Please explain contrast ratios and black level. - AVS Forum
Forum Jump: 
 2Likes
  • 2 Post By Doug Blackburn
  • 1 Post By Doug Blackburn
  • 1 Post By GeorgeAB
  • 1 Post By Michael TLV
 
Thread Tools
post #1 of 26 Old 04-29-2013, 06:10 AM - Thread Starter
Member
 
tucsondave's Avatar
 
Join Date: Nov 2010
Posts: 185
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 13
I see posts saying this projector has better blacks than that projector.
Then manufacturers claim contrast ratios of 4000:1, 20000:1, googleplex and a half to one.
If Blu Ray black level is 16 and white is 235 where do all these other numbers fit in?

thanks
tucsondave is offline  
Sponsored Links
Advertisement
 
post #2 of 26 Old 04-29-2013, 06:26 AM
Advanced Member
 
realzven's Avatar
 
Join Date: Mar 2010
Posts: 519
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 84
the contrast ratio is the delta between the MLL (aka Miminum Luminance level) and the peak white, for example :

MLL of the display 0,001 ft/L, peak white 40 ft/L => contraste ratio is 40.000:1 (40/0.001=40.000)

the dymamic range is another story : reference black is "video level 16" and reference white "is video level 235", it's two different things
realzven is offline  
post #3 of 26 Old 04-29-2013, 08:44 AM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,640
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 44 Post(s)
Liked: 175
Quote:
Originally Posted by realzven View Post

the contrast ratio is the delta between the MLL (aka Miminum Luminance level) and the peak white, for example :

MLL of the display 0,001 ft/L, peak white 40 ft/L => contraste ratio is 40.000:1 (40/0.001=40.000)

the dymamic range is another story : reference black is "video level 16" and reference white "is video level 235", it's two different things

Not to be pedantic, but terminology can be very important. Contrast ratio is calculated between black (16) and reference white (235), not peak white (255) in the video world.

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is offline  
post #4 of 26 Old 04-29-2013, 01:39 PM
Advanced Member
 
realzven's Avatar
 
Join Date: Mar 2010
Posts: 519
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 84
Sorry for my bad english but With the words "peak white" i was thinking in reality "reference white" at video level 235 with no discoloration, no color tinted, no eye strain and at 35 ftL for a flat panel display depending the environnement
realzven is offline  
post #5 of 26 Old 04-29-2013, 02:57 PM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,640
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 44 Post(s)
Liked: 175
Quote:
Originally Posted by realzven View Post

Sorry for my bad english but With the words "peak white" i was thinking in reality "reference white" at video level 235 with no discoloration, no color tinted, no eye strain and at 35 ftL for a flat panel display depending the environnement

No worries, just wanted to make sure it will be clear to anyone who finds this thread in the future.

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is offline  
post #6 of 26 Old 04-29-2013, 04:48 PM - Thread Starter
Member
 
tucsondave's Avatar
 
Join Date: Nov 2010
Posts: 185
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 13
Ok, so if black is 16 which isn't black why do people get crazy if the display has a background "glow" of 10-15. I don't remember any fuss about the "video gray" on CRT's under NTSC specs.
And how do "they" come up with a contrast ratio of 20000: 1 when dealing with a white to black ratio of 16:235?
tucsondave is offline  
post #7 of 26 Old 04-29-2013, 05:51 PM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,640
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 44 Post(s)
Liked: 175
Quote:
Originally Posted by tucsondave View Post

Ok, so if black is 16 which isn't black why do people get crazy if the display has a background "glow" of 10-15. I don't remember any fuss about the "video gray" on CRT's under NTSC specs.
And how do "they" come up with a contrast ratio of 20000: 1 when dealing with a white to black ratio of 16:235?

16 is suppose to be aligned with the minimum output of the display, it should not be gray. Ideally you would get no light output from level 16 and level 17 would start outputting light.

CRT's in the US used a 7.5 IRE pedestal, meaning 7.5 IRE was suppose to output the minimum light for the display, which directly correlates with the use of digital level 16 as video black. The use of headroom and footroom is long standing and is a by product of the old NTSC days, when this stuff actually went over the air as analog radio broadcasts.

The contrast ratio is calculated from 100% and 0%, the encoded numbers behind that are abstracted in the formulas.

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is offline  
post #8 of 26 Old 04-29-2013, 07:01 PM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,457
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 25 Post(s)
Liked: 228
The OP is confused in the extreme.

16-235 is digital values for video... there are no units.

When you MEASURE how black (or white) a TV is with a meter, you are going to measure some UNITS OF LIGHT like cd/m2 (metric) or fL (foot-Lamberts),

The thing is technology is not perfect. If the TV was perfect, 16 would ALWAYS produce total blackness from every video display (assuming you were viewing consumer video sources where black is represented by 16 and 100% white I represented by 235. But TVs always have some residual amount of light they reproduce when the screen is creating what passes for black. A few TVs (some Pioneer Kuro plasma models, not made for several years now or some brand new Panasonic plasma models) have gotten very very very close to digital 16 being truly black, but many TVs still have a visible dark gray displayed instead of black.

Let's say you measure digital level 16 (and the TV is correctly adjusted so 16 really is as black as the TV ever gets) and you measure 0.001 fL (foot-Lamberts) and you measure 100% white at 30 fL. The contrast ratio is 30 divided by 0.001, your contrast ratio would be 30,000:1

Images just plain look better and more lifelike with higher contrast ratios (as long as 100% white isn't so bright that it causes you to squint which leads to eye fatigue over the long haul.

Let's say somebody makes a PERFECT video display and digital level 16 produces total black from the screen... no measurable or visible light output from the TV when it is displaying black. In that case, the contrast ratio would be infinite because whatever level you use for 100% white will be divided by zero and anything divided by zero is infinity. A calculator will tell you that you cannot divide by 0, but any number divided by zero really results in infinity being the answer. Infinity is a valid mathematical concept and dividing by zero is an equally valid concept. Divide 30 by .001 to get 30,000. If you divide 30 by .00001, you get 300,000... make the divisor smaller and smaller and the result gets larger and larger. Zero is the "smallest" number and produces the larges result... infinity. So a "perfect" TV with truly black blacks will have an infinite contrast ratio no matter how bright or dim you set 100% white.

When it comes to projectors, manufacturers do all kinds of trickery to result in higher-than-real-world numbers. I've heard of some cases where the projector is placed so far from the screen, that a single pixel becomes large enough to measure with a meter (generally, your meter will measure some significant number of pixels, at least several thousand of the more than 2 million pixel triplets (each pixel has individual red, green, and blue pixels producing the light representing a single pixel, in projectors, those Red, green, and blie pixels are stacked on top of each other, in flat panel displays, the 3 colors are side-by-side for crt or plasma and on top of each other for LCD) that make an HD image frame. At any rate, the trick used by some projector manufacturers is to measure just ONE pixel so they get all the light and none of the dark area between pixels. That method over-states the real-world contrast ratio and often leads to the 6-digit or higher contrast ratio claims you see today.

Black level is more important to contrast ratio than the 100% white level. Example:

100% white is 100 fL and black measures 1 fL. Contrast ratio is 100:1
If 100% white s 200 fL and black is 1 fL the contrast ratio is 200:1
But if 100% white is 100 fL and black is 0.5 fL, contrast ratio is 200:1
So increasing white by 100 fL produces the same contrast ratio as reducing black by 0.5 fL... obviously black level has more of an effect on contrast ratio calculations than white level.

Depending on where you read it, human vision is capable of only about 400:1 to 800:1 contrast ratio in any one scene. But the world around us has a contrast ratio that is infinity though there probably aren't very many people who have experienced an infinite contrast ratio... even in the middle of nowhere on a moonless night, the light from stars will create a contrast ratio lower than infinity. To experience an infinite contrast ratio, you have to be in a laboratory that is capable of total blackness (much like an anechoic chamber for acoustic research attempts to produce total silence, it's really difficult), though total blackness is easier to make than total silence.

Anyway, images where black is actually dark gray instead of black look milky and lack dimension. The blacker you make the blacks, the more dimensional and realistic the images will look. That's why we like displays that produce very dark blacks rather than being obviously gray.

You can setup a video display to produce its darkest black at digital level 16 as required for consumer video sources, or you can also setup a video display to produce its blackest black for digital level 0... which is what computers do... computers operate in the 0-255 realm. If you view consumer video on a computer setup for 0-255, the consumer video source will never have a digital level below 16 (there can be occasional mistakes with numbers lower than 16 but those are mistakes not actual information). So consumer video played on a computer setup for 0-255 will never create images with true black (or as close to black as the computer display can get). Sometimes the computer video system is smart enough to know the incoming video is consumer video in the 16-235 format and the computer will switch to that mode automatically for you, but some computer video systems don't change automatically, you have to find a setting somewhere and change it for the consumer video you want to view... if you want the images to look pretty good.

Let's say the computer monitor produces 0.0001 fL for digital level 0 (zero). When you change the computer video system to work with consumer video and your lowest digital level will be 16 and you want 16 to produce the blackest black, when you have setup the computer video system to produce its best black for digital level 16, you'll still measure 0.0001 for digital level 16... just as black as digital level 0. It's all in the settings... either in the computer or in the TV (and if you can't change the computer to show consumer video correctly, you may be able to change a setting in the TV so that the TV goes into 0-255 mode).

[Edited to fix a math error]
PlasmaPZ80U and Tony_Montana like this.

"Movies is magic..." Van Dyke Parks
THX -- ISF -- HAA
Doug Blackburn is offline  
post #9 of 26 Old 04-30-2013, 03:09 PM
Advanced Member
 
hungro's Avatar
 
Join Date: Mar 2007
Posts: 628
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 27 Post(s)
Liked: 61
Quote:
Originally Posted by Doug Blackburn View Post

The OP is confused in the extreme.

16-235 is digital values for video... there are no units.

When you MEASURE how black (or white) a TV is with a meter, you are going to measure some UNITS OF LIGHT like cd/m2 (metric) or fL (foot-Lamberts),

The thing is technology is not perfect. If the TV was perfect, 16 would ALWAYS produce total blackness from every video display (assuming you were viewing consumer video sources where black is represented by 16 and 100% white I represented by 235. But TVs always have some residual amount of light they reproduce when the screen is creating what passes for black. A few TVs (some Pioneer Kuro plasma models, not made for several years now or some brand new Panasonic plasma models) have gotten very very very close to digital 16 being truly black, but many TVs still have a visible dark gray displayed instead of black.

Let's say you measure digital level 16 (and the TV is correctly adjusted so 16 really is as black as the TV ever gets) and you measure 0.001 fL (foot-Lamberts) and you measure 100% white at 30 fL. The contrast ratio is 30 divided by 0.001, your contrast ratio would be 30,000:1

Images just plain look better and more lifelike with higher contrast ratios (as long as 100% white isn't so bright that it causes you to squint which leads to eye fatigue over the long haul.

Let's say somebody makes a PERFECT video display and digital level 16 produces total black from the screen... no measurable or visible light output from the TV when it is displaying black. In that case, the contrast ratio would be infinite because whatever level you use for 100% white will be divided by zero and anything divided by zero is infinity. A calculator will tell you that you cannot divide by 0, but any number divided by zero really results in infinity being the answer. Infinity is a valid mathematical concept and dividing by zero is an equally valid concept. Divide 30 by .001 to get 30,000. If you divide 30 by .00001, you get 300,000... make the divisor smaller and smaller and the result gets larger and larger. Zero is the "smallest" number and produces the larges result... infinity. So a "perfect" TV with truly black blacks will have an infinite contrast ratio no matter how bright or dim you set 100% white.

When it comes to projectors, manufacturers do all kinds of trickery to result in higher-than-real-world numbers. I've heard of some cases where the projector is placed so far from the screen, that a single pixel becomes large enough to measure with a meter (generally, your meter will measure some significant number of pixels, at least several thousand of the more than 2 million pixel triplets (each pixel has individual red, green, and blue pixels producing the light representing a single pixel, in projectors, those Red, green, and blie pixels are stacked on top of each other, in flat panel displays, the 3 colors are side-by-side for crt or plasma and on top of each other for LCD) that make an HD image frame. At any rate, the trick used by some projector manufacturers is to measure just ONE pixel so they get all the light and none of the dark area between pixels. That method over-states the real-world contrast ratio and often leads to the 6-digit or higher contrast ratio claims you see today.

Black level is more important to contrast ratio than the 100% white level. Example:

100% white is 100 fL and black measures 1 fL. Contrast ratio is 100:1
If 100% white s 200 fL and black is 1 fL the contrast ratio is 200:1
But if 100% white is 100 fL and black is 0.5 fL, contrast ratio is 500:1
So obviously, it is much better for contrast ratio to cut the black level by 1/2 than it is to double the white level.

Depending on where you read it, human vision is capable of only about 400:1 to 800:1 contrast ratio in any one scene. But the world around us has a contrast ratio that is infinity though there probably aren't very many people who have experienced an infinite contrast ratio... even in the middle of nowhere on a moonless night, the light from stars will create a contrast ratio lower than infinity. To experience an infinite contrast ratio, you have to be in a laboratory that is capable of total blackness (much like an anechoic chamber for acoustic research attempts to produce total silence, it's really difficult), though total blackness is easier to make than total silence.

Anyway, images where black is actually dark gray instead of black look milky and lack dimension. The blacker you make the blacks, the more dimensional and realistic the images will look. That's why we like displays that produce very dark blacks rather than being obviously gray.

You can setup a video display to produce its darkest black at digital level 16 as required for consumer video sources, or you can also setup a video display to produce its blackest black for digital level 0... which is what computers do... computers operate in the 0-255 realm. If you view consumer video on a computer setup for 0-255, the consumer video source will never have a digital level below 16 (there can be occasional mistakes with numbers lower than 16 but those are mistakes not actual information). So consumer video played on a computer setup for 0-255 will never create images with true black (or as close to black as the computer display can get). Sometimes the computer video system is smart enough to know the incoming video is consumer video in the 16-235 format and the computer will switch to that mode automatically for you, but some computer video systems don't change automatically, you have to find a setting somewhere and change it for the consumer video you want to view... if you want the images to look pretty good.

Let's say the computer monitor produces 0.0001 fL for digital level 0 (zero). When you change the computer video system to work with consumer video and your lowest digital level will be 16 and you want 16 to produce the blackest black, when you have setup the computer video system to produce its best black for digital level 16, you'll still measure 0.0001 for digital level 16... just as black as digital level 0. It's all in the settings... either in the computer or in the TV (and if you can't change the computer to show consumer video correctly, you may be able to change a setting in the TV so that the TV goes into 0-255 mode).

Depending on where you read it, human vision is capable of only about 400:1 to 800:1 contrast ratio in any one scene. Can you elaborate on this further. Meaning we cannot distinguish the difference between 1000:1 or 20,000:1
hungro is offline  
post #10 of 26 Old 04-30-2013, 03:26 PM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,640
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 44 Post(s)
Liked: 175
Quote:
Originally Posted by hungro View Post

Depending on where you read it, human vision is capable of only about 400:1 to 800:1 contrast ratio in any one scene. Can you elaborate on this further. Meaning we cannot distinguish the difference between 1000:1 or 20,000:1

That's precisely what he's saying, not that you might not get better picture quality though with the better contrast ratio though.

Imagine the reverse of a 2% window size 100% intensity pattern. A small black square surrounded by 100% white at 100 candelas. Most people would not be able to perceive a difference in the black level if you had two displays side by side one with black at 0.25 candelas (400:1), and one with the black at 0.1 candelas(1000:1).

That doesn't mean there are a variety of circumstances in any given film where you have scenes, where there should be black, but the brightest object in the scene isn't even close to 100%, or that it's not dominate enough to affect your perception of black level. Having great black levels, is very important to getting a good picture and you can certainly tell the difference between those kinds of contrast ratios with certain content.

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is offline  
post #11 of 26 Old 04-30-2013, 05:28 PM - Thread Starter
Member
 
tucsondave's Avatar
 
Join Date: Nov 2010
Posts: 185
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 13
If I may confuse things even more.
I have some background in photography and am aware of both general and local brightness and color adaptation.
The local brightness adaptation being why we can see into shadows in bright sunlight and cameras can not. So I am wondering what that eye/brain effect has on
viewing bright video images in a dark environment.

On the contrast ratio question, since black to white on a PC would have 256 steps, the higher the contrast ratio, then the larger the brightness difference would be between say 127 and 128?
If you ignore the auto iris and lamp dimming contrast ratio marketing, would there be a point where a contrast ratio could be too high?
smile.gif
tucsondave is offline  
post #12 of 26 Old 04-30-2013, 06:09 PM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,640
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 44 Post(s)
Liked: 175
Quote:
Originally Posted by tucsondave View Post

If I may confuse things even more.
I have some background in photography and am aware of both general and local brightness and color adaptation.
The local brightness adaptation being why we can see into shadows in bright sunlight and cameras can not. So I am wondering what that eye/brain effect has on
viewing bright video images in a dark environment.
Viewing environment has a huge role in our perception of color. That's why there are specific recommendation for ambient light and bias lighting for content creation.
Quote:
Originally Posted by tucsondave View Post

On the contrast ratio question, since black to white on a PC would have 256 steps, the higher the contrast ratio, then the larger the brightness difference would be between say 127 and 128?
If you ignore the auto iris and lamp dimming contrast ratio marketing, would there be a point where a contrast ratio could be too high?
smile.gif

You might be getting confused here about the difference between contrast ratio and dynamic range.

100%=100, 0% = 0.1 is 1000:1 contrast ratio
100%=100, 0% =0.001 is 100,000:1 contrast ratio
but
100%=100, 0% = 0.1 is a dynamic range of 99.9
100%=100, 0% =0.001 is a dynamic range of 99.999
That's almost no difference at all, so the step size between 127 and 128 would be functionally equivalent.

Now if you had a display that could output 1500 candelas and had a black level of 1 candela, now everything changes. Now your dynamic range is 1499 candelas and you'll need much greater bit depth than 8-bit's 256 steps can provide in order to provide smooth gradients at reasonable light output.

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is offline  
post #13 of 26 Old 05-01-2013, 01:37 PM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,457
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 25 Post(s)
Liked: 228
Quote:
Originally Posted by hungro View Post

Depending on where you read it, human vision is capable of only about 400:1 to 800:1 contrast ratio in any one scene. Can you elaborate on this further. Meaning we cannot distinguish the difference between 1000:1 or 20,000:1

The iris in our eyes can only have one opening size at any one time. If the scene is mostly bright with only a little darkness in it, brightness will win and our iris will close to keep the perceived light level comfortable... but that will darken the shadow areas and we won't "see into the shadows"... but if we look away from the bright area of the scene and focus only on the darker area, our iris will open farther to allow us to see shadow detail easily, though the bright area will be uncomfortably bright if it attracts our attention... at least it will be uncomfortably bright until our iris closes enough to make it comfortable again. It's the action of our iris that allows us to tell that 100,000:1 contrast ratio makes better-looking images than 1000:1 contrast ratio.

But when you look at one scene with an "average" luminance level that keeps our iris at a fixed opening (say half way between fully open and fully closed), we can only perceive somewhere between around 400:1 and 800:1 (depending on your reference source) contrast.

Average Picture Level is a way to define how bright images are on a projector or flat panel TV. An APL of 100% could only be achieved if you displayed a 100% white pattern that filled the entire screen. A 0% APL could only be achieved on a perfect video display that produced ZERO light when it displayed 0% white (AKA black)... all the video displays we have available today are below 1% APL when displaying a full-screen black pattern, but they are also somewhere above 0% APL because their blacks aren't perfectly black. If you display an image with an APL of 10%, we can see 400:1 to 800:1 contrast within that image. If you display an image with an 80% APL we will still see THAT image with contrast of 400:1 - 800:1 because our iris attempts to keep dark scenes bright enough to see detail and bright scenes dim enough to not cause discomfort (i.e. squinting, shading eyes, blinking uncontrollably, etc.).

Bottom line... we can see over a very wide range of luminance levels, but not all at the same time. Our eyes have to adapt (via the iris opening changing sizes). This is most easily observed when you turn off bright lighting in a room and it becomes very very dark all at once. As you stand there waiting, the room changes from being so dark you can't see anything to being able to make out at least shapes and sizes of objects as you are in the dark environment longer and longer. And when a flash goes off in a dark room, we're essentially blinded momentarily until our iris opens again. So we WANT video displays with high contrast ratios and the higher the better... infinity would be great! We just can't see black and near black (i.e. shadows) and white and near-white (i.e. highlights) clearly all at the same time in the same scene.
h2c likes this.

"Movies is magic..." Van Dyke Parks
THX -- ISF -- HAA
Doug Blackburn is offline  
post #14 of 26 Old 05-01-2013, 02:57 PM
AVS Special Member
 
GeorgeAB's Avatar
 
Join Date: Feb 2002
Location: Denver, CO
Posts: 3,320
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 21 Post(s)
Liked: 110
Quote:
Originally Posted by tucsondave View Post

If I may confuse things even more.
I have some background in photography and am aware of both general and local brightness and color adaptation.
The local brightness adaptation being why we can see into shadows in bright sunlight and cameras can not. So I am wondering what that eye/brain effect has on
viewing bright video images in a dark environment.

On the contrast ratio question, since black to white on a PC would have 256 steps, the higher the contrast ratio, then the larger the brightness difference would be between say 127 and 128?
If you ignore the auto iris and lamp dimming contrast ratio marketing, would there be a point where a contrast ratio could be too high?
smile.gif
Any discussion of the theoretical limits of the human visual system relating to video performance must take into consideration the effect of time and motion when viewing video displays. Dark/light adaptation can take up to a half hour or so to complete. This also goes beyond iris diameter size. There are electrochemical changes that occur in the retina over time that are much slower than the response speed of the iris. Ambient illumination level also effects how we perceive a video image. Program mastering technicians are encouraged to allow for about a half hour to adapt to the darker conditions typically maintained in their work spaces.

Best regards and beautiful pictures,
G. Alan Brown, President
CinemaQuest, Inc.
A Lion AV Consultants affiliate

"Advancing the art and science of electronic imaging"
h2c likes this.
GeorgeAB is offline  
post #15 of 26 Old 05-01-2013, 09:23 PM
Member
 
3dmaven's Avatar
 
Join Date: Oct 2011
Posts: 111
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 11
Remember too that there is a difference between ANSI and ON/OFF contrast. Almost any contrast measurement you see listed will be ON/OFF (the difference between a pure black screen and a pure white screen), and there are some projectors that even have an infinity:1 rating in that area (high end LED ones). All the projectors I am aware of are sub 1000:1 in ANSI contrast (the difference between black and white when displayed at the same time, using a checkerboard pattern). It's the ANSI contrast that is comparable to the way your eyes perceive contrast. ON/OFF contrast is generally meaningless (a projector with 100,000:1 ON/OFF contrast may have lower ANSI contrast than a projector with 10,000:1 ON/OFF), but ON/OFF seems (to me at least) to stand in as a decent indicator of how good the black level will be. Good luck getting ANSI contrast numbers though, they are notoriously hard to measure from what I understand.
3dmaven is offline  
post #16 of 26 Old 11-17-2013, 11:58 AM
 
MonarchX's Avatar
 
Join Date: Jul 2012
Posts: 669
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 30
Is it true that properly calibrated TV using 16-235 range may provide a higher contrast ratio than the same set properly calibrated using 0-255 range?
MonarchX is offline  
post #17 of 26 Old 11-17-2013, 12:26 PM
 
MonarchX's Avatar
 
Join Date: Jul 2012
Posts: 669
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 30
I re-calibrated my PC and TV to use 16-235. The contrast ratio is exactly the same, BUT my blacks seem deeper! I do not have a colorimeter good enough to properly measure black levels, so it could be a placebo. Is there any info on that? Does well calibrated 16-235 range TV provide better blacks than the same TV but calibrated to use 0-255?
MonarchX is offline  
post #18 of 26 Old 11-17-2013, 01:08 PM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,640
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 44 Post(s)
Liked: 175
Quote:
Originally Posted by MonarchX View Post

Is it true that properly calibrated TV using 16-235 range may provide a higher contrast ratio than the same set properly calibrated using 0-255 range?

What would change?

If you properly calibrate your brightness control your TV's black is as black as it can be, it can't get any darker, so you can't lower the black level.

For the white level properly calibrated video includes some headroom, but if you compare peak white to peak white, once again it will be the maximum output of the TV.

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is offline  
post #19 of 26 Old 11-17-2013, 01:16 PM
Advanced Member
 
spacediver's Avatar
 
Join Date: May 2013
Location: Toronto
Posts: 943
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 120 Post(s)
Liked: 75
Quote:
Originally Posted by sotti View Post

16 is suppose to be aligned with the minimum output of the display, it should not be gray. Ideally you would get no light output from level 16 and level 17 would start outputting light.

CRT's in the US used a 7.5 IRE pedestal, meaning 7.5 IRE was suppose to output the minimum light for the display, which directly correlates with the use of digital level 16 as video black. The use of headroom and footroom is long standing and is a by product of the old NTSC days, when this stuff actually went over the air as analog radio broadcasts.

I'm bumping up against this issue here. Not sure how to proceed.
spacediver is online now  
post #20 of 26 Old 11-17-2013, 01:29 PM
AVS Special Member
 
Michael TLV's Avatar
 
Join Date: Feb 2000
Location: THX/ISF Calibrationist/Instructor, Calgary, AB, Canada
Posts: 6,774
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 28 Post(s)
Liked: 138
Greetings

Yes he can get better black levels ... it's called turning the TV off. Now it is as good as it gets. Invite your friends over to see the black levels. biggrin.gif

regards
PE06MCG likes this.

Michael Chen @ The Laser Video Experience
ISF/THX/TLV Video Instructor
The Video Calibration Education Hub - www.TLVEXP.com

Michael TLV is offline  
post #21 of 26 Old 11-18-2013, 05:07 PM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,457
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 25 Post(s)
Liked: 228
If the TV does not have a design flaw, YCbCr will always be 16-235 and RGB will have a setting that allows you to select 0-255 or 16-235. Again, unless the TV has a design flaw, both video levels will produce the same black level (or minimum light level). The black level has more effect on Contrast ratio than the white level so getting the black level as low as the display will go is what you want for best contrast ratio.

For the white level, if you pick 35 fL for 100% white, you SHOULD get 35 fL for both 16.235 or 0-255 if, again, the TV doesn't have a design problem. You cannot assume there is not a design problem though. If this is critical for your application, you will have to measure all the conditions that are important to you to confirm or prove that there is a design problem you have to work around some way.

You will not get a different contrast ratio from 16-235 or 0-255 unless the TV has a design problem and unless you have a higher 100% white setting for one of the two modes.

"Movies is magic..." Van Dyke Parks
THX -- ISF -- HAA
Doug Blackburn is offline  
post #22 of 26 Old 11-18-2013, 05:10 PM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,457
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 25 Post(s)
Liked: 228
If OLED technology turns out to be durable and is a worthy replacement for LCD and plasma, the MLL will be zero and contrast ratios will be infinite, just as they are in the real world. So on an OLED display, just the fact that black is "zero" foot-Lamberts means contrast ratio will be infinity all the time even if your 100% white is just 1 fL (very very dim). So once we have a display technology that produces true black levels, all the talk about contrast ratio will seem silly. All we will be interested in then is how bright the technology can make white and whether that's enough or not enough for quality viewing.

"Movies is magic..." Van Dyke Parks
THX -- ISF -- HAA
Doug Blackburn is offline  
post #23 of 26 Old 11-18-2013, 05:15 PM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 7,291
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 112 Post(s)
Liked: 204
Quote:
Originally Posted by Doug Blackburn View Post

If the TV does not have a design flaw, YCbCr will always be 16-235 and RGB will have a setting that allows you to select 0-255 or 16-235. Again, unless the TV has a design flaw, both video levels will produce the same black level (or minimum light level). The black level has more effect on Contrast ratio than the white level so getting the black level as low as the display will go is what you want for best contrast ratio.

For the white level, if you pick 35 fL for 100% white, you SHOULD get 35 fL for both 16.235 or 0-255 if, again, the TV doesn't have a design problem. You cannot assume there is not a design problem though. If this is critical for your application, you will have to measure all the conditions that are important to you to confirm or prove that there is a design problem you have to work around some way.

You will not get a different contrast ratio from 16-235 or 0-255 unless the TV has a design problem and unless you have a higher 100% white setting for one of the two modes.

the PS4 has Full and Limited options for YCbCr output; however, I tested full and it is still video levels and includes BTB/WTW... not sure if limited is clipping BTB/WTW, as I didn't try it
PlasmaPZ80U is offline  
post #24 of 26 Old 11-18-2013, 05:35 PM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,457
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 25 Post(s)
Liked: 228
Usually when that happens, the levels do not actually change when you have selected YCbCr... the control is there for RGB but it isn't grayed out when you select YCbCr and you can change it. I've never bothered checking on the PS3 whether the Full Limited option actually does anything when you are in YCbCr mode, but it is not SUPPOSED to do anything since YCbCr should always be 16-235 regardless of other settings. I've always gotten proper black levels from a PS3 set to YCbCr so there was never a reason to check on the full-limited thing.

"Movies is magic..." Van Dyke Parks
THX -- ISF -- HAA
Doug Blackburn is offline  
post #25 of 26 Old 11-19-2013, 09:12 AM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 7,291
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 112 Post(s)
Liked: 204
Quote:
Originally Posted by Doug Blackburn View Post

Usually when that happens, the levels do not actually change when you have selected YCbCr... the control is there for RGB but it isn't grayed out when you select YCbCr and you can change it. I've never bothered checking on the PS3 whether the Full Limited option actually does anything when you are in YCbCr mode, but it is not SUPPOSED to do anything since YCbCr should always be 16-235 regardless of other settings. I've always gotten proper black levels from a PS3 set to YCbCr so there was never a reason to check on the full-limited thing.

to clarify, there's actually one control for RGB and a second for YCbCr... I set former to limited and the latter to full
PlasmaPZ80U is offline  
post #26 of 26 Old 11-21-2013, 07:13 AM
Member
 
Miguelronda's Avatar
 
Join Date: Sep 2010
Posts: 59
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 26 Post(s)
Liked: 24
Please can you indicate the correct way to measure the black level on a JVC projector to apply BT.1886?

Thank you
Miguelronda is offline  
Reply Display Calibration

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off