Why You Should Ignore Contrast Ratio Specs:

Scottnot

Contrast Ratio seems to be a frequent topic on this forum, and although a great deal of information has been written on the topic newcomers to HDTV still seem to be easily baffled, confused and misled by this ubiquitous specification. I will try to explain why the very notion of a specification for contrast ratio is of no practical value when purchasing a new HDTV and suggest an alternative criteria of greater significance to the consumer. Note: it is suggested that the wiki entry and the included Notes and External Links at the end of the article be used as source references for those unfamiliar with the topic - here is a link to the wiki article:
http://en.wikipedia.org/wiki/Contrast_ratio

What is Contrast Ratio? Simple: it is the ratio of the TVs "white luminance" to it's "black luminance" . . . under some viewing condition, which in an "ideal" measurement would be a totally darkened room.

Isn't the "ANSI Checkerboard" the best method to accurately represent contrast ratio? No, the ANSI checkerboard is simply the only recognized standard of measurement. Measurement of black level will be heavily dependent on room darkness. Measurement of white level may be correct, however may represent a white luminance which can actually be painful to the eyes.
While it is a "standard", it does not and can not accurately predict which of two different TVs, with identical ANSI CRs will provide the better picture with respect to contrast ratio in a particular viewing environment.

White Luminance: White luminance can and should be understood at three different levels:
First: "Peak Brightness" is the specification that you will find in the specification section of your set's user manual. It represents the white luminance that your set can achieve with the contrast control set to 100%. Sufficient peak brightness will allow for adjustment as the screen ages, and provide a margin during calibration, that's about all.
Second: "Calibrated Brightness" is where a set is capable of being calibrated to achieve maximum white luminance while still being able to display the differences between shades of white such as a white button on a white shirt, a polar bear in a snow storm, and such.
Increase white luminance beyond this point and you lose detail in bright content; calibrated brightness (white luminance) will almost always be considerably lower than "peak brightness".
Even so, calibrated brightness is usually far brighter than most people find comfortable to watch . . . it quite literally can hurt the eyes and cause headaches.
Third: "Normal Viewing Brightness" is where the viewer finds the displayed picture content comfortable and enjoyable to watch. It is affected by room lighting, program content and personal preferences. It will almost always be lower than lower than either calibrated brightness or peak brightness . . . perhaps by as much as several times lower.

At this point, it should be clear that since "white luminance" is the numerator in the "contrast ratio" equation one may wonder which value to use . . . "peak", "calibrated", "normal viewing".
"peak" would be bogus, since it is not "real world" in any normal viewing environment.
"calibrated" might be useful when considering that a set may be used in a very bright room and displaying dimly produced content. But, of course, the ANSI specification does not call for "calibrated" white luminance, and it is not likely that all manufacturers are about to volutarily move to this measurement method.

Black Luminance: also comes in more than one flavor.
First: "Screen Black" will be the blackest black that your screen can display while the set is turned on and the brightness control is set to "0".
Second: "Calibrated Black Level" is where a set is capable of being calibrated to achieve minimum black luminance while still being able to display the differences between shades of black such as details in a dark alley, a black cat's fur, and such. Unfortunately, calibrated black level is extremely dependent on room lighting and the calibration level will vary considerably if done in a totally darkened room versus a brightly lit room.
Now, here is an interesting point - (assuming fixed room lighting) once black luminance has been properly set, it should not be changed:
Lowering it will result in loss of detail as different "shades of black" will simply merge together.
Increasing it will result in the "blackest black" becomming "less" black.

Again, it should be clear that since "black luminance" is the denominator in the "contrast ratio" equation we wonder which value is used in the specifications. It sure seems that measured black luminance at the "calibrated black level" would be most meaningful . . . of course, the ANSI specification does not require that, and we can be certain that it is not used.
And, even if it were, what about room lighting?

Room Lighting: Unless you do all of your TV viewing in a totally darkened room, this matters. Most television viewing is done with at least some ambient light, and very often with a lot of ambient light. Consider the effects of room lighting: It has been quite well documented it the earlier referenced sources that even the smallest amount of ambient light can render even the most impressive contrast ratios to be merely academic.
Now, add to this, the fact that a first order factor on the impact of ambient light is screen reflectivity and things start getting really mucked up! Have you ever seen a manufacturer specify the "screen reflectivity" of their products?

Now what? Well, unless the manufacturer(s) become willing to specify the conditions under which they have measured "calibrated white luminance", "calibrated black luminance", "ambient light", and "screen reflectivity", forget about so-called contrast ratio because it has no meaning. Instead, consider:

Most modern HDTV products provide at least "pretty darn good" contrast under "normal" viewing conditions.
If you watch most of your television in a brightly lit room you should be more concerned about screen brightness than contrast ratio.
If you prefer to watch television in a "lights out" environment, then look for sets that offer the "best black levels".

And keep in mind (although it is a matter of some debate), the human eye when viewing normal television content can probably not discern much better than 200:1 to 500:1 of contrast ratio (some argue as high as 700:1).
You just "ain't never" gonna see those big numbers!!

pappylap

But I saw where Samsung was boasting about a set with 1,000,000 to 1 CR....you mean I aint goin' to see a great picture with numbers like them?...Tee....Hee...Hee...

Great post Scott maybe the new management will make a sticky since this comes up about 20 times every day or two...

__________________

You can checkout anytime you like but you can never leave.."

50"Pioneer KURO 111FD Pro Elite
Pioneer BD320 Blu-ray player
Pioneer VSX-919AH A/V Receiver
32" & 37" Olevia LCDs
Sony DHG HDD 500 gig DVR
Samsung DTBH260 Digital Tuner
One Roll of Charmin Quilted Plus in each Bathroom...

rbinck

It is stuck. I was considering moving it because it pertains to more than just flat panels, but I don't know which forum would be better. Really though since this comes up more for flat panels than anything else, I guess here is as good as any.

It is a greatly mis-understood spec for sure.

There are elements in the OP that seem to be under-rated as well, like the reflectivity of the screen. That is more important to me than black levels because I don't watch much dark TV and when those scenes do come on I hate watching myself watch TV.

But others may and do have different ideas about black levels as it relates to picture quality. I have watched many hours on both plasmas and LCDs both and I prefer LCD myself, even with the lower contrast ratios and reduced black level definition. But then the shows and sports I watch don't have much black in them.

__________________
www.highdefinitionblog.com
www.BlogOfHawaii.com

macmarkus

Good job scottnot ! This is awesome...hopefully members will direct those with questions regarding this issue to this thread, and maybe they'll even read it, ha ha !

pappylap, I think I just read an article in Sound & Vision for April/08 about that exact thing, 1,000,000 - 1 contrast. I believe it was pertaining to the oled display's. very funny though, considering what is stated in this thread on contrast ratio.

peace

macmarkus

Jim Bob Jones

I disagree with the proposition that contrast ratio specs should be totally ignored just as I find objectionable the theory that amplifier/specs are meaningless. Measurable data that may have a direct impact on performance should be disclosed to potential buyers. The degree of impact may be debated but that does not justify discarding the information.

rbinck

Part of the problem with contrast ratio measurements is the lack of a standard method of doing the testing. Sure they can disclose their number, but they should also describe the method used for arriving at that number. Meanwhile, in practice their number is about as useful as the miles ratings for antennas. Only useful, at best, to compare to other models of the same manufacturer.

__________________
www.highdefinitionblog.com
www.BlogOfHawaii.com

crysisworm

Looks like Cathywang changed her stripes.........

__________________
.....................
The Underlings are upon us

woody westerfi

what is the exact purpose of the dynamic contrast control on a tv and when would you ever need to use it?

Scottnot

I think it varies from manufacturer to manufacturer depending on how they attempt to impliment the control. In some sets, it may reduce the backlight when very bright content is sensed; in others, in may make slight tweeks to either brightness or contrast depending on the overall picture content.

I believe the overall consensus is that more often than not, it does more harm than good and can in some cases act to the detriment of a well calibrated black level.

Bottom line - turn it "OFF".

woody westerfi

scottnot:thanks

BlackFoxx

wow lol im nice but it sounds good i will ignor what about dynamic contrast on the 40" samsung in the loft when i turn dynamic up to high it looks better...or does it?

xuchenxixu

so 4k and 40k from the same company is not gona make a difference?

Scottnot

Correct . . . for two reasons.

First: it is not likely that either is a "correct" value in that you are not being told how the values were arrived at, nor is it likely that they would be used for "normal" viewing.

Second: even if you had two sets, one with 4k CR and the other with 40k CR after being set up for "normal viewing" in some arbitrarily lit (or unlit) room, the white luminance will be on the order of 300-400 cd/m2 (let's assume a dark room so use 300 cd/m2. Now, the black level will have to be .075 cd/m2 on the 4k set and .0075 on the 40k set.
Now, a decent LCD TV will be capable of "calibrated black luminance" on the order of .75 cd/m2 (note how that yields a CR of 400:1 which is about right for a LCD display). A plasma set may be capable of levels as low as .075 cd/m2, but certainly not .0075.

In any event, the human eye, when viewing television content simply does not have the contrast appreciation level required to distinguish between 4K and 40K contrast levels.

ZincPony

i read somewhere it said you should use CR to compare models from the same manufacturer

New User

As it has just been explained in great detail, that is not the case.

__________________
Yes, I am an agent of Satan, but my duties are largely ceremonial.