Advanced: ANSI, Gamma, and on/off Contrast - OH MY! - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #1 of 94 Old 06-03-2008, 12:41 PM - Thread Starter
AVS Special Member
 
SOWK's Avatar
 
Join Date: Dec 2004
Location: Wauwatosa, WI
Posts: 4,073
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 213 Post(s)
Liked: 126
What is this thread for?

Questions and answers for Ansi Contrast, Display Gamma, on/off Contrast.

Why does it have advanced in the title?

For you to measure any of the above with any real accuracy you need calibration equipment. Colorimeters / Light meters

How do I feel about the matter?

My current thoughts on the 3 and how they work together for a 3 Dimensional Image.

Gamma > 2.3 = Number 1 Contributor
Ansi = Number 2 Contributor
ON/OFF = Number 3 Contributor

Why do I think this?

Most user opinions that have seen the Marantz VP-11S2 vs JVC RS2 at a Gamma of 2.2 say the Marantz is "more" 3D. (Ansi Advantage)

Greg R's Opinion on most material with the VP-11S2 at 2.2 gamma and RS2 at 2.4 gamma, RS2 wins in depth. (Gamma Advantage)

Now I have been able to create a gamma curve of ~2.4 on my Marantz VP-11S2 (crushing blacks by using brightness at -5) and the increase in 3D effect from the Brightness at 0/2.2 gamma is a major increase in depth.

I think if we/greg were to put the Marantz VP-11S2 in gamma 2.4 mode (external VP only) it would again be more 3D then the RS2, and the advantage at this point may be the Ansi.

Where does this leave on/off contrast and where does it fit in?

A very High On/Off of 30,000:1 + gives the ability to use a higher gamma through out the 0-100 IRE range like 2.3-2.4 with little to no shadow detail crush.

Where does this leave everyone else under 30,000:1?

If you have to have the ability within the hardware or external processor to use a lower gamma at low IRE ranges about 2.2, and then slowly raise up gamma to about 2.4 for mid to high level IRE levels this would allow you to get most of the advantages of higher gamma.
SOWK is online now  
Sponsored Links
Advertisement
 
post #2 of 94 Old 06-03-2008, 12:58 PM
AVS Addicted Member
 
R Harkness's Avatar
 
Join Date: May 2001
Location: Toronto, Ontario, Canada
Posts: 12,032
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 132 Post(s)
Liked: 359
Well I don't want to wade into the debate as to which is most important, but when I was doing every image experiment under the sun to adjust my plasma's image I was completely blown away by the effect of lowering the brightness control, in terms of the dimensionality of mid-tone scenes. I believe it was a gamma-like change. But it was just amazing how sculpturally dimensional actor's faces became when I'd plunge the brightness control down (and compensate somewhat by raising contrast - raising contrast by itself not achieving the effect of lowering the brightness control).

Unfortunately I couldn't live with all the crushed black detail so had to compromise. Which makes me giddy about the RS2 that I'll be able to use the gamma control to add dimensionality while not compromising black detail significantly.
R Harkness is online now  
post #3 of 94 Old 06-03-2008, 01:01 PM
AVS Special Member
 
HoustonHoyaFan's Avatar
 
Join Date: Jan 2003
Location: Houston, TX
Posts: 3,964
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 20
CR Numbers:
8000.00 30,000 30,000
2.20 2.2 2.4
100 8000.00 30000.00 30000.00
90 6344.88 23793.31 23297.18
80 4896.52 18361.97 17560.51
70 3650.11 13687.90 12745.49
60 2600.30 9751.11 8804.09
50 1741.10 6529.13 5683.94
40 1065.67 3996.26 3327.10
30 565.92 2122.21 1668.06
20 231.93 869.74 630.37
10 50.48 189.29 119.43
5 10.99 41.20 22.63
4 6.72 25.21 13.25
3 3.57 13.39 6.64
2 1.46 5.49 2.51
1 0.32 1.19 0.48
HoustonHoyaFan is offline  
post #4 of 94 Old 06-03-2008, 01:09 PM
AVS Special Member
 
Steve Dodds's Avatar
 
Join Date: May 1999
Location: San Francisco, now Australia
Posts: 2,884
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 16
I don't know how you can watch an image with the blacks deliberately crushed. It is inaccurate and goes against what we are trying to achieve. Black level detail is, to me, as important a part of the image as anything else.
Steve Dodds is offline  
post #5 of 94 Old 06-03-2008, 01:37 PM - Thread Starter
AVS Special Member
 
SOWK's Avatar
 
Join Date: Dec 2004
Location: Wauwatosa, WI
Posts: 4,073
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 213 Post(s)
Liked: 126
I didn't say I watch it like this. Just that I was able to recreate a gamma of 2.4
SOWK is online now  
post #6 of 94 Old 06-03-2008, 02:15 PM
AVS Club Gold
 
Mark Petersen's Avatar
 
Join Date: Dec 2000
Posts: 4,671
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by HoustonHoyaFan View Post

CR Numbers:
8000.00 30,000 30,000
2.20 2.2 2.4
100 8000.00 30000.00 30000.00
90 6344.88 23793.31 23297.18
80 4896.52 18361.97 17560.51
70 3650.11 13687.90 12745.49
60 2600.30 9751.11 8804.09
50 1741.10 6529.13 5683.94
40 1065.67 3996.26 3327.10
30 565.92 2122.21 1668.06
20 231.93 869.74 630.37
10 50.48 189.29 119.43
5 10.99 41.20 22.63
4 6.72 25.21 13.25
3 3.57 13.39 6.64
2 1.46 5.49 2.51
1 0.32 1.19 0.48

HHF, can you exaplain you how calculated these numbers? I assume that the contrast is derived from the ratio of the gamma coded white level with the black level held at a constant for each point along the curve. Is this correct? If so then it's worth mentioning that this only applies to a very low APL test pattern or image because as the overall APL increases the black level will increase too.
Mark Petersen is offline  
post #7 of 94 Old 06-03-2008, 02:39 PM
AVS Special Member
 
HoustonHoyaFan's Avatar
 
Join Date: Jan 2003
Location: Houston, TX
Posts: 3,964
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 20
Quote:
Originally Posted by Mark Petersen View Post

HHF, can you exaplain you how calculated these numbers? I assume that the contrast is derived from the ratio of the gamma coded white level with the black level held at a constant for each point along the curve. Is this correct? .

Mark your correct. The contrast calculations are decoded luminance ratioed against the black level calculated by the on/off. As such they are maxes, with no attempt to account for washout.
Quote:
Originally Posted by Mark Petersen View Post

If so then it's worth mentioning that this only applies to a very low APL test pattern or image because as the overall APL increases the black level will increase too.

Does the luminance calc change based on high APL or will the black level change due to washout?
HoustonHoyaFan is offline  
post #8 of 94 Old 06-03-2008, 03:42 PM
Advanced Member
 
Otto J's Avatar
 
Join Date: Feb 2003
Location: Denmark
Posts: 559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
I'm a bit puzzled about this whole "gamma X needs Y on/off contrast to avoid crushing blacks" thing... The Samsung SPA-800 supposedly has a gamma of 2.8. According to Joe, this is done to make the jump out of black small enough to be similar to CRT. And one thing I can say about this PJ, it is certainly NOT crushing blacks... On the contrary, it's among the best in the area of shadow detail I have ever seen, overshadowing (no pun intended) some much more expensive PJ's I have seen. According to the theory, this shouldn't be possible... So what am I missing here?

EDIT: BTW, I agree with the guess about these three issues, however I do believe that there are other factors that play a role at least as much as these issues, brightness being a big one, detail/sharpness being another. And, perhaps the biggest one of all: Homogeneity. I believe that a PJ that is pretty good in all areas, will have a better three-dimensionality or "realness", or whatever you like to call it, than a PJ that is fantastic in certain areas, but crappy in others. One single factor can break the image - if you have perfect gamma, super-high ansi, super-high on/off, perfect brightness, perfect color reproduction, but heavy edge enhancement, you will not get a three-dimensional picture. So, if we want to quantify the performance of a projector, you need to be very sure you take _all_ relevant factors into account. Leave a single one out, and you have a big risk of making the wrong conclusions.

Mvh. Otto, AV Precision
Installer/Calibrator
Otto J is offline  
post #9 of 94 Old 06-03-2008, 03:49 PM - Thread Starter
AVS Special Member
 
SOWK's Avatar
 
Join Date: Dec 2004
Location: Wauwatosa, WI
Posts: 4,073
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 213 Post(s)
Liked: 126
Quote:
Originally Posted by Otto J View Post

I'm a bit puzzled about this whole "gamma X needs Y on/off contrast to avoid crushing blacks" thing... The Samsung SPA-800 supposedly has a gamma of 2.8. According to Joe, this is done to make the jump out of black small enough to be similar to CRT. And one thing I can say about this PJ, it is certainly NOT crushing blacks... On the contrary, it's among the best in the area of shadow detail I have ever seen, overshadowing (no pun intended) some much more expensive PJ's I have seen. According to the theory, this shouldn't be possible... So what am I missing here?

Not to sound rude but are you an ISF calibrator?

What does your sig, mean:

Mvh. Otto, AV Precision
ISF Calibrator



If so what are you asking this question for, and do you not have the test equipment to get the real readings from the Samsung DLP for gamma in the lower IRE ranges?

If you were getting a real display gamma of 2.8 from the samsung, you would be crushing shadow detail significantly. You need to have about 30,000:1 and more just to use 2.4 with little to no shadow detail loss.

I am truly not trying to be insulting at all here, just asking.
SOWK is online now  
post #10 of 94 Old 06-03-2008, 04:46 PM
AVS Club Gold
 
Mark Petersen's Avatar
 
Join Date: Dec 2000
Posts: 4,671
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by HoustonHoyaFan View Post

Mark your correct. The contrast calculations are decoded luminance ratioed against the black level calculated by the on/off. As such they are maxes, with no attempt to account for washout.
Does the luminance calc change based on high APL or will the black level change due to washout?

For full 100% white, there shouldn't be an appreciable change based on overall APL (but there can be small changes, for example a CRT might go down in brightness with a full white screen because of the high current draw while LCOS actually measures slightly brighter (maybe 5%) with windowed patterns). Ignoring these relatively minor effects though, full 100% white luminance is mostly constant and if the gamma is known the luminance along the curve is known (exactly as how you've done it). It's the black level that goes though a drastic change from high APL to low APL.

As an example if we compare your numbers (which assumes a small white test pattern) vs an ANSI test pattern the white measurements will be close to the same in both cases but the black level in the ANSI case will be much higher which is why the contrast @ 100% can vary from say 300:1 (ANSI) to 30,000:1 (your low APL case). This can only happen when the full white pattern is very small (ie the image is dark and low APL), so that the washout doesn't effect the black level measurement.

In the case of the in between luminance steps, I assume (but haven't measured) that the washout effect can effect the very low luminance steps. This is dependent of course on how bright the overall image is and also the closeness to bright content. This is also true for projectors with poor native contrast performance, for example back in the days of digitals with 800:1 contrast there wasn't much contrast below about 10% because the black level eclipsed the whites even in the low APL case.

One last note related to this discussion and thread, people talk about how easy it is to spot washout on an image based on ANSI performance alone. It would be interesting to construct a synthetic test pattern that measures the degree of washout. For example say we take the extreme case of a 100% full white screen with a small rectangle in the center and then we sweep the small rectangle from 100% to 0% it would be interesting to see where on the curve the whites are washed out. It would be interesting to compare this with high ANSI projectors and also mid ANSI projectors. The results wouldn't be applicable to any particular scene but it could prove or disprove the degree of washout that people claim because it would be the case of maximum washout.
Mark Petersen is offline  
post #11 of 94 Old 06-03-2008, 05:22 PM
AVS Special Member
 
mdputnam's Avatar
 
Join Date: Sep 2003
Location: Kalifornia
Posts: 1,346
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 14
What would be more informative in determining picture quality is a specification that measures contrast at all feature sizes from full screen down to alternating pixels. The results would define a curve, the area under which could be a figure of merit for the information content in the image. The curve itself would give us far more information on image quality that just two or three measurements. From one end of the spectrum would be the largest feature ie on/off contrast at the other extreeme would be data similar to MTF indicating small feature sharpness. MTF is a little understood measurement by most hobbist here is a good explanation of it. Some of the references at the end are also excellent.

Human perception is not a direct consequence of reality, but rather an act of imagination. - Michael Faraday
mdputnam is offline  
post #12 of 94 Old 06-03-2008, 05:28 PM - Thread Starter
AVS Special Member
 
SOWK's Avatar
 
Join Date: Dec 2004
Location: Wauwatosa, WI
Posts: 4,073
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 213 Post(s)
Liked: 126
Well until someone make a program that can display contrast driven images from frull screen down to alternating pixels, we will never know.
SOWK is online now  
post #13 of 94 Old 06-03-2008, 05:34 PM
AVS Club Gold
 
Art Sonneborn's Avatar
 
Join Date: Jun 2000
Location: Battle Creek,MI USA
Posts: 22,307
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 49
Quote:
Originally Posted by mdputnam View Post

What would be more informative in determining picture quality is a specification that measures contrast at all feature sizes from full screen down to alternating pixels. The results would define a curve, the area under which could be a figure of merit for the information content in the image. The curve itself would give us far more information on image quality that just two or three measurements. From one end of the spectrum would be the largest feature ie on/off contrast at the other extreeme would be data similar to MTF indicating small feature sharpness. MTF is a little understood measurement by most hobbist here is a good explanation of it. Some of the references at the end are also excellent.

But isn't MTF really most applicable at one end of the curve as was stated from FF on to FF off to a single on off pixel grid ? MTF being almost entrely at the pixel grid end. I mean sequential is at one end ANSI in the middle and MTF at or near the opposite end.



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 #14 of 94 Old 06-03-2008, 07:55 PM
AVS Special Member
 
mdputnam's Avatar
 
Join Date: Sep 2003
Location: Kalifornia
Posts: 1,346
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 14
Quote:
Originally Posted by Art Sonneborn View Post

But isn't MTF really most applicable at one end of the curve as was stated from FF on to FF off to a single on off pixel grid ? MTF being almost entrely at the pixel grid end. I mean sequential is at one end ANSI in the middle and MTF at or near the opposite end.Art

Yes, that's why a curve representing a continuum of measurements from full screen on/off down to single pixel would be most informative. I included the MTF stuff because most folks (including me) have a difficult time getting a warm and fuzzy feeling understanding it and the far end of this graph would allow you to reach similar conclusions about a projectors image quality as a MTF number does.

Human perception is not a direct consequence of reality, but rather an act of imagination. - Michael Faraday
mdputnam is offline  
post #15 of 94 Old 06-03-2008, 07:58 PM
AVS Special Member
 
mdputnam's Avatar
 
Join Date: Sep 2003
Location: Kalifornia
Posts: 1,346
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 14
Quote:
Originally Posted by SOWK View Post

Well until someone make a program that can display contrast driven images from frull screen down to alternating pixels, we will never know.

Given all the smart people in this business I think that collectively they could come up with the methodology, if the industry had the will to do it.

Human perception is not a direct consequence of reality, but rather an act of imagination. - Michael Faraday
mdputnam is offline  
post #16 of 94 Old 06-03-2008, 09:20 PM
AVS Special Member
 
gremmy's Avatar
 
Join Date: Sep 2003
Location: St. Louis, MO
Posts: 3,942
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
Quote:
Originally Posted by Otto J View Post

I'm a bit puzzled about this whole "gamma X needs Y on/off contrast to avoid crushing blacks" thing... .

I think another side of this that rarely gets discussed is that "gamma X needs Y ft. Lamberts" to avoid dimming certain scenes to unpleasant levels. Even if the projector has enough on/off or ANSI to support a high gamma without crushing blacks, the high gamma may not be feasible once the bulb has dimmed.

For example, my Pearl is too dark in many scenes on Gamma 3 once the pop has fizzled out of the bulb.
gremmy is offline  
post #17 of 94 Old 06-03-2008, 09:32 PM
AVS Addicted Member
 
darinp2's Avatar
 
Join Date: Sep 2003
Location: Seattle, WA
Posts: 21,206
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 65
Quote:
Originally Posted by SOWK View Post

My current thoughts on the 3 and how they work together for a 3 Dimensional Image.

Gamma = Number 1 Contributor
Ansi = Number 2 Contributor
ON/OFF = Number 3 Contributor

Why do I think this?

Most user opinions that have seen the Marantz VP-11S2 vs JVC RS2 at a Gamma of 2.2 say the Marantz is "more" 3D. (Ansi Advantage)

Greg R's Opinion on most material with the VP-11S2 at 2.2 gamma and RS2 at 2.4 gamma, RS2 wins in depth. (Gamma Advantage)

Now I have been able to create a gamma curve of ~2.4 on my Marantz VP-11S2 (crushing blacks by using brightness at -5) and the increase in 3D effect from the Brightness at 0/2.2 gamma is a major increase in depth.

I think if we/greg were to put the Marantz VP-11S2 in gamma 2.4 mode (external VP only) it would again be more 3D then the RS2, and the advantage at this point may be the Ansi.

I doubt it would. Setting both to 2.2 gamma is taking away much of the advantage the RS2 can provide, while I think going to 2.4 gamma with a lower on/off CR projector is more likely to show the weakness that lower on/off CR brings. The Sharp 20k has multiple gamma choices and can go higher than 2.2 with internal controls. As far as what Greg said in his review, he didn't seem very confident that the 11S2 could support a 2.4 gamma without too much negative side effect from the lower on/off CR. And 2.2 gamma down low and then 2.4 gamma up high isn't the same thing as 2.4 throughout.
Quote:
Originally Posted by SOWK View Post

A very High On/Off of 30,000:1 + gives the ability to use a higher gamma through out the 0-100 IRE range like 2.3-2.4 with little to no shadow detail crush.

If the higher on/off CR allows the user to use a higher gamma, then how is gamma more important than on/off CR (as it seems you meant above)? That is, if one is a gating factor to the other, then that alone makes it important if the other factor is important. High on/off CR doesn't require higher gamma to see the advantage of it. Of course, there are multiple factors to images, as always. Higher gamma does require higher on/off CR to not bring out some negative side effects. And unfortunately, we can't just buy external processors to change our on/off CRs in the way we can buy external processors to change our gammas.
Quote:
Originally Posted by SOWK View Post

If you have to have the ability within the hardware or external processor to use a lower gamma at low IRE ranges about 2.2, and then slowly raise up gamma to about 2.4 for mid to high level IRE levels this would allow you to get most of the advantages of higher gamma.

I don't think this is what your experiment really showed. I think what you seem to want is higher gammas for bright images and lower gammas for dim images (come out of black faster), at least overall. That is different than wanting the low IRE levels to have a lower gamma and the higher IRE levels to have higher gamma. As an example, in your experiment with crushing blacks you lowered the level that 5 IRE produced and it looks like you then saw an advantage from doing that in bright mixed images. But if you made the lower IREs 2.2 then you wouldn't have lowered the level of 5 IRE and so wouldn't have seen that effect for that portion of the images, even in bright mixed scenes. The problem with low on/off CR then comes in for darker images where you now want a bigger separation between 0 IRE and 5 IRE. A lower gamma gives a bigger separation there so I think you want a lower gamma down lower during dark scenes and higher gamma down low during bright scenes, from your description.

Basically, I think what you really want is a dynamic gamma. That is something that can help overcome lower on/off CR, since much of what higher on/off CR brings is higher simultaneous CR in darker images, but if all the levels can be raised dynamically without crushing up against 100 IRE, then you can get higher on/off CR in those darker images with a dynamic gamma. It likely isn't completely free from an imaging standpoint (much like dynamic on/off CR from dynamic irises isn't completely free compared to native CR that high) and would require a processor with good processing power and algorithms to do right. This is way beyond just having a gamma you can adjust to different amounts at different points along the curve, but that is something the RS2 supports internally (you can pick points along the curve for each primary and move them up and down).

Given your hypothesis about what is important and Greg Roger's measurements where the Sharp 20k measured higher ANSI CR but lower on/off CR than the Marantz 11S2, and that the Sharp 20k has gamma choices internally, which do you think would have more image depth between the 20k and the 11S2 (at least if somebody wasn't having problems with rainbows)?

If you use the iris open choice for the 11S2, how does the depth of image compare to the iris closed choice?

--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 online now  
post #18 of 94 Old 06-04-2008, 04:30 AM
AVS Special Member
 
Lawguy's Avatar
 
Join Date: Sep 2004
Posts: 5,709
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 11
Quote:
Originally Posted by Mark Petersen View Post

One last note related to this discussion and thread, people talk about how easy it is to spot washout on an image based on ANSI performance alone. It would be interesting to construct a synthetic test pattern that measures the degree of washout. For example say we take the extreme case of a 100% full white screen with a small rectangle in the center and then we sweep the small rectangle from 100% to 0% it would be interesting to see where on the curve the whites are washed out. It would be interesting to compare this with high ANSI projectors and also mid ANSI projectors. The results wouldn't be applicable to any particular scene but it could prove or disprove the degree of washout that people claim because it would be the case of maximum washout.

Someone, I forget who, suggested an easy way to demonstrate this washout effect by using a PC. First, set your desktop to black. Then, create a new Word document (which is an empty white screen). Remove all toolbars. Then, just resize that window dynamically taking it from small to large and see how the black background changes. There is a point on the RS1 where blacks are clearly affected. This is, I recall, when the white portion of the image occupies between 1/3 and 1/2 of the screen.

Affable Nitwit
Lawguy is offline  
post #19 of 94 Old 06-04-2008, 04:42 AM
Advanced Member
 
Otto J's Avatar
 
Join Date: Feb 2003
Location: Denmark
Posts: 559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by SOWK View Post


If you were getting a real display gamma of 2.8 from the samsung, you would be crushing shadow detail significantly. You need to have about 30,000:1 and more just to use 2.4 with little to no shadow detail loss.

I am truly not trying to be insulting at all here, just asking.

No offence taken, I understand why you ask this. I simply haven't measured specifically at the low IRE's yet, and I was just lazy and wondered if anyone could clarify, obviously I should just measure the damn thing and find out for myself. I just received it, so I haven't had time to test everything yet. However, just looking at low IRE test patterns, the rise from black definately does look like the gamma is quite high, and no, it is not crushing black. I'm probably just overlooking some obvious point here, I could try to measure it and share the measurements, to get your opinion on what's going on. I agree that within all logic, the gamma slope would have to be non-linear (or whatever you would call it), to make it work, I just don't see it, that's why I'm asking... I'll make some measurements, might be a couple of days before I get around to it though. However, it is quite visually obvious that the SPA-800: 1. Does not crush black level detail, on the contrary it is better than most anything else I have seen. 2. Comes quite slowly out of black, and 3. Has a fairly high overall gamma. Either I'm missing something, or the math doesn't add up.

I shall return with some measurements.

Mvh. Otto, AV Precision
Installer/Calibrator
Otto J is offline  
post #20 of 94 Old 06-04-2008, 04:56 AM - Thread Starter
AVS Special Member
 
SOWK's Avatar
 
Join Date: Dec 2004
Location: Wauwatosa, WI
Posts: 4,073
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 213 Post(s)
Liked: 126
Quote:
Originally Posted by darinp2 View Post


Given your hypothesis about what is important and Greg Roger's measurements where the Sharp 20k measured higher ANSI CR but lower on/off CR than the Marantz 11S2, and that the Sharp 20k has gamma choices internally, which do you think would have more image depth between the 20k and the 11S2 (at least if somebody wasn't having problems with rainbows)?

If you use the iris open choice for the 11S2, how does the depth of image compare to the iris closed choice?

--Darin


I think the Shap 20K would have more image depth.

On my VP-11S2 I use Iris 1 (most closed) I can open it for you to see the effect it has on the depth of the image for you. Should I try this in brightness -5 mode, or calibrated brightness at 0?


You could be right about the Dynamic Gamma thing too. I need someone to make a $1000.00 VP that can look at the incoming images and apply gamma baised upon overall brightness of the scene. If most of the scene is in low Ire range, it lowers gamma throughout the range, and for mixed and bright scenes raises gamma up to 2.5 throughout the range. Darin start working on that asap!
SOWK is online now  
post #21 of 94 Old 06-04-2008, 04:59 AM - Thread Starter
AVS Special Member
 
SOWK's Avatar
 
Join Date: Dec 2004
Location: Wauwatosa, WI
Posts: 4,073
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 213 Post(s)
Liked: 126
Quote:
Originally Posted by Otto J View Post

No offence taken, I understand why you ask this. I simply haven't measured specifically at the low IRE's yet, and I was just lazy and wondered if anyone could clarify, obviously I should just measure the damn thing and find out for myself. I just received it, so I haven't had time to test everything yet. However, just looking at low IRE test patterns, the rise from black definately does look like the gamma is quite high, and no, it is not crushing black. I'm probably just overlooking some obvious point here, I could try to measure it and share the measurements, to get your opinion on what's going on. I agree that within all logic, the gamma slope would have to be non-linear (or whatever you would call it), to make it work, I just don't see it, that's why I'm asking... I'll make some measurements, might be a couple of days before I get around to it though. However, it is quite visually obvious that the SPA-800: 1. Does not crush black level detail, on the contrary it is better than most anything else I have seen. 2. Comes quite slowly out of black, and 3. Has a fairly high overall gamma. Either I'm missing something, or the math doesn't add up.

I shall return with some measurements.

I guess this would be very interesting if you are getting readings of 2.4 or higher throughout the range with no Image degradation. If so I will need to find a cheap VP that can change my VP-11S2 to gamma 2.4 mode.
SOWK is online now  
post #22 of 94 Old 06-04-2008, 05:11 AM - Thread Starter
AVS Special Member
 
SOWK's Avatar
 
Join Date: Dec 2004
Location: Wauwatosa, WI
Posts: 4,073
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 213 Post(s)
Liked: 126
Quote:
Originally Posted by darinp2 View Post

I doubt it would. Setting both to 2.2 gamma is taking away much of the advantage the RS2 can provide, while I think going to 2.4 gamma with a lower on/off CR projector is more likely to show the weakness that lower on/off CR brings. The Sharp 20k has multiple gamma choices and can go higher than 2.2 with internal controls. As far as what Greg said in his review, he didn't seem very confident that the 11S2 could support a 2.4 gamma without too much negative side effect from the lower on/off CR. And 2.2 gamma down low and then 2.4 gamma up high isn't the same thing as 2.4 throughout.



Someone with a Marantz VP-15S1 please try the projector in Gamma 2.4 mode, What are the problems you are seeming with the image if any?

I really want to know what the draw backs would be? Just a minor loss of shadow detail? Like taking brightness and going to -1 from calibrated?

If so that is still worth it.

Or is there much more to it?
SOWK is online now  
post #23 of 94 Old 06-04-2008, 05:22 AM - Thread Starter
AVS Special Member
 
SOWK's Avatar
 
Join Date: Dec 2004
Location: Wauwatosa, WI
Posts: 4,073
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 213 Post(s)
Liked: 126
What is very weird to me is from gamma 1.8 - 2.2, there is little jump in image depth, only when you hit an average of 2.3 in my testing do the image really start to gain depth.

Anyone with a knowledgeable answer for this?
SOWK is online now  
post #24 of 94 Old 06-04-2008, 06:53 AM
Advanced Member
 
Otto J's Avatar
 
Join Date: Feb 2003
Location: Denmark
Posts: 559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by SOWK View Post

I guess this would be very interesting if you are getting readings of 2.4 or higher throughout the range with no Image degradation. If so I will need to find a cheap VP that can change my VP-11S2 to gamma 2.4 mode.

I just did a measurement, the overall gamma is pretty consistent throughout the range, about 2.55 (10-100 IRE). I'm a bit uncertain as to what you mean when you say "crushing" blacks, there is definately no clipping of details, however I'm not really sure my EyeOne Pro delivers measurements that are accurate enough near black that they will be of much use. I cannot rule out the possibility that the gamma is lower at very near black, however, it does look darker than most other PJ's, but not crushed/clipped. I will say that I do not get the same near-black performance from other PJ's adjusted to 2.55 (or close to it), so I do believe that just because the overall gamma is 2.55, the near-black performance may still vary wildly. I'm not sure just putting in a Lumagen and setting it up for 2.55 gamma, will give the same picture.

Mvh. Otto, AV Precision
Installer/Calibrator
Otto J is offline  
post #25 of 94 Old 06-04-2008, 06:57 AM
Advanced Member
 
Otto J's Avatar
 
Join Date: Feb 2003
Location: Denmark
Posts: 559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by SOWK View Post

What is very weird to me is from gamma 1.8 - 2.2, there is little jump in image depth, only when you hit an average of 2.3 in my testing do the image really start to gain depth.

Anyone with a knowledgeable answer for this?

Probably not what you will call a knowledgeable answer, however, in general my impression is that the closer you get to "accurate", the more difference a small change will make. So, if a gamma lower than 2.2 looks "wrong", subjectively it doesn't really matter much how "wrong" it is, it looks wrong whatever you do. The change from "a little bit off" to "just right", makes a lot of difference, even if it may look like a small change on paper.

Mvh. Otto, AV Precision
Installer/Calibrator
Otto J is offline  
post #26 of 94 Old 06-04-2008, 07:06 AM
Advanced Member
 
Otto J's Avatar
 
Join Date: Feb 2003
Location: Denmark
Posts: 559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
I think the issue that is putting us off track here, is that we usually measure gamma by measuring 0-10-20etc IRE. I believe it is quite possible to have a gamma of, say, 2.5 from 10-100 IRE, while still having a much lower gamma from 0-10 IRE. More importantly, I'm not quite sure that increasing gamma neccessarily crushes black, I believe it would be possible to increase the gamma from 10-100 IRE, but still maintain near-black detail by still having a lower gamma near black.

Mvh. Otto, AV Precision
Installer/Calibrator
Otto J is offline  
post #27 of 94 Old 06-04-2008, 07:23 AM - Thread Starter
AVS Special Member
 
SOWK's Avatar
 
Join Date: Dec 2004
Location: Wauwatosa, WI
Posts: 4,073
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 213 Post(s)
Liked: 126
Quote:
Originally Posted by Otto J View Post

I think the issue that is putting us off track here, is that we usually measure gamma by measuring 0-10-20etc IRE. I believe it is quite possible to have a gamma of, say, 2.5 from 10-100 IRE, while still having a much lower gamma from 0-10 IRE. More importantly, I'm not quite sure that increasing gamma neccessarily crushes black, I believe it would be possible to increase the gamma from 10-100 IRE, but still maintain near-black detail by still having a lower gamma near black.

I mentioned something similar to that in the first post

Where does this leave everyone else under 30,000:1?

If you have to have the ability within the hardware or external processor to use a lower gamma at low IRE ranges about 2.2, and then slowly raise up gamma to about 2.4 for mid to high level IRE levels this would allow you to get most of the advantages of higher gamma.


And I mentioned the exact same thing 2 days ago in the VP-11S2 thread.

Post # 1439 http://www.avsforum.com/avs-vb/showt...4#post13996764

Great minds think alike!

I was thinking that may be correct, but now I wonder if only Dynamic Gamma like Darin mentioned would really work to the same effect.
SOWK is online now  
post #28 of 94 Old 06-04-2008, 07:32 AM - Thread Starter
AVS Special Member
 
SOWK's Avatar
 
Join Date: Dec 2004
Location: Wauwatosa, WI
Posts: 4,073
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 213 Post(s)
Liked: 126
Quote:
Originally Posted by Otto J View Post

Probably not what you will call a knowledgeable answer, however, in general my impression is that the closer you get to "accurate", the more difference a small change will make. So, if a gamma lower than 2.2 looks "wrong", subjectively it doesn't really matter much how "wrong" it is, it looks wrong whatever you do. The change from "a little bit off" to "just right", makes a lot of difference, even if it may look like a small change on paper.

I guess I'm not saying the gamma of 1.8 looks wrong,

I did seperate calibrations for my VP-11S2 on for gamma C (1.8) & gamma Standard (2.2) Getting them both as acurate as possible to D65 and found little difference in depth between the 2.

But when I take 2.2 settings and bring brightness down -3 (gamma 2.3) / -5 (gamma ~2.4) the image depth increases a significant amount.

I am wondering if a real display gamma of 2.4 throughout would produce the same results though.
SOWK is online now  
post #29 of 94 Old 06-04-2008, 09:24 AM
AVS Special Member
 
CaspianM's Avatar
 
Join Date: Sep 2000
Location: Land of Cardinals
Posts: 5,841
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
I believe gamma only complements the ansi and on/off. If hi ansi and sequential is not there to begin with gamma alone will not do it.

It is all about quality...that is the picture

JVC & NEC 8" CRT with 106" wide Stewart screen. All NHT speakers driven by Pioneer Elite AVR and bluray

Custom dedicated 8 seat theater

CaspianM is offline  
post #30 of 94 Old 06-04-2008, 10:08 AM - Thread Starter
AVS Special Member
 
SOWK's Avatar
 
Join Date: Dec 2004
Location: Wauwatosa, WI
Posts: 4,073
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 213 Post(s)
Liked: 126
Correct, but I think Gamma and On/off have more to do with each other then Ansi.

You need very high on/off to get higher gamma with no image faults (as far as we know so far)

And Ansi I believe really only matters if you compare 2 projectors in the same gamma mode.
SOWK is online now  
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