19607969


Since it sounds like you're taking measurements, you can use a meter to look at discoloration. One item I like about ColorHCFR is the option to display RGB. Using the RGB values you might find that one of the colors in the grayscale may tend to stop increasing before the other colors. For example if you begin with a low contrast setting and do free measures on a 100% (or higher) pattern as you increase contrast, you may find that R tends to reach a point where it goes no higher, while G and B can continue to increase as contrast is turned up. Grayscale balance would generally show the same thing as the color would drop, but the ColorHCFR RGB representation makes it very clear when a color runs out because the displayed value basically stops increasing.

This sounds like an interesting method to try. After purchasing a ColorMunki, I have been primarily using Calman. I'll give HCFR a shot with the Eye-One Display LT. I am interested to see where one color runs out of gas.

Quote:
Originally Posted by Doug Blackburn /forum/post/19608548


If you simply set the contrast to the highest setting with no linearity problems, you are very likely to have the Contrast set way too high (too bright) for dark room viewing.


What you really want to know is:


- How bright can the TV get without any one color "running out of gas" - on some, but not all, TVs, as you increase contrast, at some point, one color is likely to fall behind allowing the other 2 colors to get brighter. But there are a lot of new TVs where this never happens. This "brightest possible" setting is WRONG for dark room viewing - it will cause headaches. But it is worth knowing if you need a "bright room" setting.


- What Contrast setting produces comfortable viewing in a dark room (no eyestrain)? This is impossible to "eyeball" with a test pattern. If you don't have a meter, all you can do is watch a 2 hour movie and see if your eyes feel fatigued at any time. If you do have a meter, setting the 100% white level to 30-40 foot-Lamberts is where you want to be for dark room viewing. I generally will use 35 fL unless there's a reason to use a lower or higher setting. Without a meter, you never know what level is 35 fL - you can't guess at it. You're only recourse is to watch TV or a movie for a while and see if your eyes are "tired" afterwards. In severe cases, you might get a headache (100% white much too bright), but as you get closer to a reasonable level for dark room viewing, it will get harder to tell if you have eyestrain or not. Using a meter is really the only way to remove the guesswork

If one has a newer display where one color doesn't run out of gas at the highest contrast setting, why not set the contrast at the highest setting and just turn down the backlight to obtain 35 fl? The higher contrast setting would give a larger dynamic range and your dark room viewing recommendation would still be kept.
 

·
Registered
Joined
·
12,646 Posts

Quote:
Originally Posted by Razz1 /forum/post/19611617


If one has a newer display where one color doesn't run out of gas at the highest contrast setting, why not set the contrast at the highest setting and just turn down the backlight to obtain 35 fl? The higher contrast setting would give a larger dynamic range and your dark room viewing recommendation would still be kept.

Probably on displays that don't have backlights (or a backlight setting) like plasmas.
 

·
Banned
Joined
·
3,587 Posts

Quote:
Originally Posted by Razz1 /forum/post/19611617


If one has a newer display where one color doesn't run out of gas at the highest contrast setting, why not set the contrast at the highest setting and just turn down the backlight to obtain 35 fl? The higher contrast setting would give a larger dynamic range and your dark room viewing recommendation would still be kept.

In general, you want the Backlight set as low as possible to get the darkest blacks the TV can create. The dimmer the backlight, the darker the 0% screen will be.


You have to investigate a bit (with a meter), though. You can't just assume that the TV's response will remain linear with lower Backlight settings. Some TVs will calibrate well with the Backlight set to 40% of maximum, but it you set it to 20% of maximum, the grayscale gets so funky you can't get a good calibration any more.


But let's say you have an "ideal" TV with setting ranges of 0-100 for Backlight and Contrast. Your best images would happen with Backlight at 0 and Contrast high enough to produce a satisfying luminance level for a specific viewing condition... let's assume dark room or dark room with bias lighting. You want about 30-35 fL to avoid eyestrain (the bias light will also help reduce eyestrain). So you pick a Contrast setting and see where you are in relation to 30-35 fL, then adjust Contrast up or down to get the 30-35 fL you are looking for. Depending on how the TV works, Contrast might end up at 50 or 100... it all depends on how bright the backlight is when it is set to zero.
 

·
Registered
Joined
·
38 Posts

Quote:
Originally Posted by Razz1 /forum/post/19611617


This sounds like an interesting method to try. After purchasing a ColorMunki, I have been primarily using Calman. I'll give HCFR a shot with the Eye-One Display LT. I am interested to see where one color runs out of gas.

Provided your Eye-One Display LT is in good working order and not under reading red as a lot of them tend to do including mine. Also ColorMunki is not supported by ColorHCFR therefore you won't be able to profile it
This is why I ended up buying the enhanced ColorMunki CalMAN bundle and profile my iffy i1LT to the Colormunki.
 

·
Registered
Joined
·
199 Posts
Discussion Starter · #15 ·

Quote:
Originally Posted by Doug Blackburn /forum/post/19612586


In general, you want the Backlight set as low as possible to get the darkest blacks the TV can create. The dimmer the backlight, the darker the 0% screen will be.


But let's say you have an "ideal" TV with setting ranges of 0-100 for Backlight and Contrast. Your best images would happen with Backlight at 0 and Contrast high enough to produce a satisfying luminance level for a specific viewing condition... let's assume dark room or dark room with bias lighting. You want about 30-35 fL to avoid eyestrain (the bias light will also help reduce eyestrain). So you pick a Contrast setting and see where you are in relation to 30-35 fL, then adjust Contrast up or down to get the 30-35 fL you are looking for. Depending on how the TV works, Contrast might end up at 50 or 100... it all depends on how bright the backlight is when it is set to zero.

I have not seen this workflow suggested before where one strives to adjust the backlight down to the lowest setting that still gives a decent grayscale and then adjusting the contrast to a specified luminance. The Calman workflow has one adjust the contrast prior to adjusting the luminance. I'll give your suggestion a try and see how things look.


I have also seen where others have suggested keeping the contrast setting higher in order to achieve a wider dynamic range. It appears that you place the emphasis on a darker 0% screen. Why is this preferable?
 

·
Registered
Joined
·
12,646 Posts
I don't think a backlight of zero would be anywhere near 30fL even with contrast maxed out on a LCD. I think contrast should be set as high as possible without clipping or discoloration in the high end (post calibration) and then backlight should be set to produce 30-40fL or whatever the target is. I find my TV produces a completely unwatchable picture when backlight is 0-3 and 4-5 is just barely bright enough. 6-10 is plenty bright, though. This is with contrast in the low 90s.
 

·
Registered
Joined
·
38 Posts

Quote:
Originally Posted by PlasmaPZ80U /forum/post/19618098


I don't think a backlight of zero would be anywhere near 30fL even with contrast maxed out on a LCD. I think contrast should be set as high as possible without clipping or discoloration in the high end (post calibration) and then backlight should be set to produce 30-40fL or whatever the target is. I find my TV produces a completely unwatchable picture when backlight is 0-3 and 4-5 is just barely bright enough. 6-10 is plenty bright, though. This is with contrast in the low 90s.

My general pratice is to keep the backlight as low as possible while trying to obtain a stable greyscale. As was pointed out, the higher you set your backlight the less impact your blacks will have. On my Samsung the backlight sweetpoint is 4/10. I watch TV in a dim room. Contrast is 92/100 one click higher and then discoloration creeps in. The HD content is sharp and crisp with deep blacks where detail can be seen when it is supposed to. Like looking through a window. With this setup I'm able to achieve 33fL at 100IRE which is just right for my room. Any higher and I'll need sunglasses.
 

·
Banned
Joined
·
3,587 Posts

Quote:
Originally Posted by Razz1 /forum/post/19617880


I have also seen where others have suggested keeping the contrast setting higher in order to achieve a wider dynamic range. It appears that you place the emphasis on a darker 0% screen. Why is this preferable?

OK... beginning to feel like spoon feeding here... I said already that the highest possible Contrast setting is NOT the best Contrast setting. It's headache-city (or at least eye-strain inducing) when you are viewing movies in a dark room or even a room with bias lighting. You don't WANT the widest "dynamic range" possible. What you really mean is contrast ratio - the highest possible contrast ratio is a good thing unless whites are so bright you end up with eyestrain or headaches in a dark room - in that case you have to reduce the highest Contrast setting to something more reasonable... like the 30-35 fL I keep talking about. What you want is the highest possible contrast ratio that doesn't cause eyestrain/headaches and that's around 30-35 fL. Black level affects contrast ratio more than peak white levels so you always want to get the black level as black as it can be. What happens on the other end (peak white) is all about comfortable viewing. It doesn't really matter if a TV can reach 200 fL (exceptionally bright, brighter than the brightest LCD panel I've ever seen) because you'll never use that output level in a dark theater room (maybe it would be good if the TV was by the outdoor swimming pool). You'll still want that TV to be around 30-35 fL in a dark or bias-lit room, or maybe 35-40 fL in a bias-lit room as you can get away with a little more light if you are using a bias light.
 

·
Banned
Joined
·
3,587 Posts

Quote:
Originally Posted by Razz1 /forum/post/19617880


I have also seen where others have suggested keeping the contrast setting higher in order to achieve a wider dynamic range. It appears that you place the emphasis on a darker 0% screen. Why is this preferable?

I said already that the highest possible Contrast setting is NOT the best Contrast setting. It's headache-city (usually, or at least eye-strain inducing) when you are viewing movies in a dark room or even a room with bias lighting. You don't WANT the widest "dynamic range" possible. What you really mean is contrast ratio - the highest possible contrast ratio is a good thing unless whites are so bright you end up with eyestrain or headaches in a dark room - in that case you have to reduce the highest Contrast setting to something more reasonable... like the 30-35 fL I keep talking about. What you want is the highest possible contrast ratio that doesn't cause eyestrain/headaches and that's around 30-35 fL. Black level affects contrast ratio more than peak white levels so you always want to get the black level as black as it can be. What happens on the other end (peak white) is all about comfortable viewing. It doesn't really matter if a TV can reach 200 fL (exceptionally bright, brighter than the brightest LCD panel I've ever seen) because you'll never use that output level in a dark theater room (maybe it would be good if the TV was by the outdoor swimming pool). You'll still want that TV to be around 30-35 fL in a dark or bias-lit room, or maybe 35-40 fL in a bias-lit room as you can get away with a little more light if you are using a bias light.
 

·
Banned
Joined
·
20,735 Posts

Quote:
Originally Posted by Razz1 /forum/post/19617880


I have not seen this workflow suggested before where one strives to adjust the backlight down to the lowest setting that still gives a decent grayscale and then adjusting the contrast to a specified luminance. The Calman workflow has one adjust the contrast prior to adjusting the luminance. I'll give your suggestion a try and see how things look.


I have also seen where others have suggested keeping the contrast setting higher in order to achieve a wider dynamic range. It appears that you place the emphasis on a darker 0% screen. Why is this preferable?

White level and black level need to be maximized on a digital display (not really on a plasma and not on a CRT though).


If it is too bright then adjust irises or backlighting to net the appropriate light output. You generally don't want to bring white level down (contrast) to dim the image because you're throwing away CR. So only do that if you really have to. If you have a projector that's way too bright (rare), get a lower-gain screen, or use an ND filter if iris control is not available.
 
1 - 20 of 50 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