Opinion on Contrast - Page 2 - AVS Forum
Forum Jump: 
Closed Thread
 
Thread Tools
post #31 of 195 Old 06-30-2006, 07:41 PM
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 HoustonHoyaFan
People may not like it, but most medical research supports the conclusion that the simultaneous contrast range for the HVS is < 200:1!
Do you have some medical research that shows this for the simultaneous contrast range (not the inverse of the simultaneous contrast range). Most people just get confused and quote things that are 1:200 as if they have a contrast ratio of 200:1. If I told people that I wanted to know the biggest step a person could take, would they use research that tested the smallest step a person could take in order to answer that question? For instance, let's say there was some research that the smallest step was 1/30th of an inch, but the way the test was scored was to use the inverse or reciprocal (just like CSF scores do) and so that would be a score of 30. Anybody who then took that 30 to use as an answer toward the question of what the biggest step a person could take would obviously not understand the research. And it seems like the same thing happens here with people using CSF scores of things like 100:1 or 200:1 that are CRs of around 1.01:1 to relate to maximum CR a person can see. They don't relate and a CSF score of 200:1 tells you pretty much zero about what the maximum CR before a person can detect higher CRs is.
Quote:
Originally Posted by HoustonHoyaFan
The problem occurs when people then try to use those numbers to say one only needs 200:1 on/off or even 200:1 ANSI CR, which is clearly incorrect.
They are clearly incorrect for multiple reasons, but the main one is probably because 200:1 CSF is like testing if a person can see a white cat in front of a white wall and 200:1 CR is like looking at a black cat in front of a white wall. Pretty much unrelated and one doesn't tell you anything about the other.
Quote:
Originally Posted by HoustonHoyaFan
Put another way, if one had a device which could produce 200:1 per pixel checkerboard, then we can have a conversation. :)
If it had a low on/off CR then the blacks would still suck in many scenes. And why a 200:1 per pixel checkerboard when the CSF test isn't anything like that (it is more like a 1.005:1 per pixel checkerboard)?

--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  
Sponsored Links
Advertisement
 
post #32 of 195 Old 06-30-2006, 08:02 PM
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 tbrunet
At a zone or tone of zero (black) the illumination is still 3.1%.
In the past you told us that the highest CR possible with 8 bits was 219:1, which would make black just a little under 5% of the illumination of white. Are you now changing your answer to about 32:1 for the highest CR (black at 3.1% of white)?
Quote:
Originally Posted by Gary Lightfoot
I did a comparison of some scenes I'm familiar with on a Barco 808 CRT and found that I couldn't see any more detail than I could with my own DLP projetor which only has around 2700:1 on/off CR (in fact the DLP had more shadow detail due to an incorrect black level on the Barco, but that was later corrected). I then had the opertunity to directly compare the same DLP with a Sony G90 CRT doing an A/B test (in the same room) and found the same results.
While I think tbrunet should probably be banned for continuing to try to mislead people, it is sometimes funny to see what kinds of ridiculous tactics he will use in his quest. One of them is to quote Dr. Soneira's credentials when he wants to claim that Dr. Soneira must be correct, and ignore them when he is claiming Dr. Soneira is wrong (tbrunet claims that the highest CR possible with 8 bits is 219:1 and Dr. Soneira claims something like 192,000:1, both of which are incorrect, BTW).

On the subject of shadow detail most people here can do a simple test to see that the stuff tbrunet quotes from Dr. Soneira about the bottom 8 steps or so being indistinguishable from each other on a display with just 2000:1 is just not true, just like you found Gary. There are quite a few displays here with close to that CR and all people have to do is plug in AVIA, DVE, THX or do many other tests which will show that these levels are not indistinguishable. In fact, if the displays Dr. Soneira used in his comparison that had less than 2000:1 on/off CR did have those levels as indistinguishable from video black (level 16) then they were severely miscalibrated. If it were true then the first pluge bar above video black would be gone and it will be visible on a properly setup display with 2000:1 CR. It's basically a 5 minute test to see if the statement is true or not and anybody doing things right will find that it isn't true. But I don't expect the fact that it is so easy to test and disprove to keep tbrunet from trying to make people believe it is true, given his past behavior.

--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 #33 of 195 Old 07-01-2006, 06:45 AM
 
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
Computers do use log base2 arithmetic for their internal data representations. The properties of light based on the inverse square law are also based on log base2 arithmetic. There is no correlation at all between these two fundamental properties. There is no such thing as digital gamma. Digital encoding does not compress tones.

http://www.rags-int-inc.com/PhotoTechStuff/TonesnZones/
Quote:
The most serious misconception is that because of the computer’s binary ordering, there are digitally more highlight tones in an image than shadow tones. This misguided perception is based on the premise that each bit in a digital number represents twice or half the values of the adjacent bits.

Each zone represents double or half of the light (one exposure stop) in the adjacent zone, so they follow a logarithmic progression (powers of two) in terms of the amount of light. But they still represent a linear progression in terms of their tonal values.
The data demonstrates the fact that the exposure verses tone relationship is not linear or logarithmic. Over exposure will clip tonal details quickly. Under exposure also distorts tonal values, but it retains shadow detail longer and can be recovered easier. This should not be news to any digital photographer.
It also demonstrates the fact that a relatively small miscalculation in the base exposure can result in a somewhat significant shift in tonal values. Coupled with the fact that each color channel (RGB) reacts differently, some (Darin :p ) might consider digital imaging pure magic.
A typical curve for film will have a characteristic S shape. It represents a mathematical non-linear periodic function similar to the curves in a sine wave. Mathematically, Fourier transforms are sometimes used to generate a curve of this shape. These should not be confused with gamma as they track sensitivity limits near the extremes of a function instead of logarithmic progressions. As with gamma most electronic and chemical processes exhibit some sort of s-curve function at their practical design limits. As the voltage reaches its clipping points, minimum and maximum, changes in brightness have smaller effects. We call this tone compression.

***There is tone compression in the shadows. But it is due to the properties of light. It has nothing to do with whether the numbers have been digitized or not. It has nothing to do with the minimum or maximum density of any single device. It has nothing to do with the gamma of any single device, including the eyeball.
tbrunet is offline  
post #34 of 195 Old 07-01-2006, 07:02 AM
 
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 darinp2
(tbrunet claims that the highest CR possible with 8 bits is 219:1 and Dr. Soneira claims something like 192,000:1, both of which are incorrect, BTW).
Actually No!..one is defined by the theoretical, and the other "real world"


Computer Systems, Colourspaces, Monitors and Printers:
The most commonly used image formats like Jpeg don't have any embedded colourspace, but assume an implicit gamma to compensate for the non-linearity of CRT monitor phosphors. This gamma value is typically 1.8 or 2.2 depending on which system you are using. Attempts to use a linear colourspace (gamma=1.0) are controversial, and have not been adopted by the mainstream. Monitors are not perfectly linear and cannot accurately display a full 256:1 ratio of brightness levels (badly calibrated ones in bright ambient light show even less useful range). Printed paper has quite a low range (less than 50:1).

Human Visual Perpection:
The human visual system is non-linear and is roughly logarithmic over large ranges in brightness. An object in shade might seem "almost" as light as the same thing in sunlight, but an exposure meter might show 2.3 stops less light (5 times darker). The typical range of brightness levels of adjacent areas that can be distinguished by eye is about 100:1 (with dark areas being swamped by bright areas). The human visual system has a very large dynamic range, but limited precision within that range.
tbrunet is offline  
post #35 of 195 Old 07-01-2006, 07:42 AM
 
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
Darin (before I'm banned :)) please view the image Figure 4 in this link. Thanks in advance!

http://www.rags-int-inc.com/PhotoTechStuff/DigitalFilm/
This is a contrived image intended only to illustrate the concept of dynamic range. It is possible to see the moon and the sun in the same sky, because they fit within the dynamic range of our vision. It is impossible to see the stars behind the moon in daylight because the sun has overwhelmed any contrast. The earth bound objects we see are illuminated by the sun. This city shoreline had an exposure value of approximately EV +15. The starry sky would be approximately EV -6. The moon itself is about EV+14, but it would only illuminate this shoreline at about EV -2. The direct noon sun itself is at least EV +22. The dynamic range of this scene (if it could exist) would be about EV 28. No single film, sensor, or eyeball can take it all in at one view.
Dynamic Range and Contrast
The next important metric would be the dynamic range or the ability to detect and discriminate contrast. The eye can resolve about 7 to 10 stops of light (contrast) at a single glance. But a sun and shade daylight scene can easily contain 15 stops of light. So even with the eye, dark and light tones can wind up compressed with loss of detail. The eye can adjust from darkness to a bright scene in about 5 minutes. It takes up to 30 minutes to fully adjust from strong light to darkness. We do the same thing with film or digital images by varying the exposure. And we see the same artifacts, blown highlights or loss of shadow detail.
tbrunet is offline  
post #36 of 195 Old 07-01-2006, 07:59 AM
AVS Club Gold
 
Art Sonneborn's Avatar
 
Join Date: Jun 2000
Location: Battle Creek,MI USA
Posts: 22,302
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 44
Quote:
Originally Posted by Gary Lightfoot
So I would say that having greater than 2000:1 on/off CR will mean that you are seeing all the detail that is available from the source, but if you want a better black level you will need higher CR.


Gary
Bottom line ! :) If a picture area has no detail in the areas that are supposed to be very very dark or black and it can be rendered as black then the three dimensionality, punch and how engaging it is goes up dramatically.

Of course ,this doesn't really start to come up until one gets into moderately low to low APL but anyone who enjoys very very high sequential contrast then sees a display that doesn't have it will be disappointed.

Art

My HT


iRule rules my theater
 

"If she's amazing she won't be easy,if she's easy she won't be amazing"

 

Bob Marley

Art Sonneborn is offline  
post #37 of 195 Old 07-01-2006, 11:04 PM
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 tbrunet
Darin (before I'm banned :)) please view the image Figure 4 in this link. Thanks in advance!
If you want others to review and comment on things you request then you should answer simple questions that have been asked of you and not play games to avoid them like you are doing. I've answered a lot of things from you while you played your games of avoidance. I will comment on your stuff if you answer one of skogan's questions with a real answer:
Quote:
Originally Posted by skogan
So, tbrunet, what is the highest on/off CR possible with 8 bit video?
You already proved me right once when I said that you wouldn't give actual answers to questions like that, so you can of course prove me right again since I think you will continue avoiding that question from skogan as you have avoided other questions for months.

In fact, even after all the trashing you have done of some of the most knowledgeable people around here on this subject matter, I bet you don't even know enough about the subject matter to fill in correct numbers for what the ??s should be in the following with the 8 bit type of video encodings on DVD and HD DVD and the 10 ft-lamberts for the 100%stim case:

100%stim (video 235) = 10 ft-lamberts
50%stim (video 125) = ?? ft-lamberts
25%stim (video 71) = ?? ft-lamberts
0%stim (video 16) = ?? ft-lamberts

You can of course go ahead and prove me right once again by not providing any answers for those.

--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 #38 of 195 Old 07-01-2006, 11:56 PM
Senior Member
 
faterikcartman's Avatar
 
Join Date: Jan 2005
Location: San Diego
Posts: 379
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
I've skimmed this and won't spend my time reading all the links in the debate between Darin and the crazy guy.

Still, I will use my genius to offer this possible reconcilliation:

Let's say the crazy guy is right and the human eye can spot or recognize 100 steps of gradiation in contrast. But electronic sensors can spot 10,000.

It may be that the human eye can only spot all 100 steps when exposed to the 10,000 step stretch observed by electronic sensors.

Or, the human eye's recognition of steps may be relative so that the human eye can only spot 100 steps between 1,000 to 1 electronic sensor measured CR or 100 steps between 10,000 to 1 electronic sensor measured CR but that the 100 steps measured between the 10,000 to 1 range from an absolute brighter white and blacker black so that while the human eye may only recognize 100 steps in that range their greater difference between absolutes is more pleasing and results in an impression of more depth and 3D like image.

I'm guessing the crazy guy did not consider this thinking.

I leave it to you all to consider and feel free to let me know if I need to further explain what I'm saying. While I didn't stay at a Holiday Inn Express last night, I really am a genius.

YMMV

By definition 50% of people are below the median.
faterikcartman is offline  
post #39 of 195 Old 07-02-2006, 12:32 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 faterikcartman
Or, the human eye's recognition of steps may be relative so that the human eye can only spot 100 steps between 1,000 to 1 electronic sensor measured CR or 100 steps between 10,000 to 1 electronic sensor measured CR but that the 100 steps measured between the 10,000 to 1 range from an absolute brighter white and blacker black so that while the human eye may only recognize 100 steps in that range their greater difference between absolutes is more pleasing and results in an impression of more depth and 3D like image.
Yep, except that higher on/off CR doesn't mean brighter whites, so it could just be the blacker black that the person can perceive and appreciates just because it looks better or because it is more like real life. And if there are only 5 steps instead of 100 and they are spread out way more because of higher on/off CR then people could perceive an improvement (with the amount of extra spread needed varying depending on how those 5 steps are layed out). The Just-Noticable-Difference (JND) comes in also, where somebody may claim that in a particular image a person can't perceive some transition down low and so high CR isn't needed, but having a weak on/off CR can lead to a person not seeing a transition they could have seen if the CR had been higher (because the darker background would have meant more difference between the background and the thing just a little brighter and possibly getting over the JND threshold). Just doing math for an example JND would show the advantage of higher CR on the ability to see certain low transitions in some images. There are a lot of things that go into this, but when a person won't even answer a straightforward question about how much light should be shot on the screen for black, I think it is pretty obvious that they don't actual want to comprehend these things.

The main reason for high on/off CR being an advantage in normal video isn't for seeing 100 steps at one time, it is for seeing darker images in ways that are more like the real world where even dark images can have very large simultaneous CRs. Heck, even if a person only wanted 100:1 in any image, it would take something like 10k:1 on/off CR to have 100:1 or more simultaneous CR in a 10%stim/0%stim image and even more to maintain that 100:1 in images darker than that.

--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 #40 of 195 Old 07-02-2006, 06:44 AM
 
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 darinp2
You can of course go ahead and prove me right once again by not providing any answers for those.
FWIW I have learned that a displays CR is an attribute to a certain point (…how much is perceived or needed I leave that to your resident experts. Although grayscale tracking is probably the single most important..I suggest anyone interested google the name Peter Putman and contrast ratio.

DARIN for the love of GOD!! The density for 2^8 bits (linear/non-linear) the CONTAST Maximum is:

2^8 = CR 255:1
Log (2^8) = Density ~ 2.4 = 255:1

“Gamma†has no affect on this figure. :eek:

Darin, I provided a link http://www.rags-int-inc.com/PhotoTechStuff/DigitalFilm/
& image in Figure 4 will put to rest the SM of the performance envelope of the HVS. Look don’t look ..a picture is worth a thousand 8 bit words:)

http://www.rags-int-inc.com/PhotoTechStuff/DigitalFilm/
Quote:
This is a contrived image intended only to illustrate the concept of dynamic range.
The dynamic range of a device is the difference (contrast) between the minimum and maximum signal it can faithfully record. It is sometimes expressed as a ratio between the minimum and maximum radiance (decibels). Image density is frequently expressed as brightness measured with a densiometer on a logarithmic scale of 0 to 4. A density of 3.0 is 10 times greater intensity than a density of 2.0. A contrast range of 100:1 is a density range of 2.0, and 1000:1 is a range of 3.0
tbrunet is offline  
post #41 of 195 Old 07-02-2006, 09:39 AM
 
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 MrWigggles
tbrunet is nuts and while he posts information that is often correct, it is often not applicable for the given situation. I have not even tried to read the links he posted this time.
WOW ;)
I honestly don't understand why this SM is soo provocative! Mention the word CR and Darin goes on and on about ANSI and ON/OFF!

I will concede, marginal ANSI headroom is a benefit.. (Peter Putman does as well)

But no matter how you cut it, the CIELAB has a reasonable grasp on the SM, & I suggest they understand at what point contouring rises above the perception threshold. It been accepted for some time now that ~1% delta is at the HVS threshold of perception.

If it were as Darin suggest .1%, then 8 bits would not be acceptable!
tbrunet is offline  
post #42 of 195 Old 07-02-2006, 11:52 AM
AVS Special Member
 
tvted's Avatar
 
Join Date: Oct 2003
Location: Toronto, Canada
Posts: 3,573
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by tbrunet
http://www.rags-int-inc.com/PhotoTechStuff/DigitalFilm/
The dynamic range of a device is the difference (contrast) between the minimum and maximum signal it can faithfully record.
Though I think this is an absolute waste of typing to ask...

Thomas,
If I remember correctly, your long time contention has been that bit depth = tonality = dynamic range. I also seem to recall that you do not believe that film is encoded such that a dynamic range of 10 bits can be represented within an 8 bit space. Correct me if my recollection is mistaken.

From the page you linked to above:
Quote:
The next metric is the bit-depth. This tells us how many discrete tonal differences can be represented in a pixel. Of course each pixel represents only one of these tones. More bits per pixel results in more tones and larger file sizes (Mega Bytes). Don’t confuse bit-depth with dynamic range. Dynamic range defines the difference between the darkest and lightest tones captured from a scene. Bit-depth defines the granularity of the individual tones in between. If dynamic range was a ladder, bit-depth would define the number of steps on the ladder, not its height. Dynamic range would describe the height of the ladder.
I've also recall you claiming the eye's ability to resolve dynamic range after adaption to be in the range of 50 to 100:1

From within the same article:
Quote:
The next important metric would be the dynamic range or the ability to detect and discriminate contrast. The eye can resolve about 7 to 10 stops of light (contrast) at a single glance.
Even his low estimate is higher than a maximum of 100:1

Will you comment on the bit depth vs. dynamic range quote as well as the seeming contradiction of your thoughts regarding the eye's abilty to resolve dynamic range at a single glance?.

Your thoughts not some web mobius strip.

ted
tvted is offline  
post #43 of 195 Old 07-03-2006, 07:11 AM
 
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 tvted
Though I think this is an absolute waste of typing ...
Agreed!

Quote:
Originally Posted by tvted
I've also recall you claiming the eye's ability to resolve dynamic range after adaption to be in the range of 50 to 100:1
Is lying part of your agenda as well..I have said repeatedly :eek: ..and it's back up by countless research on the SM (~100 or 150:1)

According to Darin 1,000,000:1 :o BTW I base my opinion on empirical data

Quote:
Originally Posted by tvted
Will you comment on the bit depth vs. dynamic range quote as well as the seeming contradiction of your thoughts regarding the eye's abilty to resolve dynamic range at a single glance?.
A very important consequence of gamma correction is perceptual uniformity of the corrected tristimulus signals (R', G', B') and their relative luminance. The perception of contrast levels in human vision is quite nonlinear and is called lightness. The basis of the argument comes from this excerpt from the classical psychological experiments which led to the formulation of Weber-Fechner's Law:

Lightness (a perceptual quantity) is approximately luminance raised to the 0.4 power.
tbrunet is offline  
post #44 of 195 Old 07-03-2006, 07:19 AM
 
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 tvted
I also seem to recall that you do not believe that film is encoded such that a dynamic range of 10 bits can be represented within an 8 bit space. Correct me if my recollection is mistaken.
Correct you are mistaken!!
I suggest you research the concept of DR *Compression* ;)
tbrunet is offline  
post #45 of 195 Old 07-03-2006, 07:39 AM
AVS Special Member
 
tvted's Avatar
 
Join Date: Oct 2003
Location: Toronto, Canada
Posts: 3,573
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by tbrunet

Is lying part of your agenda as well..I have said repeatedly :eek: ..and it's back up by countless research on the SM (~100 or 150:1)
Oh my!

As I said "recollection...correct me if I'm wrong".
Agenda? :rolleyes:

You did, however, fail to comment on the very claims made in the article you linked to. Unfortunately this is not surprising given past behavior.

Quote:
Originally Posted by tbrunet
A very important consequence of gamma correction is perceptual uniformity of the corrected tristimulus signals (R', G', B') and their relative luminance. The perception of contrast levels in human vision is quite nonlinear and is called lightness. The basis of the argument comes from this excerpt from the classical psychological experiments which led to the formulation of Weber-Fechner's Law:

Lightness (a perceptual quantity) is approximately luminance raised to the 0.4 power.
Thanks for the lecture - none of this is unknown to me or I suspect most trying to decipher your thinking. Nice to see you agreed with my first instinct. My fault for not taking it seriously.

Instead of addressing the simple questions derived from the sources you post, you choose to attack individuals posting to this thread. Are you sure the agenda is not yours? You could be the most knowledgeable person on earth in these matters but you would deserve dismissal simply because of your decorum.

I truly would like to know what displays you have calibrated and what tools you have utilized in said endeavors. But, never mind, I'm not sure that the answers gained would tell anything.

ta ta,
ted
tvted is offline  
post #46 of 195 Old 07-03-2006, 07:46 AM
AVS Special Member
 
tvted's Avatar
 
Join Date: Oct 2003
Location: Toronto, Canada
Posts: 3,573
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by tbrunet
Correct you are mistaken!!
I suggest you research the concept of DR *Compression* ;)
As links I've directed you to in the past have confirmed.

So are you now agreeing that source can provide more Contrast than linear 8 bit can provide if linearly decoded? If so, how do we regain that CR when displayed, or do you believe that it remains compressed for display?

ted
tvted is offline  
post #47 of 195 Old 07-03-2006, 07:49 AM
 
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 MrWigggles
tbrunet is nuts
Quote:
Originally Posted by tvted
You could be the most knowledgeable person on earth in these matters but you would deserve dismissal simply because of your decorum.
decorum?
tbrunet is offline  
post #48 of 195 Old 07-03-2006, 10:37 AM
AVS Special Member
 
tvted's Avatar
 
Join Date: Oct 2003
Location: Toronto, Canada
Posts: 3,573
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by tbrunet
decorum?
You've always seemed ready with web links. Why not a dictionary?

http://www.m-w.com/dictionary/decorum

Quote:
4 plural : the conventions of polite behavior
"decorum" is both singular and plural

It is also flattering that you place Mr. Wigggles' comment and mine in the same text body. However, in the nearly three years that I've stumbled through these forums, I've learned he is considerably more astute at these matters than me. Hence, I trust his judgement. In these matters he is an engineer, I am more of a dilettante art-fag type. I suspect you are neither.

Anyway, I tire quickly. Too many questions asked with too many vague, evasive replies if not outright non sequiturs. Definitely not dialogue as I understand it. Doesn't fit my "agenda".

going dark to match my clothes.
ted
tvted is offline  
post #49 of 195 Old 07-03-2006, 10:52 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 tbrunet
DARIN for the love of GOD!! The density for 2^8 bits (linear/non-linear) the CONTAST Maximum is:

2^8 = CR 255:1
Log (2^8) = Density ~ 2.4 = 255:1
You are like somebody who continually claims that 2+2=5 and refuses to take 2 or one object, 2 of another object and count how many objects there are. The only reason you have been able to delude yourself about this is because after telling me that you would measure the light coming off the CRT monitor you claim to use for your work in the industry, you then decided not to for some reason. One of your excuses included that you couldn't do it unless you had a light meter that measured accurately down to a light level that was the equivalent of contrast ratio of thousands to one. And yet you continue to claim that the darkest black cannot be any darker than 1/255th of the white level. I'm sure people can see how ridiculous that excuse is. Maybe you can remind people of all the reasons you said you would measure the light for the black level with a display and haven't in 6+ months.
Quote:
Originally Posted by tbrunet
“Gamma†has no affect on this figure. :eek:
Gamma has no effect on the highest CR possible because the correct black level is zero no matter the gamma. You don't understand enough to know that and so probably think that when experts say that the gamma doesn't change the range possible of the endpoints that they are supporting your ridiculous 255:1 number.
Quote:
Originally Posted by tbrunet
“Darin, I provided a link http://www.rags-int-inc.com/PhotoTechStuff/DigitalFilm/
& image in Figure 4 will put to rest the SM of the performance envelope of the HVS.
You only think that because you have so little comprehension of this subject matter. It is interesting how you link to things that contradict your own claims, but claim that they support your claims for some reason. In images like that one, some stuff on the left is likely to be invisible to a person who isn't able to block out the bright side largely because (as what you quoted says), images like that can easily contain 15 stops or light (or around 32,000:1 CR). And according to that, the upper limit for the human eye at a single glance is around 1000:1 (10 stops). Only somebody who didn't understand that stuff would then claim that the 1000:1 and 32,000:1 support claims of the human eye not being able to see more than 100-150:1 under any circumstances. Yes, there are images were shadow detail disappears because the differences don't get over the just-noticable-difference, but that doesn't support your claims and we've been over this and how a lighter background can actually make things invisible that wouldn't have been with a darker background.

And I assume that people here have watched actual movies. Sometimes it seems like you haven't because images like above are only a part of movies. When the sun goes down and things go dark then that single glance limit of 1000:1 moves to a darker range. But according to your claim above that you have never measured for, if the brightest thing in a scene is 1/10th of the white level then the contrast ratio in that scene cannot be any higher than 25.5:1 with 8 bit video. And if the brightest thing in a scene is 1/50th of the white level, then the contrast ratio in that scene cannot be any higher than about 5:1. If you are confused, those are both direct results of your claim that the on/off CR cannot be any higher than 255:1. They're all ridiculous and anybody who has ever measured displays for light levels would know that.

It also seem that you continue to confuse capture devices with display devices as relates to dynamic range or contrast ratio. I explained some of this in my article linked above.
Quote:
Originally Posted by tbrunet
"The eye can adjust from darkness to a bright scene in about 5 minutes. It takes up to 30 minutes to fully adjust from strong light to darkness."
I know we've been over this before. The range being talked about here is huge and real world experience should tell people that using this to try to claim that people won't be able to adjust during a movie doesn't hold much water. Anybody who has left a movie during a bright day and gone outside will know that it took their eyes some time to adjust. The reason is that their eyes are generally pretty dark adapted while in the movie theater. The same length of adjustment during the movie doesn't occur because the white levels in theaters are generally way darker. As in, the white levels in theaters are generally quite a bit darker than blacktop when they get outside and it is sunny, let alone the thousands of ft-lamberts common outside on sunny days with brighter surfaces. In the above "strong light" for our eyes doesn't refer to 7 ft-lamberts or so that is common for movies. Also, the 30 minutes is for full adjustment there. Adjusting from around 7 ft-lamberts to one seven hundred thousandths of that would take less than 10 seconds for people with normal vision according to testing I've done with a system that can do that and which I can show people (at a friend's house). I can show them how it takes even less time for one ten thousandths of the original level (10k:1 on/off CR). The fact that people can adjust down to seeing even much darker than 1/100000th of a ft-lambert after 30 minutes doesn't preclude smaller adjustments that are much quicker in any way.
Quote:
Originally Posted by tbrunet
If it were as Darin suggest .1%, then 8 bits would not be acceptable!
Nobody has ever said that 8 bits was enough to do all the gradations in real life correctly, but yet you continue to misuse this to claim that the highest CR possible is 255:1. Without dithering at edges people will be able to see many single steps in our 8 bit sources if the display doesn't soften things too much. That doesn't support your claims about on/off CR though no matter how many times you try to claim it does.
Quote:
Originally Posted by tbrunet
Lightness (a perceptual quantity) is approximately luminance raised to the 0.4 power.
Meaning that with your claim of only 255:1 being possible with 8 bits, if people put up a full "black" screen and a full white screen, the white screen would look about 9 times brighter than the "black" screen and that is the most range they could ever get to their eyes.

BTW: You are proving me right again by not being able to provide answers to:

100%stim (video 235) = 10 ft-lamberts
50%stim (video 125) = ?? ft-lamberts
25%stim (video 71) = ?? ft-lamberts
0%stim (video 16) = ?? ft-lamberts

Since this is long I'll include something else I think people who think there is any credence to tbrunet's claims of 255:1 on/off CR being the limit should read, in my next post.

--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 #50 of 195 Old 07-03-2006, 11:09 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
This stuff is long and I don't expect people to read it all, but for those who want to understand why some of us are so tired of tbrunet's claim about 255:1 being the limit and his trashing of some of the most knowledgeable people on the subject, I'll try to give a simple example. tbrunet's position is that the highest on/off CR possible is the number of steps, or the number of different states minus 1 (2^8-1=255). Meaning that the least amount of light possible for video black is 1/255th of the white level. Or 1/219th for video sources with video black at 16 and video reference white at 235.

I imagine that most people here own cars and that those cars have headlights which can be set to high, low, or off. tbrunet's "logic" would say that the highest CR possible is 2:1, because there are 3 settings. So, if high puts out 100 lumens, then off cannot put out any less than 50 lumens. Anybody who goes out and sets their headlights to off will know instantly that they can go a whole lot darker than half the level for the high setting and that in fact the darkest that the headlights can go for off is no visible light (unless there is a problem with the car). By tbrunet's claims, the only way the designers could get the headlights to go dark for the "off" position would be to give the user a ton of choices for how bright the headlights will be. More choices doesn't affect whether the lights can actually go off when the user selects off though.

Despite how obvious this headlight example should be to people after they think about it, based on his past behavior, I expect tbrunet to continue trying to mislead people into believing that having 255 steps means that the lowest light level cannot be any darker than 1/255th of the highest light level and to use his misdirection tricks in that process.

--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 #51 of 195 Old 07-03-2006, 12:10 PM
Senior Member
 
faterikcartman's Avatar
 
Join Date: Jan 2005
Location: San Diego
Posts: 379
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by darinp2
Anybody who has left a movie during a bright day and gone outside will know that it took their eyes some time to adjust. The reason is that their eyes are generally pretty dark adapted while in the movie theater. The same length of adjustment during the movie doesn't occur because the white levels in theaters are generally way darker. As in, the white levels in theaters are generally quite a bit darker than blacktop when they get outside and it is sunny, let alone the thousands of ft-lamberts common outside on sunny days with brighter surfaces.
--Darin
Just to clarify/explain a little more for the average Joe like me who may be reading all this; Even if there is a scene in a movie where there is a bright bright white it is usually relatively small (spatially) compared to the overall darkness around the viewer so that the overall light level reaching the eye is still not all that much relatively. Thus, no long burn-in which washes out the rest of the movie and takes a long time for recovery. Plus, bright scenes/areas of the picture are relatively short in duration usually and while adjustment can take a while in recovery, it also takes some time to leave an imprint.

Finally, I referred to the "crazy" guy for a reason.

I'm sure we've all met someone like this who the dictionary refers to as a "crank".

I've learned over the years that people who want to enter into long, complex, and detailed discussions like this but who refuse to participate in an honest dialogue (here, for example, ignoring the points of contradiction you've brought up) are either nutty agitators, folks who don't really know what they are talking about, or are educated/employed beyond their intelligence.

They become the proverbial tar-baby that you, Darin, just keep hitting as if it were a real person.

To lighten things up I'll give you a quote from Napolean Dynamite which you may want to take to heart here:

"Break the wrist, and just walk away."

The internet provides a lot of information and even if it is accurate it can cause a lot of problems when someone doesn't really understand it. It is like giving someone a list of math formulas and then having them try to solve a complex word problem involving integral calculus. Just because one holds the answer in their hand doesn't mean they can make use of it.

By definition 50% of people are below the median.
faterikcartman is offline  
post #52 of 195 Old 07-04-2006, 08:22 AM
AVS Special Member
 
danielo's Avatar
 
Join Date: Feb 2002
Location: Utrecht, Netherlands
Posts: 4,187
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 50
Quote:
Originally Posted by darinp2

Despite how obvious this headlight example should be to people after they think about it, based on his past behavior, I expect tbrunet to continue trying to mislead people into believing that having 255 steps means that the lowest light level cannot be any darker than 1/255th of the highest light level and to use his misdirection tricks in that process.

--Darin
You explained it so many times including how people can test it themselfs using a cheap lux meter that at this point its clear he doesn't accept how you (and others) define cr or he's just trolling its that simple.

Alot of people have made the 255 levels is 255 cr mistake until explained and explaining it to new people on the forum is very useful.

Maybe its better to help people with your massive amount of practical information who are not trying to troll you like this and tricking you into explaining it again and again and again and...

Just my opinion, One thing i like about hometheatherhifi is that they have a line of articles explaining things to a silly level to combat myths (THX one comes to mind) your article on these issues will be referenced by many people who can then make up their own mind.

Daniel.

for men to evolve we have to upgrade
danielo is offline  
post #53 of 195 Old 07-04-2006, 01:44 PM
Senior Member
 
faterikcartman's Avatar
 
Join Date: Jan 2005
Location: San Diego
Posts: 379
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Danielo, I agree and that is why I deleted my posts in this thread. That guy is not interested in a rational back and forth which actually considers points and facts he doesn't like.

Darin is hitting the proverbial tar baby in tangling with him.

The dictionary defines a "crank" as "an unbalanced person who is overzealous in the advocacy of a private cause." :(

By definition 50% of people are below the median.
faterikcartman is offline  
post #54 of 195 Old 07-04-2006, 04:11 PM
AVS Special Member
 
tvted's Avatar
 
Join Date: Oct 2003
Location: Toronto, Canada
Posts: 3,573
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by faterikcartman
Danielo, I agree and that is why I deleted.......
Agreed, and in the spirit of........well most. :)

ted
tvted is offline  
post #55 of 195 Old 07-04-2006, 04:52 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:
This message is hidden because tbrunet is on your ignore list.
Ah. :)
ChrisWiggles is offline  
post #56 of 195 Old 07-04-2006, 06:18 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
Do you have some medical research that shows this for the simultaneous contrast range (not the inverse of the simultaneous contrast range). --Darin
I might have mentioned in the past that I am on the BOD, and a significant investor in an Artificial Vision startup, so I have a financial vested interest in this area.

If you know of any medical research, or any straight forward test which could invalidate this "HVS conventional wisdom", I would be very interested in hearing about it.
HoustonHoyaFan is offline  
post #57 of 195 Old 07-04-2006, 06:30 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
I think you need to clarify what HVS 'conventional wisdom' is. AFAIK it is conventionally right along the lines of what Darin has been stating it is.

There are some much lower numbers that occur right AT the edge of a high contrast boundary, and there is also the 100:1 number which is really 1.01:1 Contrast Sensitivity Threshold, which is extremely different than a contrast ratio between black and white.

If you can drive at night, you can see a lot more than 100 or 200:1 or whatever, at once in your full field of view. If you look right AT a high contrast boundary (a bright light in darkness), immediately adjacent to that spill in the eye etc greatly reduces what you can see. That's why stoplights have those large black borders around them, and why tunnels are much more brightly lit at the ends of the tunnel than in the middle, during the day, etc.
ChrisWiggles is offline  
post #58 of 195 Old 07-04-2006, 06:31 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:
The problem occurs when people then try to use those numbers to say one only needs 200:1 on/off or even 200:1 ANSI CR, which is clearly incorrect.
It sounds like we're in strong agreement, hoyafan.
ChrisWiggles is offline  
post #59 of 195 Old 07-04-2006, 07:20 PM
Advanced Member
 
JJay's Avatar
 
Join Date: Aug 2002
Location: San Diego
Posts: 856
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I know darin has come at this a variety of ways :) but for the hopefully very few people that don't understand how tbrunet is wrong in his theories:

"The Weber-Fechner law
--------------------------------------------------------------------------------

This law hails from the middle of the nineteenth century. It is mainly founded on experiments where persons were given two nearly identical stimuli (for example, two similar weights) and tested whether they could notice a difference between them. It was found that the smallest noticeable difference was roughly proportional to the intensity of the stimulus. Ie, if a person could consistently feel that a 110 g weight was heavier than a 100 g weight, he could also feel that 1100 g was more than 1000 g.

There is a lot of sense in the Weber-Fechner law, but in the long run it has turned out to be too inflexible. It is therefore rarely used nowadays, but it is very vulgar not to know about it.

Nowadays, the formula most used is Stevens' formula. "

www.neuro.uu.se


As wigggles stated 'tbrunet is nuts and while he posts information that is often correct, it is often not applicable for the given situation'.

In our dealings with cr and pjs, we are concerned with the maximum threshold of cr perception by humans, not the minimum as tbrunet keeps throwing out there in a variety of different ways.

Jaime

If you can't say anything kind at least have the decency to be vague...
JJay is offline  
post #60 of 195 Old 07-04-2006, 07:53 PM
AVS Addicted Member
 
millerwill's Avatar
 
Join Date: Apr 2004
Location: Berkeley, CA
Posts: 11,383
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 21 Post(s)
Liked: 29
I have a more practical question re constrast, specifically ANSI CR. I've read much of what Darin has written on this issue, and learned a great deal (starting from 0!). My question: at night, suppose the only ambient light I have is that which is reflected from my light-colored ceiling and walls, and that I have a very bright projector (e.g., like the Mits wd-2000, which supposedly produces 1600 calibrated lumens). I understand that the only way to increase ANSI CR is to use a high gain (and thus narrow viewing angle) screen (such as the Highpower), so that some of the reflected light is rejected, or a gray screen (such as the Firehawk, which also has a narrow viewing angle) that reduces the reflected light. With the HP screen one could use an ND filter (because it is so high gain). Question: would the HP + ND filter be as good as the FH in producing good ANSI CR? HP + ND filter has the advantage that one can remove the ND filter as the lamp dims, while this option is not available with the FH screen. So what would be the recommendation: HP + ND filter, or FH?
millerwill is online now  
Closed Thread 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