Some people are still being told that 200:1 is all we can see.... - Page 10 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #271 of 505 Old 01-04-2007, 07:21 AM
Member
 
epstewart's Avatar
 
Join Date: Jul 2003
Posts: 174
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by ChrisWiggles View Post

I can't speak for HoustonHoya, but I doubt that is a typo. That is, in essence, the basic claim that started this thread.

The essence of some of the claims in the thread is that we cannot see more than a few hundred to one across a single scene. Therefore displays with simultaneous CR performance greater than say 200:1 or so are unnecessary because they already exceed our visual capabilities. There have been a variety of specific numbers suggested that follow that basic claim, 100:1, 130:1, 200:1, 250:1, 300:1. Maybe I'm missing some. But they're all a few hundred to one basically.

It turns out that the ISF is not actually stating this at all and is instead misinterpretation by some dealers who were ISF-trained, and Silver instead cited Poynton's figure of 1,000:1 simultaneous capability across a scene. This is an order of magnitude lower than the 10^5 figure from the Brightside papers which seem more accurate to me intuitively. However, despite the fact that the original post of the thread which seemed to insinuate that the ISF was unfortunately parroting this low 200:1 capability figure was actually not accurate in terms of what the ISF is saying, clearly many people have come to try to defend these low figures as correct. At least for myself, it's very difficult to take such low claims as 200:1 across a scene as the max human capability seriously, but apparently this is fairly commonly held belief. Which is why this thread is already 9-pages long. I am not sure how people are capable of driving around at night with only 200:1 capability, yet somehow...

"Tone mapping for high dynamic range images" is a doctoral dissertation that says (pp. 3-4):

Quote:


The dynamic ranges of several natural scenes were measured and 1:160 was found to be an average contrast ratio (Hunt 1995). Outdoors scenes usually have a larger dynamic range, which can reach a contrast ratio of three orders of magnitude (1:1000) or more. Scenes with fog tend to have a small contrast ratio.

1:1000 or more for outdoors scenes ... that seems to confirm Poynton and refute the idea that the human visual system (HVS) is limited to 1:300 or less. But, as what I am about to say shows, I think the dynamic range capability of the HVS is actually irrelevant.

Folks, look at it this way. If the HVS is limited to a CR of 1:300 or less, then it is intrinsically "low dynamic range" (LDR). If that is so, then the dissertation I just quoted is pointless. There would be no point in developing high dynamic range (HDR) displays for LDR eyeballs.

I believe a source of confusion on this topic is the fact that a "standard display" is LDR. That is, until very recently all TVs and projectors CRTs, LCDs, etc. were LDR. This indeed is an assumption Poynton makes. I am sure it is the assumption the Imaging Science Foundation makes. It is the assumption made by the ATSC and NTSC television broadcasting standards, by Rec. 709 and Rec. 601, etc.

If the display is, by common assumption, LDR, then the signals that are input to it must be encoded accordingly. It's no good inputting a 1:1000 image into a 1:100 display. That would provoke clipping of whites and/or swallowing of shadow detail.

So the dynamic range of the image is reduced in the video camera from what it typically is in the original scene. If the original scene is an HDR one, taken in broad daylight, DR compression is needed. I assume that the same DR compression has to be done even for LDR scenes as well, for the sake of consistency.

I further assume that even if the video camera or other capture device is inherently HDR, the DR of the output image still has to be reduced for an LDR display.

And I further assume that the above is true even if the display is HDR. That is, even though we now have displays and projectors capable of simultaneous contrast ratios well in excess of 1000:1, the video system as a whole can't assume that. It has to assume the end user's display is LDR, since that indeed is the assumption under which the relevant standards were developed.

So it seems fair to summarize the above as: the video systems we have today (ATSC, NTSC, Rec. 709, Rec. 601, etc.) are inherently LDR. Ergo, signals input to the TV are without fail optimized for LDR displays ... even if the display in "my" home theater just happens to be HDR. Consequently, the display will perform its best with these input signals when it is adjusted to reproduce an image using a low dynamic range.

Notice that the question of the actual dynamic range capability of the human visual system completely drops out of the above discussion. It is irrelevant.

Eric Stewart
Catonsville, MD
epstewart is offline  
Sponsored Links
Advertisement
 
post #272 of 505 Old 01-04-2007, 10:17 AM
AVS Club Gold
 
GlenC's Avatar
 
Join Date: Jun 2003
Location: Los Angeles - Whittier, CA
Posts: 2,555
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by epstewart View Post

.....If the display is, by common assumption, LDR, then the signals that are input to it must be encoded accordingly. It's no good inputting a 1:1000 image into a 1:100 display. That would provoke clipping of whites and/or swallowing of shadow detail.

So the dynamic range of the image is reduced in the video camera from what it typically is in the original scene. If the original scene is an HDR one, taken in broad daylight, DR compression is needed. I assume that the same DR compression has to be done even for LDR scenes as well, for the sake of consistency.

I further assume that even if the video camera or other capture device is inherently HDR, the DR of the output image still has to be reduced for an LDR display.

I am not sure how the source CR is an issue, video that we have now is mastered to a standard. SMPTE guidelines were, as I recall, in the range of 16fL. Now, there is a big difference in a 42" screen at 16fL and a 120" screen at 16fL (14'x8' wall). With the 42" at 6sqft, and the wall at 112sqft, it is 5% of your field of view, if is at .5fL our eyes will probably average the light level it is seeing, possibly around 1.27fL. We should still be able to see, @300:1, .004fL to 381fL, however 381/.004=95250:1 On the other hand, if the screen is 120", 49sqft, that is 44%, equating to an average level of 7.32fL, again at 300:1, .024 to 2196fL and 2196/.024=91500:1.

Does this make any sense?

Glen Carter
Home Theater Calibration
www.ISFHT.com
GlenC is offline  
post #273 of 505 Old 01-04-2007, 11:27 AM
Senior Member
 
mdtiberi's Avatar
 
Join Date: Oct 2005
Location: Woodland Hills, CA
Posts: 489
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Confirmation of 1000:1 from Brian A. Wandell, Stein Family Professor and Chair Psychology and Electrical Engineering, Stanford University.

"Within the image reproduction pipeline, display devices limit the maximum dynamic range. The dynamic range of natural scenes can exceed three orders of magnitude. Many films and newer camera sensors can capture this range, and people can discriminate intensities spanning this range. Display devices, however, such as CRTs, LCDs, and print media, are limited to a dynamic range of roughly one to two orders of magnitude."

From the paper Rendering high dynamic range images

I would consider him an unimpeachable source.
mdtiberi is offline  
post #274 of 505 Old 01-04-2007, 12:33 PM
Member
 
armadillo's Avatar
 
Join Date: Jul 2005
Posts: 108
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
mdtiberi: Yes, natural scenes can exceed three orders of magnitude and yes people can discriminate more than three orders of magnitude. Nowhere does he state that the HVS can do so simultaneously without adaptation. Also note the following citation in the discussion:

"In reviewing this material and evaluating algorithm performance, we have noted that an important limitation is that we have only a modest understanding of the illumination and surface reflectance properties in natural scenes. What is the real dynamic range of a scene over the range that a human can discriminate? How does the dynamic range in a scene vary spatially? What are the properties of typical shadow edges? What about surface edges? Obtaining a more precise understanding of scene properties should be a priority for developing good test samples as well as evaluating practical algorithms."
armadillo is offline  
post #275 of 505 Old 01-04-2007, 02:24 PM
AVS Special Member
 
HoustonHoyaFan's Avatar
 
Join Date: Jan 2003
Location: Houston, TX
Posts: 3,963
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 19
Quote:
Originally Posted by darinp2 View Post

Hopefully HHF can clarify, but I think his position might be that for one scene off the screen we need more than 200:1 because we can move our eyes. But that if our eyes stayed looking at one place and the scene was static, we couldn't pick up any differences outside of about 200:1. It is possible that our eyes adjust so fast that it seems to me that I can see multiple steps at once that are well outside 200:1, but that it really isn't instant.

Agreed!

I would also add that since contrast gain adaptation (<100ms>, can be triggered by motion in the visual frame, not just the eye moving, we clearly need >> 200:1 off screen CR, before the HVS becomes a limiting factor in viewing motion pictures!

My summary of the situation

A[pj/screen]-->CR(A)-->B[photoreceptors]-->CR(B)-->C[ganglion]-->CR(C)-->D[visual cortex]

Luminance Output Limits A B C D
Dynamic Range: A 15,000:1(1) B <10^14 C <200:1 D <200:1
Instantaneous CR: A <100:1(2) B < 10^5 C <200:1 D <200:1

Notes
1) JVC RS1. CRT are much higher, 700,000:1??
2) LDR cameras, HDR >1,000:1 ??
CR(A) seems to be source limited to < 100:1 discernable, because of current LDR cameras? Note ANSI CR does not directly measure this. We can clearly see the difference between ANSI CR of 200:1 vs ANSI CR of 1,300:1.

CR(B) has been stated to be 10^5 by some posters. I have not verified.

CR(C) has been referenced/CW as <100:1 in HVS papers. I have posted at least 2.

The CR that we "see" in the visual cortext is limited by the CR that the ganglion cells can output.
HoustonHoyaFan is offline  
post #276 of 505 Old 01-04-2007, 02:31 PM
 
tbrunet's Avatar
 
Join Date: Jul 2005
Location: Savannah, GA
Posts: 2,750
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by mdtiberi View Post

Many films and newer camera sensors can capture this range, and people can discriminate intensities spanning this range. .

mdtiberi, capturing 1000:1 latitude on film is rarely if ever done. 500-600:1 being a more typical peak. Film records information logarithmically, much like the human eye, and not unlike the human (eye/brain) own compression characteristic. HDR digital formats store this dynamic range with special transfer functions. As I have stated many times, it is impossible to represent much more than 100:1 with standard video gamma regardless of the bit depth. It is very limited in both the upper and lower ends of the gamma curve.

To capture the full dynamic range on negative film requires 14 bits per channel in a linear representation. The raw linear representation (LIN) is often converted into a logarithmic representation (LOG) to transfer more dynamic image information across a smaller number of bits.

http://www.tek.com/Measurement/App_N...ary_part_1.pdf
Dynamic Range, Image Capture - The range of luminances actually captured in the image is defined and limited by the transfer function which is usually nonlinear. Capture and recording systems traditionally limit their linear response to a central portion of their dynamic range, and may have extended nonlinear shoulder and toe regions. For any scene, it is usually possible to place the luminances of interest on a preferred portion of the transfer function, with excursions into higher and lower limits rolled off or truncated by the respective shoulder and toe of the curve.
tbrunet is offline  
post #277 of 505 Old 01-04-2007, 02:55 PM
Senior Member
 
mdtiberi's Avatar
 
Join Date: Oct 2005
Location: Woodland Hills, CA
Posts: 489
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Thanks for the feedback guys but I'm just a humble reporter.

Too bad Prof. Wandell did not answer the questions he posed since they would be quite useful as they are the very issues many of us are grappling with (except for Chris cuz he knows everything )
mdtiberi is offline  
post #278 of 505 Old 01-04-2007, 03:06 PM
 
tbrunet's Avatar
 
Join Date: Jul 2005
Location: Savannah, GA
Posts: 2,750
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by mdtiberi View Post

except for Chris cuz he knows everything

tbrunet is offline  
post #279 of 505 Old 01-04-2007, 06:13 PM
 
ChrisWiggles's Avatar
 
Join Date: Nov 2002
Location: Seattle
Posts: 20,730
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:


mdtiberi, capturing 1000:1 latitude on film is rarely if ever done. 500-600:1 being a more typical peak.

Any sources for this claim? I hate to even ask you for references, but given your long and immaculate history of posting complete and utter bullsh*t, it's pretty amazing you expect anyone to believe you. You also continue to believe that 8-bit video can only do 220:1 and thus that no displays can have an on/off CR greater than 220:1.

EDIT: I should probably add that in terms of film being displayed, it's probably not uncommon to see an on/off CR that low in an average theater. That's quite different from what's being captured, however.
ChrisWiggles is offline  
post #280 of 505 Old 01-04-2007, 07:12 PM
Member
 
armadillo's Avatar
 
Join Date: Jul 2005
Posts: 108
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Chris: Is it really asking too much to keep your posts free of personal insults? I believe that most of us have a genuine interest in understanding the issues. Most of the people posting in this thread are fairly open-minded. I would really like to see you respecting differences in opinion as much as we respect your opinion.
armadillo is offline  
post #281 of 505 Old 01-04-2007, 08:07 PM
Member
 
armadillo's Avatar
 
Join Date: Jul 2005
Posts: 108
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
HoustonHoyaFan: I think that is an accurate summary of the discussion so far.
armadillo is offline  
post #282 of 505 Old 01-04-2007, 09:49 PM
Senior Member
 
mdtiberi's Avatar
 
Join Date: Oct 2005
Location: Woodland Hills, CA
Posts: 489
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by HoustonHoyaFan View Post

Agreed!

I would also add that since contrast gain adaptation (<100ms>, can be triggered by motion in the visual frame, not just the eye moving, we clearly need >> 200:1 off screen CR, before the HVS becomes a limiting factor in viewing motion pictures!

My summary of the situation

A[pj/screen]-->CR(A)-->B[photoreceptors]-->CR(B)-->C[ganglion]-->CR(C)-->D[visual cortex]

Luminance Output Limits A B C D
Dynamic Range: A 15,000:1(1) B <10^14 C <200:1 D <200:1Instantaneous CR: A <100:1(2) B < 10^5 C <200:1 D <200:1

Notes
1) JVC RS1. CRT are much higher, 700,000:1??
2) LDR cameras, HDR >1,000:1 ??
CR(A) seems to be source limited to < 100:1 discernable, because of current LDR cameras? Note ANSI CR does not directly measure this. We can clearly see the difference between ANSI CR of 200:1 vs ANSI CR of 1,300:1.

CR(B) has been stated to be 10^5 by some posters. I have not verified.

CR(C) has been referenced/CW as <100:1 in HVS papers. I have posted at least 2.

The CR that we "see" in the visual cortext is limited by the CR that the ganglion cells can output.

Have some questions about some of your statements.

Photoreceptors are not 10^14 according the paper I cited in post #253 Contrast Gain in the Brain ; Geoffrey M. Boynton, The Salk Institute. See page 1, 2nd column paragraph 4. He states clearly one to two orders of magnitude for photoreceptors. I also think it impossible to state a ratio for the visual cortex. The papers that I have read make it very clear that there is much we do not understand regarding how the brain processes information let alone assigning a ratio to contrast perception, bit of a stretch I think. Also your last statement that the brain is limited by ganglion output really needs a reference. Is that's your support for the visual cortex at 200:1 as well?
mdtiberi is offline  
post #283 of 505 Old 01-04-2007, 10:26 PM
AVS Special Member
 
HoustonHoyaFan's Avatar
 
Join Date: Jan 2003
Location: Houston, TX
Posts: 3,963
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 19
[quote=mdtiberi]He states clearly one to two orders of magnitude for photoreceptors. QUOTE]

Above you posted
Quote:
Originally Posted by mdtiberi View Post

Confirmation of 1000:1 from Brian A. Wandell, Stein Family Professor and Chair Psychology and Electrical Engineering, Stanford University.

I would consider him an unimpeachable source.

Which do you believe is correct over all > 10^3 or that the photoreceptors (rods and cones) at the light receiving/facing layer of the retina only have a dynamic range of 10 to 100 (10^1 to 10^2)?

Photoreceptors have the ability to capture a range from a few photons to damage level. That's Bio 101.
HoustonHoyaFan is offline  
post #284 of 505 Old 01-04-2007, 10:35 PM
 
ChrisWiggles's Avatar
 
Join Date: Nov 2002
Location: Seattle
Posts: 20,730
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by armadillo View Post

Chris: Is it really asking too much to keep your posts free of personal insults? I believe that most of us have a genuine interest in understanding the issues. Most of the people posting in this thread are fairly open-minded. I would really like to see you respecting differences in opinion as much as we respect your opinion.

I'm sorry, but you must not be familiar with thomas's posting history. My only mistake was responding at all. Characterizing his posts as "BS" is the most polite way I could possibly put it. If you would like me to be completely honest, I will use much stronger language than that. That user is on my ignore list and has been for a long time. So it actually is asking too much of me to respect the presence and behavior of that user. Given your recent arrival to the forum and these discussions, that's understandable, but no I am not going to respect tbrunet's behavior and it's not fair for you to ask me to. I won't, period. There is a long history there that you are not aware of.
ChrisWiggles is offline  
post #285 of 505 Old 01-05-2007, 04:32 AM
Member
 
epstewart's Avatar
 
Join Date: Jul 2003
Posts: 174
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
The central bone of contention in this thread would seem to be the question of whether the human visual system can make use of a simultaneous contrast ratio in excess of 300:1.

We know that certain TV displays can produce such a ratio. For example, my Sony KDL-40XBR2 LCD panel has an advertised simultaneous or ANSI contrast ratio of 1300:1. PC Magazine measured the ANSI ratio (see this review) at 1205:1.

That was with the backlight all the way up. "With the LCD's backlight set to minimum and peak white levels reduced for dark-room viewing (100 candela/m2 or lower), contrast ratio dropped to a still impressive 550:1."

If the peak white levels were reduced still further, the Sony's simultaneous CR could presumably be lowered to 300:1.

It seems to me that those who say 300:1 represents the limit of human vision must also be saying that we cannot tell the difference between 300:1 and 550:1. That is, adjusting the "picture" control (Sony's name for the contrast control) to produce a measurable simultaneous contrast ratio above 300:1 would make no discernible difference to the eye.

I don't have the instruments to conduct a formal experiment, but I can tell you that with the backlight set to minimum in a dark room (and with the brightness control adjusted properly such that black level is optimized) the Sony's picture control produces a discernible difference across its entire range. I have to assume that as some point in that range, simultaneous or ANSI CR is 300:1, and at some higher point, 550:1. My eyes can definitely see the difference.

Ergo, human vision can in fact make use of a simultaneous contrast ratio well in excess of 300:1.

Eric Stewart
Catonsville, MD
epstewart is offline  
post #286 of 505 Old 01-05-2007, 07:15 AM
Member
 
armadillo's Avatar
 
Join Date: Jul 2005
Posts: 108
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
epstewart: You should do the 5-square test and tell us what happens to the faint square at the various CR.
armadillo is offline  
post #287 of 505 Old 01-05-2007, 09:23 AM
AVS Special Member
 
HoustonHoyaFan's Avatar
 
Join Date: Jan 2003
Location: Houston, TX
Posts: 3,963
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 19
Quote:
Originally Posted by epstewart View Post

I don't have the instruments to conduct a formal experiment, but I can tell you that with the backlight set to minimum in a dark room (and with the brightness control adjusted properly such that black level is optimized) the Sony's picture control produces a discernible difference across its entire range. I have to assume that as some point in that range, simultaneous or ANSI CR is 300:1, and at some higher point, 550:1. My eyes can definitely see the difference.

Are you saying you can tell the difference in a a 4 x 4 ANSI checkboard by changing contrast (white level), or that you can tell the difference in images?
Quote:
Originally Posted by epstewart View Post

Ergo, human vision can in fact make use of a simultaneous contrast ratio well in excess of 300:1.

Note, neither case gives us the data to come to that conclusion.

Just because a display gives one say 1,300:1 measurement on a ANSI 4 x 4 black/white checkerboard does not say that one is getting anywhere close to that on any other type of image!
HoustonHoyaFan is offline  
post #288 of 505 Old 01-05-2007, 09:54 AM
Member
 
armadillo's Avatar
 
Join Date: Jul 2005
Posts: 108
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Think about the following: The HVS takes advantage of an iris to adapt to incident light (no sane person will dispute that). So as the brightness of a scene increases, the iris will close by a certain amount (unless you drop some some atropine into your eyes). This will reduce the number of photons that hit your retina. It is inevitable that faint objects will then fall below the threshold of perception. The open question in this thread is where that threshold is. I can say that on my LCD with nominal 700:1 CR, I can see a faint square that is 2% brightness against the background. Under the best circumstances, this would translate to 350:1 (assuming 100% brightnes delivering 700:1). I can then place a white square next to that 2% square and it will disappear into the background. In fact, even a 3% square will disappear for me (but not a 4% square). So for my particular HVS, the 700:1 CR of my monitor will not allow me to see a 3% square, essentially limiting the effective simultaneous CR to 700/3=233:1. Mind you, I don't think I'm getting 700:1 from my monitor (both contrast and brightness settings are at ~80%). So at least for my particular circumstances, the HVS simply does not provide for more than 200:1 at best.
armadillo is offline  
post #289 of 505 Old 01-05-2007, 10:27 AM
AVS Club Gold
 
GlenC's Avatar
 
Join Date: Jun 2003
Location: Los Angeles - Whittier, CA
Posts: 2,555
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by armadillo View Post

Think about the following: The HVS takes advantage of an iris to adapt to incident light (no sane person will dispute that). So as the brightness of a scene increases, the iris will close by a certain amount (unless you drop some some atropine into your eyes). This will reduce the number of photons that hit your retina. It is inevitable that faint objects will then fall below the threshold of perception. The open question in this thread is where that threshold is. I can say that on my LCD with nominal 700:1 CR, I can see a faint square that is 2% brightness against the background. Under the best circumstances, this would translate to 350:1 (assuming 100% brightnes delivering 700:1). I can then place a white square next to that 2% square and it will disappear into the background. In fact, even a 3% square will disappear for me (but not a 4% square). So for my particular HVS, the 700:1 CR of my monitor will not allow me to see a 3% square, essentially limiting the effective simultaneous CR to 700/3=233:1. Mind you, I don't think I'm getting 700:1 from my monitor (both contrast and brightness settings are at ~80%). So at least for my particular circumstances, the HVS simply does not provide for more than 200:1 at best.

I agree that this is the main issue we face watching video. Size of screen and percent of field of view also a factor.

Glen Carter
Home Theater Calibration
www.ISFHT.com
GlenC is offline  
post #290 of 505 Old 01-05-2007, 10:33 AM
AVS Special Member
 
skogan's Avatar
 
Join Date: Feb 2004
Location: Koror, Palau
Posts: 4,415
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by armadillo View Post

I can then place a white square next to that 2% square and it will disappear into the background.

What happens if you don't put the two squares right next to each other? What if one square is on one side of the screen, and the other is on the other side of the screen? That is important because you want the display to accurately depict both images, depending on which one you look at.

For example, imagine that the black box is instead a picture of a human head, with dark black hair. You would want the display to show the strands of hair on the head, not one big black blob. On the other side of the screen is a window with a girl in a white dress in the back ground. You want to be able to see the wrinkles on her dress.

When focusing in on either image, the other image goes out of focus, and the monitor doesn't know which you are focusing on. So it must be able to replicate each correctly - and that would take greater than 300:1 on screen Ansi CR.

I understand you made concession for this on big screens, but I think that it is an error to think it doesn't apply to smaller screens as well. What really matters is the ratio of "screen size to distance from screen". In other words, sitting close to a small screen is the same as sitting far away from a big screen. And either way, you have the ability to focus down on the area you are interested in.

So the question becomes, does your "fading square" trick work only if the squares are right next to each other on the screen? Or, if one is on the far left, and one is on the far right, do you view each in independent frames? I suspect you do. I suspect the frames are smaller than what you are giving credit for, and for normal TV viewing size and distances, you would have a separate frame for things on the extreme sides of the screen. And for that reason alone, you would want higher CR on screen then what is available per frame in the HVS.
skogan is offline  
post #291 of 505 Old 01-05-2007, 10:57 AM
 
ChrisWiggles's Avatar
 
Join Date: Nov 2002
Location: Seattle
Posts: 20,730
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by armadillo View Post

Think about the following: The HVS takes advantage of an iris to adapt to incident light (no sane person will dispute that). So as the brightness of a scene increases, the iris will close by a certain amount (unless you drop some some atropine into your eyes). This will reduce the number of photons that hit your retina. It is inevitable that faint objects will then fall below the threshold of perception. The open question in this thread is where that threshold is. I can say that on my LCD with nominal 700:1 CR, I can see a faint square that is 2% brightness against the background. Under the best circumstances, this would translate to 350:1 (assuming 100% brightnes delivering 700:1). I can then place a white square next to that 2% square and it will disappear into the background. In fact, even a 3% square will disappear for me (but not a 4% square). So for my particular HVS, the 700:1 CR of my monitor will not allow me to see a 3% square, essentially limiting the effective simultaneous CR to 700/3=233:1. Mind you, I don't think I'm getting 700:1 from my monitor (both contrast and brightness settings are at ~80%). So at least for my particular circumstances, the HVS simply does not provide for more than 200:1 at best.

But as I explained before, the problem with this test is that you do not know the relative difference of the squares with the addition of the white square. Display spill will obscure the visibility as I explained before. This is probably the primary issue that this test shows, not anything to do with your eyes per se. If the display can't render the boundary as distinct and can't maintain the contrast in that scene simultaneously, then you can't see it no matter WHAT your visual capabilities are. That's why I've repeatedly said that this test is largely irrelevant, except insofar as it describes a display's simultaneous CR capability. I've said this a number of times, but I don't think it's quite sunk in yet. The limit in this situation is the display, not your eyes. You can test this by blocking the white square from your field of view. You could also do this by physically masking the white square, but this may exacerbate spill issues as you add something to the light path, so the former method is probably preferred. If you block all the white from your field of view, can you then again see the square boundary? The answer should be no. You've lost visibility of the boundary in both cases, with the presence of white and without the presence of white in your vision, but in both cases white is present through the display system. This means the display is spilling and reducing the CR of that boundary by raising the absolute black level in that scene. The display is the limit, not your eyes.
ChrisWiggles is offline  
post #292 of 505 Old 01-05-2007, 10:59 AM
 
ChrisWiggles's Avatar
 
Join Date: Nov 2002
Location: Seattle
Posts: 20,730
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by skogan View Post

What happens if you don't put the two squares right next to each other? What if one square is on one side of the screen, and the other is on the other side of the screen? That is important because you want the display to accurately depict both images, depending on which one you look at.

For example, imagine that the black box is instead a picture of a human head, with dark black hair. You would want the display to show the strands of hair on the head, not one big black blob. On the other side of the screen is a window with a girl in a white dress in the back ground. You want to be able to see the wrinkles on her dress.

When focusing in on either image, the other image goes out of focus, and the monitor doesn't know which you are focusing on. So it must be able to replicate each correctly - and that would take greater than 300:1 on screen Ansi CR.

I understand you made concession for this on big screens, but I think that it is an error to think it doesn't apply to smaller screens as well. What really matters is the ratio of "screen size to distance from screen". In other words, sitting close to a small screen is the same as sitting far away from a big screen. And either way, you have the ability to focus down on the area you are interested in.

So the question becomes, does your "fading square" trick work only if the squares are right next to each other on the screen? Or, if one is on the far left, and one is on the far right, do you view each in independent frames? I suspect you do. I suspect the frames are smaller than what you are giving credit for, and for normal TV viewing size and distances, you would have a separate frame for things on the extreme sides of the screen. And for that reason alone, you would want higher CR on screen then what is available per frame in the HVS.

Again, I can't emphasize it enough, this test is not a test of your vision at all. It's a test of display performance. The visibility of the boundary or not is overwhelmingly related to the display performance, not visual limits per se. You lose visibility of the boundary because it falls below the visible threshold because the display system is spilling all over it from the white pattern element. This is pretty basic display stuff that we should all be pretty familiar with at this point.
ChrisWiggles is offline  
post #293 of 505 Old 01-05-2007, 11:02 AM
 
ChrisWiggles's Avatar
 
Join Date: Nov 2002
Location: Seattle
Posts: 20,730
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by HoustonHoyaFan View Post

Just because a display gives one say 1,300:1 measurement on a ANSI 4 x 4 black/white checkerboard does not say that one is getting anywhere close to that on any other type of image!

Except that if the on/off CR is higher than the ANSI CR, which is almost always the case (significantly so), then actually in most scenes you are getting higher CR than the ANSI CR. ANSI CR is a 50% APL pattern, which is actually quite high in APL, as most scenes are significantly below that, probably less than half that. So you're going to be getting a lot more instantaneous contrast in those images than even then ANSI CR pattern (which is something of an extreme pattern) achieves. See the "time to define an alternative to ANSI CR" thread for more discussions about this.
ChrisWiggles is offline  
post #294 of 505 Old 01-05-2007, 11:10 AM
 
ChrisWiggles's Avatar
 
Join Date: Nov 2002
Location: Seattle
Posts: 20,730
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by epstewart View Post

The central bone of contention in this thread would seem to be the question of whether the human visual system can make use of a simultaneous contrast ratio in excess of 300:1.

We know that certain TV displays can produce such a ratio. For example, my Sony KDL-40XBR2 LCD panel has an advertised simultaneous or ANSI contrast ratio of 1300:1. PC Magazine measured the ANSI ratio (see this review) at 1205:1.

That was with the backlight all the way up. "With the LCD's backlight set to minimum and peak white levels reduced for dark-room viewing (100 candela/m2 or lower), contrast ratio dropped to a still impressive 550:1."

If the peak white levels were reduced still further, the Sony's simultaneous CR could presumably be lowered to 300:1.

It seems to me that those who say 300:1 represents the limit of human vision must also be saying that we cannot tell the difference between 300:1 and 550:1. That is, adjusting the "picture" control (Sony's name for the contrast control) to produce a measurable simultaneous contrast ratio above 300:1 would make no discernible difference to the eye.

I don't have the instruments to conduct a formal experiment, but I can tell you that with the backlight set to minimum in a dark room (and with the brightness control adjusted properly such that black level is optimized) the Sony's picture control produces a discernible difference across its entire range. I have to assume that as some point in that range, simultaneous or ANSI CR is 300:1, and at some higher point, 550:1. My eyes can definitely see the difference.

Ergo, human vision can in fact make use of a simultaneous contrast ratio well in excess of 300:1.

The problem with this method is that ANSI CR is most heavily defined by the nature of physical light spill in the system. That is to say, it's a ratio that maintains a fairly fixed ratio based on the physical behavior of the display system. If you move the white point around, the ratio of spill happening is not changing in most all cases because you are not physically modifying the display (for most displays). If you move white/black points around(conservatively), you'll be having a serious impact on the on/off CR through the system, but minimal or no impact on the ANSI CR assuming that on/off CR is significantly greater than ANSI CR, which is almost universally is. At some point on/off could become so low that it is approaching the low ANSI CR and will be contributing in that pattern situation, so it could lower that measurement, but generally speaking with reasonable white/black points ANSI CR measures function as a kind of fixed ratio that relate primarily to the physical nature of the display and the light path, spill etc. Unless you physically modify the display or the display environment (i.e. the room), then it is unlikely that ANSI CR will really change much if at all.
ChrisWiggles is offline  
post #295 of 505 Old 01-05-2007, 11:32 AM
AVS Addicted Member
 
darinp2's Avatar
 
Join Date: Sep 2003
Location: Seattle, WA
Posts: 21,165
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 43
Quote:
Originally Posted by armadillo View Post

I can say that on my LCD with nominal 700:1 CR, I can see a faint square that is 2% brightness against the background. Under the best circumstances, this would translate to 350:1 (assuming 100% brightnes delivering 700:1). I can then place a white square next to that 2% square and it will disappear into the background. In fact, even a 3% square will disappear for me (but not a 4% square). So for my particular HVS, the 700:1 CR of my monitor will not allow me to see a 3% square, essentially limiting the effective simultaneous CR to 700/3=233:1. Mind you, I don't think I'm getting 700:1 from my monitor (both contrast and brightness settings are at ~80%). So at least for my particular circumstances, the HVS simply does not provide for more than 200:1 at best.

What is 2%, 3% and 4% brightness above. If these are luminances that is one thing, but if they are like IRE or % stimulation before gamma, then 4% with even a 2.2 gamma would be less than .1%.

I admit that I haven't run through the test, but in the above it sounds like you are saying that you can't see it at 233:1 because your CR is 700:1, but this doesn't seem to say anything about whether you would see it if the CR for the display all of a sudden went to something large like 100k:1. In other words, if you can't see it at 233:1, but would see it at higher CRs then that doesn't tell you that your eye can't see bigger CRS, it tells you that you can't see smaller CRs (at least in a particular case). Please correct me if I am misunderstanding something about your test. As I've tried to address multiple times, if we want to know the upper limit then we shouldn't test for the lower limit, we should test for the upper limit where no more improvement can be detected.

--Darin

This is the AV Science Forum. Please don't be gullible and please do remember the saying, "Fool me once, shame on you. Fool me twice, shame on me."
darinp2 is offline  
post #296 of 505 Old 01-05-2007, 11:37 AM
Member
 
armadillo's Avatar
 
Join Date: Jul 2005
Posts: 108
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
skogan: I completely agree with everything you said. Obviously, it all depends on the field of view you are inspecting. The HVS will adjust occording to that image (be it a small screen in the distance or a subimage you look at in huge projection screen). Let me re-state that higher CR is a desirable feature and I will fight companies that give them to us.
armadillo is offline  
post #297 of 505 Old 01-05-2007, 11:46 AM
AVS Club Gold
 
GlenC's Avatar
 
Join Date: Jun 2003
Location: Los Angeles - Whittier, CA
Posts: 2,555
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by skogan View Post

What happens if you don't put the two squares right next to each other? What if one square is on one side of the screen, and the other is on the other side of the screen? That is important because you want the display to accurately depict both images, depending on which one you look at.

The display is accurately displaying, the issue is that when they are inclose proximity, the eye cannot see the full range.

Quote:
For example, imagine that the black box is instead a picture of a human head, with dark black hair. You would want the display to show the strands of hair on the head, not one big black blob. On the other side of the screen is a window with a girl in a white dress in the back ground. You want to be able to see the wrinkles on her dress.

Agreed, but when looking at the dress and then you look at the head, you may not be able to see the hair strands for a moment. If the TV is small enough you may never be able to see the hair because of the light output of the white dress.

Quote:
When focusing in on either image, the other image goes out of focus, and the monitor doesn't know which you are focusing on. So it must be able to replicate each correctly - and that would take greater than 300:1 on screen Ansi CR.

Even if the display does show the hair detail, you may not see it. consider a person standing inside a open doorway with daytime sky behind, can you see the strands of black hair, or do you need to close the door or add more light to the room.

Quote:
I understand you made concession for this on big screens, but I think that it is an error to think it doesn't apply to smaller screens as well. What really matters is the ratio of "screen size to distance from screen". In other words, sitting close to a small screen is the same as sitting far away from a big screen. And either way, you have the ability to focus down on the area you are interested in.

See previous comment and if you are driving into the setting sun, the car approching looks like a shadow, no matter how you focus on it you cannot see the details until you block the sun from your eyes.

Quote:
So the question becomes, does your "fading square" trick work only if the squares are right next to each other on the screen? Or, if one is on the far left, and one is on the far right, do you view each in independent frames? I suspect you do. I suspect the frames are smaller than what you are giving credit for, and for normal TV viewing size and distances, you would have a separate frame for things on the extreme sides of the screen. And for that reason alone, you would want higher CR on screen then what is available per frame in the HVS.

What good is a higher contrast ratio on the screen if you cannot see it. Higher contrast ratios basically mean brighter whites and or blacker blacks.

Lets just say you have a projector with a on/off CR of 15000:1 If you display a 0-10 IRE stair step pattern (1 IRE steps), you will see all steps, if you put up the same 90-100, you will see all the steps. If 100 IRE is 15 fL, 0 IRE is .001 fL. The CR between 0 IRE and 3 IRE is 2.3:1, 3 to 100 is 6415:1 and 10 to 100 is 316:1 @2.5 gamma (assumes perfect performance, actual would be lower). Calculated 10 IRE is .047 fL. I would agree that if you put up a white square on one side of the screen and a 1 IRE square on the other side, you might be able to see it. As you increase the size of the white square the 1 IRE may still be there, but as the white gets bigger, your eye closes down to adjust for the brighter light and you will at one point no longer be able to see the 1 IRE square. The same may happen as they get closer together, it is not a "trick" it just a human limitation. There are three major factors in the actual situation, the persons eye, the environment and the display system performance. Some displays hold their black level better than others, this is generally seen in ANSI CR performance.

Glen Carter
Home Theater Calibration
www.ISFHT.com
GlenC is offline  
post #298 of 505 Old 01-05-2007, 11:54 AM
 
ChrisWiggles's Avatar
 
Join Date: Nov 2002
Location: Seattle
Posts: 20,730
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by GlenC View Post

The display is accurately displaying, the issue is that when they are inclose proximity, the eye cannot see the full range.

Agreed, but when looking at the dress and then you look at the head, you may not be able to see the hair strands for a moment. If the TV is small enough you may never be able to see the hair because of the light output of the white dress.

Even if the display does show the hair detail, you may not see it. consider a person standing inside a open doorway with daytime sky behind, can you see the strands of black hair, or do you need to close the door or add more light to the room.

See previous comment and if you are driving into the setting sun, the car approching looks like a shadow, no matter how you focus on it you cannot see the details until you block the sun from your eyes.

What good is a higher contrast ratio on the screen if you cannot see it. Higher contrast ratios basically mean brighter whites and or blacker blacks.

Lets just say you have a projector with a on/off CR of 15000:1 If you display a 0-10 IRE stair step pattern (1 IRE steps), you will see all steps, if you put up the same 90-100, you will see all the steps. If 100 IRE is 15 fL, 0 IRE is .001 fL. The CR between 0 IRE and 3 IRE is 2.3:1, 3 to 100 is 6415:1 and 10 to 100 is 316:1 @2.5 gamma (assumes perfect performance, actual would be lower). Calculated 10 IRE is .047 fL. I would agree that if you put up a white square on one side of the screen and a 1 IRE square on the other side, you might be able to see it. As you increase the size of the white square the 1 IRE may still be there, but as the white gets bigger, your eye closes down to adjust for the brighter light and you will at one point no longer be able to see the 1 IRE square. The same may happen as they get closer together, it is not a "trick" it just a human limitation. There are three major factors in the actual situation, the persons eye, the environment and the display system performance. Some displays hold their black level better than others, this is generally seen in ANSI CR performance.

*sigh*

I am bashing my head against the keyboard.
ChrisWiggles is offline  
post #299 of 505 Old 01-05-2007, 12:43 PM
Member
 
armadillo's Avatar
 
Join Date: Jul 2005
Posts: 108
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Chris: You asked the correct question. If I remove the white square, do I see the square again? The answer is YES. In fact, I can move my foveal area so that the white square is off center and the square reappears (I already mentioned that in a previous post; so it's not the system that causes this but my HVS).
armadillo is offline  
post #300 of 505 Old 01-05-2007, 12:51 PM
AVS Club Gold
 
GlenC's Avatar
 
Join Date: Jun 2003
Location: Los Angeles - Whittier, CA
Posts: 2,555
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by ChrisWiggles View Post

*sigh*

I am bashing my head against the keyboard.

Don't hurt your self......... I really don't understand your problem with others giving their opinions. You apparently don't like or agree with what I have said, but it is not wrong.......

I have basically stated that we, as a human factor, may not be able to see what the projector is fully capable of producing. If you can see detail 2' from the screen but you cannot see it from 15', that is not a problem of the display.

As far as I can see in this thread. is a failure to fully define the meaning of the statement "200:1 contrast is all that we can see" Without that defined, nobody has any authoritative claim on opinions being right or wrong, unless there is clearly an error. There are way too many potential variables involved. As I stated before, display a full 100 IRE field and look at it. Now replace it with a 0-20 IRE stair pattern (5 IRE steps) and see which step you can see before the eye runs out of range. That first step would be your instantaneous CR from the screen fL and percent field of view. This will be different, depending on age, health, individual and many other factors. This would be a valid test if I had said "200:1 instantaneous CR is all we can see in a theater environment."

Glen Carter
Home Theater Calibration
www.ISFHT.com
GlenC is offline  
Reply Digital Hi-End Projectors - $3,000+ USD MSRP

User Tag List

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