or Connect
AVS › AVS Forum › Display Devices › Digital Hi-End Projectors - \$3,000+ USD MSRP › Some people are still being told that 200:1 is all we can see....
New Posts  All Forums:Forum Nav:

# Some people are still being told that 200:1 is all we can see.... - Page 2

My humble apologies in advance for citing Wikipedia, but the author on the Eyemakes an interesting statement:

Dynamic range

What is interesting is that the process of iris adaptation is continuous and nonlinear much like video data itself. So as we are watching a movie the eye is adapting all the while, starting over almost from one scene to the next. The lower ratios may be more applicable to real word viewing but the extraordinarily high ratios may be for very specific and narrow conditions, only useful in comparing one display from the next and what the HVS is capable of but rarely experiences.

So is it possible that both assertions are correct?

### AVS Top Picks

Quote:

Quite an interesting paper - in essence, it provides an age-corrected equation that allows you to determine disability glare - the ratio of a point source of light (cd/m^2) to scattered light (lux) (or, as Chris has been referring to it, spill) - given an angle theta from the source which is accurate for the domain of 1°

So, given a light intensity from a "point source" on your projector, your age, and the angle of your line of sight from the point source, you could determine the intensity of the scattered light on your retina. From there you could determine a contrast of point source to scattered light - percieved contrast ratio - and then compare that to the actual contrast at the boundary. Obviously a point source - one pixel - on your projector isn't as relevant to this discussion as a larger bright object on your screen surrounded by a much darker area. In that case, you simply need to integrate the same equation over the angular aperture of the light source to find the same ratio.

Of interesting note - the paper specifically cited that disability glare across the general population increases rapidly past age 60, doubles at 70, and triples by 83. Good news to all of us uber picky AVS-types - as we get older our diminishing visual acuity will make us more accepting of lower PQ! (J/K ) By that point they'll probably have some mind-blowing displays available anyways...

I know this is all an aside from the main discussion, but I found it interesting nonetheless. I'd still like to see any studies performed on larger angle HVA contrast that don't involve disability glare. After seeing the effects of disability glare, and that it usually limits perceivable contrast in those particular situations to 150:1, I find it VERY hard to believe that static HVA is that low at large angles. In other words, the originally cited 200:1 does sound a little outlandish.
The Wikipedia entry is mistaken, and it provides no references for those claims.

Quote:

So as we are watching a movie the eye is adapting all the while, starting over almost from one scene to the next.

That is correct. But again, we are interested in the simultaneous capability, not a dynamic one over time which I cited previously as being in the range(over significant time) of 10^14.

Quote:

The lower ratios may be more applicable to real word viewing but the extraordinarily high ratios may be for very specific and narrow conditions

I think if you follow the discussion, it's fairly apparent that it's the other way around. The much smaller numbers such as Vos's 150:1 limit at a high-contrast boundary are for very specific occassions, while the larger number is more appropriate for general images as it is not restricted to all the conditions imposted to get the much lower number of 150:1 such as having a high contrast boundary and viewing right adjacent to it by a very small angle.
Quote:
That is correct. But again, we are interested in the simultaneous capability, not a dynamic one over time which I cited previously as being in the range(over significant time) of 10^14.

Where does the 10^14 number come from? That's a very large number and is certainly not referenced in your citations either.

I am not so certain Vos 1984 is making reference to boundries, would like confirmation of this. It is certainly not entirely clear from reading the Siggraph paper you cited. Tried to find the Vos paper but all I managed to come up with was a listing of his work as it relates to visual displays in the work place from 1984 which would imply a full field of view. But I am far from certain this is even the paper.

http://www.cie.co.at/framepublications.html see Publication CIE 60-1984

You should correct the Wikipedia entry if it is indeed wrong.
The way that the 200:1 was mentioned also suggested that a display that was advertised with over 200:1 was either BS or unbelievable, so that's another part if the training that perhaps should be worded differently as I think most of us agree that high on/off is pretty important, especially for determining black level (or a lack of). The differences in black levels between a low CR display and a high CR display is pretty obvious and visible between them especially if displayed side by side.

I had another question which Erik helped me with the other day:

In it I posted a scene from Gladiator which on either a DLP or CRT with certain display capabilities (as shown in the thread), will have around 2400:1 CR. That being the case, I would think that most people would be able to see all the detail in that image without finding any of it crushed into white or black. Display it on a low contrast capable LCD for example and you will see larger areas of grey without detail.

Now, if you look at that image on a monitor you will easily encompass it within your field of view and see all that detail (do you perceive large areas of crushed black or white in it?). So you are seeing around 2400:1 (if the monitor is capable of that and you are in an environment where ambient is not reducing it). I don't believe the perception of that image would change on a large screen with a display that capable of over 2400:1 on/off CR

I wonder what it would look like if it was reduced to just 200:1

If there is anything fundamentally wrong with my take on the above, I'd appreciate the input since it seems pretty rational to me but sometimes things aren't what they seem.

TIA

Gary
Quote:
Originally Posted by ChrisWiggles

The Wikipedia entry is mistaken, and it provides no references for those claims.

Someone else who got it wrong
"Whereas the usable light intensity in our visual world varies by approximately eight orders of magnitude, the range over which we discriminate differences in light intensity varies only approximately 100-fold."

http://www.iovs.org/cgi/content/full/42/11/2721

I will say for the last time; the < 300:1 CR cited for the HVS has nothing to do with the ratio between the average light measured of the white squares / average light measured of the black squares in a 4x4 checkerboard. (ANSI CR)! The fact most devices would give a much lower CR number on a 8x8, and even lower on a 16x16 checkerboard tells me that we have not yet reached the HVC < 300:1 limit, with current display devices.
Quote:
Originally Posted by mdtiberi

That sounds eerily similar to my recollections (though I was told this by someone with some scientific standing, as opposed to reading it). Whether it's accurate or not, is a good question, but this was what I was trying to explain (and recall).

Chris, I understand that you believe this is incorrect, but, just as we need to provide proof that it is correct, you need to provide proof that your conjecture is correct, as well. Especially considering there is a good amount of material out there that backs up mine (and others').

Quote:
Originally Posted by Gary Lightfoot

Now, if you look at that image on a monitor you will easily encompass it within your field of view and see all that detail (do you perceive large areas of crushed black or white in it?). So you are seeing around 2400:1 (if the monitor is capable of that and you are in an environment where ambient is not reducing it). I don't believe the perception of that image would change on a large screen with a display that capable of over 2400:1 on/off CR. I wonder what it would look like if it was reduced to just 200:1

That's just it, though. You are perceiving 2400:1, but at any given instant, your eye is only seeing 200:1 (or whatever small number). The brain is using iris adjustments (and possibly chemical adjustments, as mentioned in the Wikipedia article) to stitch together an image that you perceive to have a much higher contrast ratio.

In response to the OP, while 200:1 may be technically accurate, it isn't really applicable to our concerns as AV enthusiasts - because it's all about what our brain perceives, not what our eye technically sees during a brief blip of a second.
Quote:
Originally Posted by Lindahl

That's just it, though. You are perceiving 2400:1, but at any given instant, your eye is only seeing 200:1 (or whatever small number). The brain is using iris adjustments (and possibly chemical adjustments, as mentioned in the Wikipedia article) to stitch together an image that you perceive to have a much higher contrast ratio.

The thing is, for me to truly believe that, I would assume that as I move my eye around an image, I would be concious of the areas not in the center of my view collapsing into much lower contrast areas, but I don't. The image remains the same without any changes in brightness or darkness which I assume should happen if the CR is being limited to just 200:1.

Quote:
Originally Posted by Lindahl

In response to the OP, while 200:1 may be technically accurate, it isn't really applicable to our concerns as AV enthusiasts - because it's all about what our brain perceives, not what our eye technically sees during a brief blip of a second.

I can see exactly what you mean, and I did have to think about that. That's what made me think that there would have to be a change of magnitude in a 2400:1 scene if I were to look at the high end of luminance (sky through the trees, or Maximus's face) so the darker parts which were 2000:1 away from it would have to then become much brighter for me to still percieve them and the detail within that range, otherwise they would be outside my suggested range of 200:1 and I wouldn't be able to see them. When I do that, I look at just that part of the image and don't move my eyes but I can still see the image that I saw when I looked at it directly.

So I wonder if others can understand the confusion with the 200:1 number with real images.

I'll say it again though, the way that the number is used in the ISF training suggests that contrast numbers above that are irrelevant, yet even if the above 200:1 is true, we still need high on/off to render an entire scene so that even at a given instant we need the higher display capabilities to render a real world image for our eyes/brain to stitch it together for us to perceive an image that is similar to real life. 200:1 is not how we perceive real life so to say that having more than 200:1 because that's all the eye can see would be misleading (and it is for those who blindly beleive what they are being told).
Quote:
Originally Posted by HogPilot

Quite an interesting paper - in essence, it provides an age-corrected equation that allows you to determine disability glare - the ratio of a point source of light (cd/m^2) to scattered light (lux) (or, as Chris has been referring to it, spill) - given an angle theta from the source which is accurate for the domain of 1°

After reading this paper, it does not appear to be applicable to home theater environments. Dr. Vos makes many references to the "veiling background" or background luminance and entoptic scatter (light bouncing around the inside of your eye). All of the examples are situations like people driving, riding a bike, interior and tunnel lighting. Light coming from the periphery is significant and is factored into the derivation of the equation.

It is important to read academic papers in the complete context in which they are written to fully understand where they can be applied. It is certainly a stretch to apply this paper to a controlled viewing environment where stray light (background luminance) is painstakingly minimized or eliminated for optimum viewing.
Quote:
Originally Posted by HoustonHoyaFan

I will say for the last time; the < 300:1 CR cited for the HVS has nothing to do with the ratio between the average light measured of the white squares / average light measured of the black squares in a 4x4 checkerboard. (ANSI CR)! The fact most devices would give a much lower CR number on a 8x8, and even lower on a 16x16 checkerboard tells me that we have not yet reached the HVC < 300:1 limit, with current display devices.

Even if accurate, I wonder if the 300:1 figure is relevant for home theater purposes. A picture may have daylight out the window, while a dark corner inside the house. The projector doesn't know which one you are looking at, and so both should be projected accurately. You wouldn't want the shadow portions of the screen compressed because there is light out the window, nor would you want the everything in the window compressed because of the shadowy corner.

So even if the HVS limits was 300:1, that wouldn't mean it should be a goal of a projected image. The projected image would need to exceed that by a great margin, I would think.
Quote:
Originally Posted by mdtiberi

After reading this paper, it does not appear to be applicable to home theater environments. Dr. Vos makes many references to the "veiling background" or background luminance and entoptic scatter (light bouncing around the inside of your eye). All of the examples are situations like people driving, riding a bike, interior and tunnel lighting. Light coming from the periphery is significant and is factored into the derivation of the equation.

It is important to read academic papers in the complete context in which they are written to fully understand where they can be applied. It is certainly a stretch to apply this paper to a controlled viewing environment where stray light (background luminance) is painstakingly minimized or eliminated for optimum viewing.

In this case, the paper applies to a bright object, i.e. a streetlight, contrasted against a dark background, i.e. a dark sky. Whether it's on a screen or in real life, the effect would be the same given enough contrast between the two - the halo created on the retina outside of the bright object limits our ability to discern detail on an adjacent dark area. My intent was not to discuss a bright screen against the dark background of a light controlled room, but to discuss the ability (or rather, inability) of our eye to discern detail due to disability glare in a small angular region on the screen.

A synopsis of the equation discussed in the paper is found here, and reads:

Quote:
Three disability glare equations have been defined to extend the classic Stiles-Holladay equation to take account of the effect of age, the effect of ocular pigmentation and to extend the angular domain over which the equations are valid. They all describe the veiling luminance L veil (in cd/m²) due to a point glare source at an angle Theta (in degrees) to the line of sight that gives rise to an illuminance E glare (in lx) in the plane of the observer's eye, and they do so with adequate accuracy within their respective validity domains.

From Dr. Vos's discussions on the phenomenon early in the paper - a bright light creating a halo that obscures the surrounding image - as well as the explanation above, it's pretty clear that L eq and L veil are the luminance of the veil or halo caused by E glare of the bright light, not background light or light from the periphery, which would make it totally applicable to this discussion.
Quote:
Originally Posted by tstites

I won't add to his comments, but published scientific data that I've seen puts the upper end of the range under best conditions to be a little under 300:1 and you go down from there as you factor in age, image intensity and spatial separation of the light/dark elements.

My guess is that you are thinking of the Contrast Sensitivity Function, where people seem to have missed that this is an inverse (or reciprocal) function and so shows the opposite of what they think it shows as far as what happens as we get only. The CSF measures the inverse of the contrast threshold and therefore measures the least amount of CR we can see, not the most. Completely different.

The numbers from Brightside that were reportedly somewhere in the range of 100,000:1 without the eye's iris adapting and full range of 10^14:1 (although with lots of adaptation time). Not meaning anything personal, but the person who wrote the Wikipedia article doesn't seem to know what they are talking about with respect to that 1,000,000:1. I can demonstrate seeing way over a million to one range just measuring outside my house (where over 1500 ft-lamberts is common) and then in my theater.

Other people seem to confuse things like 300:1 as if they are ratios when a person is using them as a number of steps (which isn't a ratio). If a person is claiming some number of steps at once then they should say that and not use the number as if it is a ratio.

It is so easy to show that people can see instantaneous improvements well beyond 300:1 that I wonder how some of this stuff gets propagated. To me it almost seems a little bit like somebody claiming that people can't jump any further than 2 feet and a bunch of people just believe it without doing any testing themselves.

I'll have to come back and look at this as I have to run. I did have a person who teaches ISF courses tell me that the limit was something like 300:1, but the image they use doesn't even show close to the limits. There are conditions where people can't see as much (like very close to a bright spot as people have discussed) and others where they can see a lot, but the absolute limits are the most we can see across all images, not just on some particularly limiting images.

The difference between a projector that would do 500:1 simultaneous CR with the image that MrWigggles posted here:

http://www.avsforum.com/avs-vb/showt...&&#post9086092

and one doing 4000:1 simultaneous would be pretty profound in a light controlled environment.

--Darin
Quote:
Originally Posted by HoustonHoyaFan

I agree. The < 300:1 per visual frame is certainly the CW in the HVS community. I am on the BOD of a artificial vision startup, and that is the number the Chief Scientist uses. The joke at our board meetings is; 300:1 is the number we will use until my video projector "pals" come up with 50 years of experimental data to prove otherwise.

You've been claiming this for as long as I can remember. Have you even done a single test to see if you can see more than this or not?
Quote:
Originally Posted by HoustonHoyaFan

Note that the < 300:1 is not the same as an ANSI CR measurement. It is more a measure of the range of light levels that the HVS can discern in a single visual frame. levels outside that range are clipped into white and crushed into black.

Then why can I discern two levels that are outside of the 300:1 range? If they were both crushed to black, then they wouldn't be discernable.

Are you sure somebody isn't confusing this with the number of steps (which isn't the same as a ratio of lumens for each spot) or with testing that is about the reciprocal. If there has been 50 years fo experimental data proving that people can't see more than 300:1, then you should be able to point us to it. Shouldn't you? Pointing us to the vast amount of data out there that shows that people can't see less than the inverse of 300:1 (which is what Contrast Sensitivity Function shows) wouldn't support your point. Since we've been over this multiple times, I hope you will come up with some data this time and tell us whether the source is the Contrast Sensitivity testing that many people get confused about. A 300:1 CSF score is a contrast ratio of under 1.01:1, not a contrast ratio of 300:1.

If there is a ton of scientific data out there about the most CR people can ever see at once, then I don't know where it is. Most of what I find is about the least CR they differentiate at at once. And since the least CR we can see at once goes up with age (our eyes lose their ability to differentiate small differences in levels), we need more CR to differentiate details as our eyes age.

I hope that you won't keep telling people that the limit for human vision is 300:1 contrast ratio without actually getting some data. Please ask your Chief Scientist if he is referring to the CSF. If he is, then he is obviously wrong if he thinks this is a test of the most CR that people can see. If he is referring to other scientific data, then please get a source for that. It would seem like quite a coincidence that the CSF limit is about 300:1 and this is also the number you are claiming for the most CR people can see.

--Darin
This number of 200:1 contrast we can see in any one time is relevant to Ansi CR right? But the lower the APL of the scene, the lower the contrast level drops. So base on this site

http://home1.gte.net/res18h39/contrast.htm

If I plug in 17500:1 ONOFF CR and 260 Ansi CR, you can see that the CR at all levels are not even close to 200:1. Of course if I decrease the room reflectivity, we can hit over 250:1.
BUT MAINLY IN THE HIGH APL SCENES! At 0.0001 reflectivity, the JVC will hit over 250:1 CR at a scene with 50% APL and above. The Sharp will do twice as well as the JVC at 50% APL and only go below 200:1 at about 25% APL.

Could this be what these scientist mean by 200:1 CR at any one time? Of course with smaller checkerbox squares, I fully expect the Ansi CR to drop. So it seems that display technology still has some ways to go before it matches what we can see any one time.

Oliver
Quote:
Originally Posted by oliverlim

This number of 200:1 contrast we can see in any one time is relevant to Ansi CR right?

The ANSI CR checkerboard would be more limiting to our vision than some other images (where we could see more than we could there). Despite that, seeing more than 200:1 in the ANSI CR checkerboard is easy. Just get some black velvet or black posterboard and put it up or put your hand up between the projector and one of the "black" rectangles with a projector that does around 200:1 and you should see that it is easy to see the improvement. And movement is even less limiting than static images.

I think it is unfortunate that people are giving credence to small numbers like 100:1, 200:1, or 300:1. I hope people will do their own testing.

--Darin
Quote:
Originally Posted by darinp2

The ANSI CR checkerboard would be more limiting to our vision than some other images (where we could see more than we could there). Despite that, seeing more than 200:1 in the ANSI CR checkerboard is easy. Just get some black velvet or black posterboard and put it up or put your hand up between the projector and one of the "black" rectangles with a projector that does around 200:1 and you should see that it is easy to see the improvement. And movement is even less limiting than static images.

I think it is unfortunate that people are giving credence to small numbers like 100:1, 200:1, or 300:1. I hope people will do their own testing.

--Darin

I agree. I always felt that DLPs just show more depth to the pictures. And the only thing that seems to have some clues to why that is so seems to be Ansi CR. It does show that we still do not have all the measurements and answers to show us what we are seeing and why.

Oliver
Quote:
Originally Posted by darinp2

...Please ask your Chief Scientist if he is referring to the CSF. If he is, then he is obviously wrong if he thinks this is a test of the most CR that people can see...

I have the utmost respect for your AV knowledge, however our Chief Scientist is one the foremost retina researchers in the world. The startup is in the area of artificial vision. I, and the investors who have to date invested 10s of MM \$ in his team, will take his word, over yours, as to the single frame CR of the HVS.

As I posted above, I guess Heinrich and Bach are also confused?
Quote:
Originally Posted by Heinrich and Bach

Whereas the usable light intensity in our visual world varies by approximately eight orders of magnitude, the range over which we discriminate differences in light intensity varies only approximately 100-fold. Photochemical and neural mechanisms keep the limited range of discriminability in the range of the prevailing luminance, known as luminance adaptation. Similarly, although on a smaller scale, contrast adaptation shifts the steep part of the contrast transfer function to match the prevailing contrast condition.1 On a single-cell basis, there is a distinction between contrast adaptation and contrast gain control. The latter represents rapid changes, taking approximately 100 msec, as found in retinal ganglion cells of cats2 and MX ganglion cells of macaques.3 Conte et al.4 also found rapid changes in humans by means of pattern electroretinogram (PERG) and visual evoked potentials (VEPs). Contrast adaptation refers to relatively long-term changes (seconds to minutes).

http://www.iovs.org/cgi/content/full/42/11/2721

Can you point to a single scientific reference which claims per frame HVS CR > a few 100:1?
Quote:
Originally Posted by HoustonHoyaFan

I have the utmost respect for your AV knowledge, however our Chief Scientist is one the foremost retina researchers in the world. The startup is in the area of artificial vision. I, and the investors who have to date invested 10s of MM \$ in his team, will take his word, over yours, as to the single frame CR of the HVS.

It sounds like you are going to invest all that money and won't do a single test yourself? Is that right? Do you still have a Ruby? Do you have a light controlled room? It should have an ANSI CR of close to 300:1. Why not do some testing? The above makes me think of the people who said that heavier objects must fall faster than lighter objects because Aristotle (I believe it was him) said it was so.

I take it you won't ask him if this is related to the Contrast Sensitivity Function testing.

The guys working on the high dynamic range display at Brightside might not have invested as much as your team, but should be able to easily demonstrate high simultaneous CRs that people can discern from lower simultaneous CRs. Do you guys have a device that can show huge levels like those vs lower levels?
Quote:
Originally Posted by HoustonHoyaFan

As I posted above, I guess Heinrich and Bach are also confused?

I'll have to look at that closer later, but if you are really going to refuse to do any testing yourself, maybe you could tell us why you are going to refuse to do that. As I've pointed out before, there are conditions where we are more limited than others, so looking for the most means looking for the conditions where we can see the most. As I said, I'll have to look closer at the data there. But if the adaptation level was x and a person could discern up to 50:1 above that and 50:1 below that at the same time, then the whole range there would be 2500:1. If they could only see 100:1 total then it would have to be limited to 10:1 in either direction, or one direction would have to be more limited than the other in a way that the full range limit was only 100:1.

--Darin
Quote:
Originally Posted by darinp2

It sounds like you are going to invest all that money and won't do a single test yourself? Is that right? ...

I'll have to look at that closer later, but if you are really going to refuse to do any testing yourself, maybe you could tell us why you are going to refuse to do that.

In all honesty, I am not qualified to do any testing which would refute the claims of leading HVS scientists. My undergrad degree is in Biochemistry, which allows me to understand most of the papers, thats my limit

I am however interested in knowing what are the HVS limits, if any, of ANSI CR measurements. The new Sony 40" XBR LCD has been tested at >1,200:1 ANSI CR!
Quote:
Originally Posted by HoustonHoyaFan

In all honesty, I am not qualified to do any testing which would refute the claims of leading HVS scientists. My undergrad degree is in Biochemistry, which allows me to understand most of the papers, thats my limit

If a scientist told you that you couldn't jump any higher than one foot off the ground, would you test it or just throw up your arms and say that you weren't qualified? I'm being serious. I don't understand why you seem to want to not understand this stuff.

If a setup was showing an ANSI CR of 300:1 off the screen and that was improved to 2000:1 ANSI CR, do you think you would be able to see that the CR had improved by looking at those two images or not? If MrWigggles image was put up with 500:1 simultaneous CR and that was improved to 4000:1, do you think you would be able to see that the 4000:1 was better? If you are going to keep telling people that 300:1 is the limit, why not at least test to find out if you can see improvements beyond that?

Maybe you aren't even understanding what your chief scientist is saying. I know you said you guys joke about the testing that has been done, but in my theater I put up some images with transition points way beyond 300:1 and then we joke about how there are people out there who claim that we can't see those higher CRs since our eyes are limited to a much smaller range than is in those images, even though the transitions are easy for anybody there to see. If your company could use some money, maybe we could setup a bet to see if I can see huge improvements to simultaneous CR beyond 300:1 in certain images off a home theater projector screen with a black velvet border in a mostly black velvet room (like my theater). Of course, it would be with images were we can tend to see higher CRs, not ones where we are more limited (as I have said multiple times, our limits are based somewhat on the makeup of the images). You might want to actually do some testing before agreeing to take a straight up bet on this one though.

--Darin
Quote:
Originally Posted by HoustonHoyaFan

In all honesty, I am not qualified to do any testing which would refute the claims of leading HVS scientists.

Really?

The testing involves these steps:

2) display a standard checkerboard pattern
3) wave your hand in a black square while looking for a shadow

For variations also suggested in this thread: observe to see if you can identify screen masking or your black screen border that borders the black squares. You can also find black material or material that is of a different reflectivity of your screen surface and place it in the black square and look to see if you can still see that material.

These are experiments anyone can do quite easily in just a few minutes.

And they don't refute anything quanitified about the HVS, which is what Darin and I and others have been saying for a while now. Use a little bit of common sense and think about what the claims are. Do they make any sense at all? Do the test, it only takes a few moments.

It's unbelievable to think that this forum is filled with users who find it too difficult to wave their hand in front of their projector. Heck, you don't even need to use a checkerboard pattern if you don't have that, you could use a letterbox bar on any widescreen film. Wave your hand around in the black portion while a bright scene is being displayed. Is there a shadow? Can you see the shadow? Pretty simple.
All these claims about 300:1 being the simultaneous limit for our visual system spread around here come in VERY handy to promote and justify products with inadequate inter-pixel/intra-scene contrast capabilities. I saw it used just 3 days ago for this very purpose. As Chris and Darin put it: It is very easy for everybody to prove these claims wrong. There is a huge benefit going from a 300:1 "ANSI contrast" device to one capable of 500:1-800:1 let alone 2000:1+ (everything else including the full dynamic range of the display ("On/Off contrast") being equal and given a suitable room).

And even if we would take it as granted that ~ 300:1 is the limit the HVS can discern in a "single visual frame" we should/need to have a MUCH higher dynamic range at any given instance in time on the screen as our field of vision constantly changes during watching a movie - therefor we should have a minimum of 300:1 (assuming this number is right which is quite an assumption) of dynamic range no matter "where we look at on the screen" - therefor the full dynamic range at any given instance in time on screen (=intra-scene contrast, the ANSI contrast measurement gives us a hint about that, but depending on the content real world intra-frame contrast performance of a given FP can be much lower than measured ANSI contrast) needs to be (depending on the scene/content) MUCH higher than the limit the HVS can discern in a "single visual frame". This is what the marketing propaganda promoting projectors with 250:1 ANSI contrast doesn't tell you
Quote:

Even the Brightside people seem to agree that "the maximum perceivable
contrast is somewhere around 150:1" for a High Contrast Boundary

http://www.cs.ubc.ca/~mmt/Papers/MscThesis.pdf
Quote:

"I owe a debt of gratitude to everyone at BrightSide Technologies and the Structured Surface Physics Lab for helping me in more ways than I can count"

While different values are reported for the threshold past which we cannot make out high contrast boundaries, most agree that the maximum perceivable contrast is somewhere around 150:1. Scene contrast boundaries above this threshold appear blurry and indistinct, and the eye is unable to judge the relative magnitudes of the adjacent regions. From Moon & Spencer's original work on glare, we know that any high contrast boundary will scatter at least 4% of its energy on the retina to the darker side of the boundary, obscuring the visibility of the edge and details within a few degrees of it. If the contrast of an edge is 25:1, then details on the darker side will be competing with an equal amount of light scattered from the brighter side, reducing visible contrast by a factor of 2 in the darker region. When the edge contrast reaches a value of 150:1, the visible contrast on the dark side is reduced by a factor of 12, rendering details indistinct or invisible. Figure 2.1 shows the model by Deeley at several adaptation luminances.

L* is used in both the CIELAB and CIELUV color spaces, which target print and video respectively, and L* models contrasts approximately 100:1 and a peak luminance of somewhere around 200 cd/m2.
Quote:
Originally Posted by HoustonHoyaFan

http://www.iovs.org/cgi/content/full/42/11/2721

Thanks, your link validates accepted scientific empirical data regarding our HVS gathered over the last century!
I happen to know the laser spot from my pointer is brighter that the brightest white my Pearl can display. And my border is blacker than the black my Pearl can display.

My Pearl has ANSI contrast measure in the ~250:1 range.

I can EASILY see the laser pointer spot at the same time I see the black border.

Haven't I just demonstrated that I can easily exceed 250:1? And I'm pretty sure I'm not at my visual limits (i.e. I can make out the darker shadow the screen border casts on the black wall behind it)

If we were talking 100-200:1 as absolute limts, wouldn't I have had difficulty discerning these additional objects from the image of the Pearl itself?

I just don't think the low numbers pass the "smell test", and it's reasonable to test what the APPLY to, not necessarily their validity in some specific application (such as Chris' border-flare point).
In the test with a black border and a checkerboard pattern for a display with greater than 300:1 ANSI CR, is it possible that the eye adjusts and crushes white to enable someone to see the difference between the black in the checkerboard and the black velvet?
I agree with Chris on this one. The shadow puppet test is a perfect example. Put up an ANSI checkerboard that measures at least 300:1 and then make a shadow puppet in the dark square. I bet you can see it, especially on a projector with an elevated black floor.

The above would be a good test for intrascene contrast visual acuity. If ISF guys are actually talking about a 200:1 on/off contrast ratio (as reported in the OP), well, that's pretty stilly. If they are confusing ANSI with on/off, it's equally bad, and either way it's wrong.

The problem here is that some people aren't using common sense. They are just repeating stuff that they've been told. Come on, guys. We can do better than that.

To me, all this talk about static vs. dynamic, iris vs. non is mostly academic. The important question is whether or not a person can see the difference, since to me it sounds like these 200:1 numbers are being used to assert that higher contrast images are not visibly superior to lower contrast ones. And I assure you, they are. Try the puppet test. Report back.
Quote:
Originally Posted by mrlittlejeans

In the test with a black border and a checkerboard pattern for a display with greater than 300:1 ANSI CR, is it possible that the eye adjusts and crushes white to enable someone to see the difference between the black in the checkerboard and the black velvet?

If the human eye adjusts that fast, it hardly matters. I can actually see the light spill IN my black border if I'm sitting close enough, and I can do so on bright scenes as well as dim, and with no noticeable time for my eyes to adjust.
Quote:
Originally Posted by darinp2

If a setup was showing an ANSI CR of 300:1 off the screen and that was improved to 2000:1 ANSI CR

Why are you talking about ANSI CR? I have said time and time again that there is no direct relationship between the HVS CR number of <300:1 and the ANSI CR number of 300:1. Thats like comparing 300 meters with 300 seconds. It does not work, and you know that. Given that the Ruby does ~300:1 ANSI on a 4x4 and 80:1 ANSI on a 6x6, I would speculate it is ~2:1 ANSI on a single pixel checkerboard which is the closest analogy to a HVS CR IMO.
Quote:
Originally Posted by darinp2

Maybe you aren't even understanding what your chief scientist is saying. I know you said you guys joke about the testing that has been done, but in my theater I put up some images with transition points way beyond 300:1

300:1 ANSI CR?
I guess you are right. I don't understand, the scientists are confused, all the researchers who have mapped the bio structural and biochemical limitations of the retina are clueless.
Quote:
Originally Posted by TheLion

There is a huge benefit going from a 300:1 "ANSI contrast" device to one capable of 500:1-800:1 let alone 2000:1+ (everything else including the full dynamic range of the display ("On/Off contrast") being equal and given a suitable room).

I don't think so if your On-Off is very high. I'm not saying there is no benefit but huge? Not really. Would you have any relevant examples from films where this benefit would be huge if you have say ANSI 200:1 and go to 2000:1 with On-Off high enough to not visibly limit the picture at all?
New Posts  All Forums:Forum Nav:
Return Home
• Some people are still being told that 200:1 is all we can see....

### AVS Top Picks

AVS › AVS Forum › Display Devices › Digital Hi-End Projectors - \$3,000+ USD MSRP › Some people are still being told that 200:1 is all we can see....