The contrast ratio is the ratio between the luminance of the brightest white and the darkest black that a TV can produce. Larger contrast ratios work out to deeper blacks, which make a big difference to overall picture quality. It’s of particular importance for dark scenes in movies and TV shows.
To test contrast, we measure the luminance of both white and black, and then calculate the ratio between the two. Luminance is a metric to measure the amount of light (or intensity) present on a surface, and is expressed in cd/m2. For TVs, the brighter the light of the screen is, the higher the luminance will be.
Update 21/04/2017: We added the 'Contrast with local dimming' value to show the effect of local dimming on the contrast ratio. We also renamed the 'Contrast' value to 'Native contrast' and took out the white and black value to streamline the whole contrast section.
The contrast ratio is the best indicator of how good a picture can look, as an image will generally look better if the blacks are really dark and the whites are really bright - if you are watching a dark movie, you don't want the image to look like a washed-out gray instead of being black. Because TVs can mostly achieve the same brightness of white, the contrast ratio really represents how dark black will be, and a higher number will mean darker blacks.
Keep in mind that having the deep, dark blacks you get from good contrast is mostly a benefit when you watch TV in a dark room. In brighter rooms, reflected ambient light will reduce the perceivable difference you might get with darker blacks, and so higher contrasts won’t look so different. Good contrast is also more noticeable with dark, shadowy scenes - it probably won’t make a big difference while watching something like football, or a bright and colorful cartoon.
What it is:
Ratio of the white brightness divided by the blacks measured on our checkerboard test pattern with a white target of 100 cd/m².
When it matters:
Dark scenes in a dark room.
On TVs, contrast is the ratio between white and black and is really the only one of these tests that should be looked at. A higher contrast ratio means deeper blacks and better picture for dark scenes viewed in a dark room.
The 'Native contrast' represent the contrast ratio of the panel use to build the TV and is measured without any image processing added or features design to accentuate the contrast ratio like local dimming.
To measure the luminance of the black and the white, we use a black and white checkerboard pattern. We calibrate the television to have a white luminance of about 100 cd/m2 and then we measure the luminance of a black square on the same picture.
Once we have measured the black and the white, we are able to calculate what the contrast ratio is. To get this number, we divide the white luminance by the black luminance.
Our test pattern
Because it is a ratio, there is no unit for contrast. Instead, the norm is to express the number as ‘X:1,’ with ‘X’ being how many times brighter white is than black. For example, a 2,000,000 : 1 contrast ratio would mean that the television emits white that is two million times brighter than its black. The higher the contrast ratio, the deeper the blacks, and the better the picture will look.
Contrast with local dimming
What it is:
Ratio of the white brightness divided by the blacks measured on our checkerboard test pattern with local dimming turned on (maximum) with a white target of 100 cd/m².
When it matters:
Dark scenes in a dark room.
The 'Contrast with local dimming' is similar to the 'Native contrast' but calculated with the local dimming turned on and set to high. This is also measured on a checkerboard pattern, with the same method used for the native contrast.
OLED vs LED
OLED sets have much better blacks than LED TVs do. This is because, with OLED panels, there is no backlight; the pixels emit light themselves, so a black pixel is just left off, and remains perfectly dark. That’s why OLED TVs tend to be listed as having ‘infinite contrast’ – you can’t divide the white luminance by a black luminance measurement of ‘0.’
How to get the best results
As a first step, try using the calibration settings we recommend (provided we have reviewed your TV). This will get good, basic contrast - meaning no additional contrast-enhancing settings - and with no loss of detail in dark portions of the image. You can find links to those at the top of each review. In particular, you should copy the ‘Brightness’ and ‘Contrast’ settings.
Most TVs also have a few other settings that affect contrast. We listed a few below, along with our thoughts on whether or not they should be used.
Related settings & other notes
Contrast: Adjusting this will let you affect how much contrast the TV has. We list a recommended setting with all of our reviews, but it's almost always fine to just set this to the maximum. On rare occasion, gamma might be affected, leading to a loss of detail in highlights.
Local dimming: Dims the backlight behind darker portions of the screen. Does deepen contrast, and worth using when implemented well. Can introduce issues like light blooming off of light objects within dark areas, and when done especially poorly, can dim the entire image. We discuss local dimming in more detail here.
Dynamic contrast: Uses software to process the blacks and make them darker. Unfortunately, this removes detail from the image. Not worth using.
Full/limited RGB: Full RGB may offer slightly more detail in blacks and shadows, but it’s not a big difference when compared with limited RGB. Just make sure that both your TV and your source device use matching RGB settings.
Gamma: Gamma does not control the depth of black, but it does control the amount of detail you will see in dark portions of an image. If you find it difficult to make out detail in dark images, consider making a slight adjustment to the gamma. We discuss gamma in more detail here.
Highlight brightening: Makes highlights in images extra bright, which will affect how contrast looks. Worth using with HDR media, but not really useful for most video.
Backlight/OLED light settings do not affect contrast, and so you should set them to whatever looks best in your viewing space. With LED TVs, both white and black will become about equally brighter or dimmer when the backlight is adjusted, preserving the ratio of light to dark. With OLED, blacks will remain perfectly black, and only other colors will become brighter.
A TV’s contrast ratio indicates the depth of blacks – a higher contrast ratio means deeper blacks – with higher contrast meaning better picture. It’s a very important part of picture quality, so if you want something that looks really good (particularly in a dark room), be sure to get a TV that has good contrast. When testing for contrast, we measure the luminance of black and of white and then divide white by black to get the contrast ratio number.
There are many things that can be done to improve contrast. As a good first step, look to our recommended picture settings (listed with every review), as those can help you get a good baseline. From there, you can enable or disable a few different settings that might help deepen blacks. Just remember that some of those settings will have other consequences for picture quality.
Mega Contrast Ratio is just a marketing term that began to be used a few years ago when they started advertising a dynamic contrast ratio of 1,000,000:1 and higher. It does not refer to a particular technology. It was used by a few companies just to make their contrast ratio look cool. It is about that same time that the stated contrast ratio was becoming useless because of all the false claims.
I watched the Gamescom event in Cologne Germany and after seeing Microsoft and EA use the lg la6200 to show off their next definition games like Forza 5, FIFA 14 and Ryse Son of Rome. I decided to buy that same model but your review shows that the lg la6200 has poor blacks, poor input lag. So my question is why would a rich company like Microsoft choose a tv that you rated barely a 7 to show off the graphics of their next gen games. I am confused, please answer. By the way, this is the best website for detailed tv reviews that i have visited, bookmarked.
For a setting like a convention, their choice of TV makes sense. The LA6200 has a very wide viewing angle for an LED, which is useful when a lot of people are watching the TV and taking pictures not directly in front of it. The low contrast is not an issue because the room isn't very dark, so you do not notice the poor blacks. The input lag is one of the best we tested so far. If you game all by yourself in a dark room though, you will be better off with a Samsung.
Plasma and VA panels have much less luminescence of the black than IPS panels. I wonder about VA. If contrast ratio on VA is higher than on IPS, is it possible to say that VA panels have less luminescence of other colors (blue, red etc.) than IPS or the contrast ratio concerns only ratio between the luminance of white and black?
All colors are affected not just white and blacks. If you look closely at a TV screen, you won't actually see any white pixels. Instead, each pixel is subdivided into 3 colors (red, blue and green). Therefore, the above equation of contrast can be also written as Contrast = (Luminance Red + Luminance Blue + Luminance Green) / (Luminance Black). IPS panels have about the same maximum luminance of the white color (so also red, blue and green) but they have a higher luminance of the black color. This creates a lower contrast ratio than their VA counter part.
I had a Kuro KRP600a and was foolish to sell it.
Am I stupid to say that the Panasonic ZT60 has a very dull picture?
I bought the Samsung F8500, but I sold it after owning it for a week.
Not bad, but really not up to the Kuro.
OLED is, from what I have seen, super (the LG is good, but somehow I don't like the Samsung), but I only saw them in stores.
All the Panasonic plasmas don't get very bright, which is probably what you meant by dull. With the right calibration, they are more than bright enough for dark/dimmed rooms, though. The Samsung F8500 can get a lot brighter, but its blacks are not as good.
I have a four-year-old LG 60" plasma TV, and the picture will not come on until the set warms up. It now goes off at will. Is this an expensive thing to fix?
Usually, yes. The repair bill can ramp up to a few hundred, easily. The problem is most likely that the capacitors are bad, so they take longer to charge up and no longer retain their charges. You will need to get that board replaced (or just replace the capacitors, but few repairmen do this).
I have a question about the testing methodology. If you measure the light output from a black frame, and then a white frame, and divide them to get the contrast ratio, how do you make sure the monitor doesn't dim the black frame (interfering with the measurement) with its software?
Great site. I found some 4K options that best mimic plasma-like black levels for dark room performance from your site, but the thing is, I am using it for a monitor, so I need 4:4:4 to get sharp text etc. Can you advise? Most TV's do 4:2:2, which is apparently substandard for a monitor. I could go to 30Hz if that would resolve the issue.
Unfortunately, we did not test for Chroma 4:4:4 support in 2014 (though we will for 2015 TVs), so we can't say which TVs do and do not support it in 4k. Some TVs might support it at 60 hz on an HDMi 2.0 connection, but no HDCP 2.2 connection will allow 4k 4:4:4 at 60 hz. Either kind might be able to pull it off at 30 hz, but that will depend on the model.
Question between Micro Dimming Pro & Frame Dimming: which gives better results?
Neither is very effective. Micro dimming is a software-based solution, so it's not great at localizing the dimming effect. Frame dimming just dims the entire screen with dark scenes, which does make darks darker, but also makes everything else darker (ideally, the brights would stay bright). Learn more about this here.
Please add color gamut in the usage section. Even without a wide color gamut there are tvs that hit higher rec2020 spec than others. Please make this information easier to compare. At least add more tv models to compare side by side.
Thank you for the suggestions. We have some changes coming that you might appreciate. First, we will soon add a score to the color gamut coverage. We are also working on a different way to compare TVs by letting our visitors select the features they want to compare. That way, you will be able to sort TVs by color gamut coverage, if that is what you want. Soon is a big word (we are working on a few improvements at the same time), but we wish to be able to pull the new features out in the next following months. Thank you again for your feedback!
What type of test pattern generator are you using? Some of these TVs are HDR capable but won't display HDR unless the signal tells it to. Does your TPG have HDMI 2.0a output? Is it sending HDR metadata? Have you precalibrated the TVs for a maximum luminance? If not, this test is losing relevance in the coming HDR world.
Good points! So far, we made all of our test patterns without the help of a test pattern generator. And then we use a PC to display the patterns on the TVs. For contrast, since we use a checkboard pattern, it basically give us the native contrast of the TVs. For HDR, we measure the peak brightness of a 2% white window with all settings to the max. We also test the effectiveness of local dimming and we measure the color gamut of DCI P3 and BT.2020 (new color gamut screenshot coming soon). All of which plays a role on how HDR content is displayed on TVs. We don't test the compatibility of the signal though (if a HDR video can be played or not) and it is a good idea to add that to future reviews. We will be able to do that soon with the Samsung UHD blu-ray player that we ordered. Thank you for your input!
Although Local Dimming reduces the light output on the ANSI checkerboard patterns, I believe that the contrast ratio should be measured with it on regardless to get a more true contrast level report and closer representation of what we actually see......any thoughts on this?
We haven't tested a TV yet that has a local dimming that was good enough to be able to actually change the contrast ratio on the checkerboard pattern. For example, we measured the same contrast on the Vizio P with and without local dimming (the white and the blacks both dimmed by the same ratio). Also, our test philosophy is to isolate each component of picture quality and measure them separately. In this mindset, it makes sense to measure the static contrast ratio without local dimming.
I noticed on your test that the measured black luminance for different TVs was too different from others tests and reviews. For example, the Sony r450A in your test has the following values:
Black: 0.022 cd/m2
White: 104.5 cd/m2
Contrast: 4750 : 1
But on other sites, this TV has only 0,05 cd/m2 (or 0,056 precisely) which gives a contrast of ~ 3200 with ~170 cd/m2 for the white.
All the black luminance measures on your tests are lower to the others sites I have found especially for VA panel TVs. They are, at best, 0,04 cd/m2 (On the others sites) and the lowest black level I have found was for the Panasonic 42E6, with 0,03 cd/m2.
1- I really want to know the reason for a such big difference on the black luminance measurement?
2- I got used to the near-perfect black level of the CRT TVs. I have two LCD TVs and the dark scenes in a dark room look awful to me (either on Sony 32cx520 or on Samsung 42f5000), due to the gray blacks. What is the minimum black luminance for an LCD TV to have a relatively good black in a completely dark room?
1) Two reasons. First, we take the black measurement at a 100 cd/m2 white, which is a bit darker than other sites. The contrast ratio stays the same no matter the luminosity of the backlight, so a darker white also means a darker black. Second, we use a checkboard pattern to measure both the black and white. This isn't really important for an LED, but it does affect the result for a plasma.
2) Below 0.030 cd/m2 is relatively good (using our numbers). Of course, in a pitch black room, you will still see it.