Time to define an alternative to ANSI contrast? - Page 12 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #331 of 341 Old 02-16-2007, 09:38 PM
AVS Addicted Member
 
darinp2's Avatar
 
Join Date: Sep 2003
Location: Seattle, WA
Posts: 21,171
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 5 Post(s)
Liked: 54
Quote:
Originally Posted by GlenC View Post

I still don't see/understand the need for "special" test patterns for a DI projector. The patterns that Wm has developed seem very appropriate for measuring the maximum CR in low APL. If a DI projector doesn't improve their CR with these patterns, then I don't see much advantage to the DI.

That's the main problem. Those test patterns don't show the advantage of a DI and that advantage is real world. They don't show how much a DI can help the simultaneous CR in a dark image with just dark things (like nothing above 30%stim). If you wanted to come up with patterns to avoid seeing the advantage that DIs bring, then test patterns like these ones that just make the bright parts smaller would probably be what you would use, so part of the question is which patterns are the "special" ones in that case. I like the patterns (I think they are useful for showing one aspect well) but they are kind of special in favoring the non-DI projector.
Quote:
Originally Posted by GlenC View Post

I see DI as a gimmick to help a poor performing projector (poor black levels), necessatating reliance on the manufacturer to correctly modify the gamma curves to compensate for the reduced light output.

Do you see the DIs that we either evolved with or were created with (depending on your persuasion) as gimmicks to help poor performing areas of human eyes and vision, or as useful aspects of our vision that create a good balance? There isn't a non-DI digital projector I know of that couldn't improve their performance in certain scenes with a DI and I doubt there will be any that qualify for quite a while.

--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  
Sponsored Links
Advertisement
 
post #332 of 341 Old 02-16-2007, 09:45 PM - Thread Starter
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 darinp2 View Post

I would expect the simultaneous CRs to be reasonably close to what a non-DI projector with the same full range on/off CR and the same ANSI CR could do from a 100/0 checkerboard all the way down to a 1/0 checkerboard.

Yes this is exactly the problem. It won't capture the benefits of DI behavior much better than the APL pattern that we have now.

What is more interesting and which gauges DI performance benefits better is to see the full range of %stim changes with a low APL where the iris is closed similar to the graph in the analysis that I posted earlier. If we have a low APL (small rectangles) test pattern and start at 100% stim and work down we know from the static contrast results that the iris will be at it's smallest aperture even at 100% stim so we can easily see the effects of gamma boost as the %stim is reduced. Basically it will exactly mimic the dynamic contrast analysis that I did earlier with the exception that the %stim whites and blacks will be measured rather than extrapolating the % stim white and using the measured min black.

This case (low APL) doesn't capture the full realm of DI behavior but it does capture the case of maximum dynamic contrast (lowest aperture and largest gamma boost) which is probably the most important case.

EDIT: Darin, now that I've thought about your post more, I realize that I was thinking that you meant native projector w/o DI but with iris off rather than the one with the high on/off. So I see more of the logic behind what you're saying. I agree that the dynamic contrast for the DI in this scenario (with the ANSI pattern) will be similar to the native with high on/off at the very bottom of the %stim range, but it's still less than ideal because much of the higher % stim ranges aren't measuring dynamic contrast (ie the iris is fully open and there is no gamma boost). Instead much of the higher %stim ranges would actually be measuring static contrast.

In fact from the scenario that I posted the maximum benefit of the DI happens at ~60% stim with the iris fully closed (low APL). Just venturing a very rough guess, I wouldn't be surprised if the iris doesn't fully close until the %stim with the ANSI style pattern is down around 20% or so. So all we have done is captured dynamic contrast for a relatively small range of % stim and for a mix of iris apertures.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
Mark Petersen is offline  
post #333 of 341 Old 02-17-2007, 02:30 AM - Thread Starter
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 GlenC View Post

I still don't see/understand the need for "special" test patterns for a DI projector. The patterns that Wm has developed seem very appropriate for measuring the maximum CR in low APL. If a DI projector doesn't improve their CR with these patterns, then I don't see much advantage to the DI.

I see DI as a gimmick to help a poor performing projector (poor black levels), necessatating reliance on the manufacturer to correctly modify the gamma curves to compensate for the reduced light output.

We will not be seeing moveis released that are "optimized for DI projectors"

I don't really see DIs as a gimmick. I think the scenario that I posted shows the benefit of a DI quite well, especially if you compare the curves of the native iris off projector which equates to what the projector would be without DI. In this example, the projector is only ~3100:1 on/off and for dark details in low APL scenes it performs as well as a projector with 14000:1 on/off. So there is clearly an advantage. The price of such a system however is that the white levels are doubly modulated (both iris aperture and gamma is being modified at the same time) so the white levels might not be as stable. The gamma relationship between different % stim pixels also gets skewed with BC being an extreme case. The other drawback of course is that bright details in low APL scenes don't have the full contrast as a high native on/off projector.

So I think to properly determine simultaneous contrast for a mix of projector technologies it's important to capture the pros and cons of DI behavior. I think it's an important aspect of this project and it's been a healthy if at times contentious discussion. What I've personally found interesting in this discussion is that DI opponents have tended to dismiss the advantages of DI's and the DI proponents have tended to overcredit the advantages of DI. In fact I've heard several times in this and other threads that a DI offers the same performance in low APL scenes as a high native on/off and this is clearly not the case for all pixel IREs. What this tells me is that people are still grappling with how DI affects simultaneous contrast and how to communicate these effects. I think I've personally taken a lot of heat for not communicating a balanced view of DI performance but up until I posted the dynamic contrast scenario I hadn't really seen anyone else do a better treatment of it.

So hopefully as this project progresses we'll get better at describing and quantifying contrast performance and I think as this happens there will be less frustration and contention. It also goes to show how poor the traditional merits of ANSI and on/off CR are at describing these technologies which is why this thread was started.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
Mark Petersen is offline  
post #334 of 341 Old 02-17-2007, 10:44 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
Mark, you are right, I was probably a little unfair on my opinion. It was mostly targeted at those who tend to "over-credit" the advantages. A discussion like this needs to be realistic and factual.

I haven't lived with a DI projector, the concept and function just don't attract me. The one thing I haven't seen is a mapping of the Iris function. Just when does the iris start to close and how much. Seems like a test with a white window or white field, reducing the level by 1 IRE increments would allow observation and measurement of the iris and gamma modifications. At the point the iris is fully closed, the luminance value should be the threshold for CR measurement. If the luminance value was, say, 1.5%, then, with a full black field, and a 100% white window, area totaling 1.5%, should be the CR measurement point for "MAX DI low APL CR". Does this make sense?

Glen Carter
Home Theater Calibration

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
GlenC is offline  
post #335 of 341 Old 02-17-2007, 02:27 PM - Thread Starter
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 GlenC View Post

I haven't lived with a DI projector, the concept and function just don't attract me.

I also haven't lived with a DI projector so it's been a bit of a struggle for me to understand the pros and cons. I think most people would prefer a higher native on/off CR projector than one that achieves the same on/off CR with a DI, but this isn't really a fair comparison. A better comparison is two projectors with similar native on/off CR but one with a DI that can increase the on/off CR by many times. Many people will prefer the added contrast benefits of the DI even with it's cons than the lower contrast of the native projector.

Quote:
Originally Posted by GlenC View Post

The one thong I haven't seen is a mapping of the Iris function. Just when does the iris start to close and how much. Seems like a test with a white window or white field, reducing the level by 1 IRE increments would allow observation and measurement of the iris and gamma modifications. At the point the iris is fully closed, the luminance value should be the threshold for CR measurement. If the luminance value was, say, 1.5%, then, with a full black field, and a 100% white window, area totaling 1.5%, should be the CR measurement point for "MAX DI low APL CR". Does this make sense?

I haven't seen a mapping of the iris function either and I agree it would be interesting to plot. The problem is that it's non-trivial to do so. For example if one were to trigger the iris using a %stim pattern (non 100% white) how does one ensure that the iris isn't closing while the gamma is being boosted. In other words it's hard to tell when the in-between iris apertures are being used. It's also complicated because of the varied nature of the DI algorithm. Using a full white pattern may cause the DI to reduce gamma boost to mitigate BC. From the static contrast results though we are able to see where the iris clamps down for a full white pattern (it's between the 2% and 5% APL patterns).

If a person isn't able to capture all of the in between iris apertures and gamma boost combinations, then the next best thing to do is capture the best case and worse case scenarios which delineate the in between regions and this is what the iris off and low APL test pattern/scenario that I talked about earlier accomplish.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
Mark Petersen is offline  
post #336 of 341 Old 02-18-2007, 09:51 AM
AVS Special Member
 
sethk's Avatar
 
Join Date: Jan 2003
Posts: 1,285
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 10
Quote:
Originally Posted by Mark Petersen View Post

As we discussed, I want to update the contrast project thread with DI dynamic contrast information which in combination with the static contrast information will provide a balanced feel for DI behavior. As suggested by Bob Sorel, I'm soliciting feedback, comments and criticisms from forum DI afficionados before updating the contrast project thread.

Since the dynamic contrast test patterns aren't finished yet, I decided to do the next best thing which is to use the measured VW50 full white (iris 1, iris off) and black level (iris 1, iris off) data from the static contrast benchmark and then to use this data to extrapolate a range of white levels (%stim) as specified by Erik Garci's contrast calculator (for a 2.5 gamma projector).

Since this uses a mix of measured and calculated values it's purpose is for illustration uses only, actual results may vary. This provides only one scenario for DI behavior, but I think it serves the purpose for a general illustration quite well.

The scenario chosen assumes that the overall APL is low so that the iris aperture is at it's smallest. The % stim data points correspond to pixel intensities that might exist in a mixed, but low APL scene. The assumption is that gamma is boosted enough to overcome light loss due to iris reduction if sufficient headroom is available. Plots are done in both linear and logarithmic scale as linear scale helps to show what happens with bright detail and logarithmic scale helps to show what happens with dark detail.

This graph shows a native 3100:1 projector (VW50 data iris off), this same projector with a simple DI added that only adjusts iris aperture, the same projector with a full featured DI that performs gamma boost and lastly a native 14000:1 projector which is suggested by the on/off CR of the projector with DI.



Mark,
Thanks for posting this. I'm somewhat confused by the numbers and captions. How does a "Native 14000:1" projector achieve a 14000:1 "intra-image simultaneous" CR with 100% stimulus?
Intra-image contrast is the contrast of darkest spot in the image vs the brightest spot in the same image (not in sequential frames, because that wouldn't be intra-image contrast.)
Are you telling me that a 14000:1 native CR projector could actually measure 14000:1 with total black and total white in the same image? What does this graph really plot?
sethk is offline  
post #337 of 341 Old 02-18-2007, 11:09 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 sethk View Post

Mark,
Thanks for posting this. I'm somewhat confused by the numbers and captions. How does a "Native 14000:1" projector achieve a 14000:1 "intra-image simultaneous" CR with 100% stimulus?
Intra-image contrast is the contrast of darkest spot in the image vs the brightest spot in the same image (not in sequential frames, because that wouldn't be intra-image contrast.)
Are you telling me that a 14000:1 native CR projector could actually measure 14000:1 with total black and total white in the same image? What does this graph really plot?

Assuming the use of about a 2% APL, for the 14000:1, think of this as 98% of the screen black and a 100% white square that is 2% of the total screen area (basically a 10" square on a 110" diagonal screen).

Glen Carter
Home Theater Calibration

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
GlenC is offline  
post #338 of 341 Old 02-18-2007, 11:22 AM
AVS Addicted Member
 
stanger89's Avatar
 
Join Date: Nov 2002
Location: Marion, IA
Posts: 17,397
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 112 Post(s)
Liked: 142
Quote:
Originally Posted by sethk View Post

Are you telling me that a 14000:1 native CR projector could actually measure 14000:1 with total black and total white in the same image? What does this graph really plot?

Yes, the best way to put it though is an x:1 native contrast PJ can approach x:1 as APL approaches 0. It's sort of a limit function. If you look at the measurements in the other thread you'll see this born out in the data (in fact the DiLA RS1 actually does better than On/Off with "0%" APL patterns).

See what an anamorphoscopic lens can do,
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
stanger89 is online now  
post #339 of 341 Old 02-18-2007, 11:48 AM - Thread Starter
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 sethk View Post

Mark,
Thanks for posting this. I'm somewhat confused by the numbers and captions. How does a "Native 14000:1" projector achieve a 14000:1 "intra-image simultaneous" CR with 100% stimulus?

Yes, GlenC has it right. A person cold also think of the extreme intra-image contrast case of an all black screen with a single 100% pixel.

Quote:


Intra-image contrast is the contrast of darkest spot in the image vs the brightest spot in the same image (not in sequential frames, because that wouldn't be intra-image contrast.)
Are you telling me that a 14000:1 native CR projector could actually measure 14000:1 with total black and total white in the same image?

Yes, theoretically a 14000:1 native CR projector could measure close to 14000:1 within the same image. I say theoretically because it's difficult to measure very small areas of full white. The static contrast measurements here show this sort of a relationship although we currently only have data down to 0.5% white (by area). Wm has taken measurements as low as 0.1% white (with correspondingly higher contrasts), but it becomes more difficult to fill a probe sensor with that small of an area.

Quote:


What does this graph really plot?

It shows the theoretical maximum dynamic contrast for a DI vs the same projector without a DI (iris off) vs a native projector with the same on/off contrast as the DI is able to achieve. The maximum dynamic contrast can only happen when the APL of the image is low so that the blacks aren't washed out. So this means a very small area of "white". This small area of white is then varied in brightness from full white (100%) to 0% and the maximum effects of gamma boost for the full range of white stimulus is plotted. This scenario uses measured VW50 data from the static contrast benchmark for iris off and iris 1 modes, but with extrapolated white levels for a 2.5 gamma along with the assumption that the blacks at each step are unchanged from the measured minimum full black. Actual measured results may vary although for illustration purposes it describes the relationships and what one would expect quite well.

In essence the DI (with dynamic gamma) allows a projector to better utilize the limited contrast that it has with a wider range of pixel brightness than what would be the case otherwise (ie w/o the DI). Although this flexibility comes with a price (brightness compression for example).

EDIT: I see Stranger89 and I are thinking along the same lines


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
Mark Petersen is offline  
post #340 of 341 Old 02-19-2007, 08:03 AM
AVS Special Member
 
sethk's Avatar
 
Join Date: Jan 2003
Posts: 1,285
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 10
Thanks for the replies, guys. I got the comparisons of the different curves, but I thought the points on each individual curve were plotting average APL versus contrast. I didn't get that it was plotting stimulus of a 2% patch versus contrast (probably because I didnt read back far enough).

A relative version of this graph would be interesting:
At 10% stimulus the maximum theoretical CR would be 10% of manufacturer quoted spec (assuming linear measurements).
So for a native 14,000:1 pj, if this measured value is 1400:1, it would be 100% of quoted spec.
If for a 15000:1 DI enable pj it was 1000:1, it would be 66.67% of theoretical max spec (at that stimulus) and so on.
Plotting the %age of measured / theoretical contrast vs stimulus of a 2% patch would also provide some interesting insight, although its just another view of the same data.

I find this thread very interesting, and I certainly am learning new things - I didn't realize that for small bright patches the contrast on non-DI projectors was this high.
sethk is offline  
post #341 of 341 Old 02-19-2007, 07:39 PM - Thread Starter
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
It's also interesting to look at the theoretical performance of these projectors with a real image. If you check out Erik Garci's Intrascene contrast calculator you can see how projectors with various ANSI and ON/OFF contrasts perform as far as theoretical contrast between the brightest and darkest pixels with a real scene. The calculator is here:


The space scene below has a low APL which reduces the washout effect so that high contrasts can be achieved.



It's interesting to compare the results of the Intrascene calculator with the plot that I posted. The Intrascene calculator predicts 7446:1 contrast between the brightest pixel (81%) and the black background. Since this is well above the maximum Intra-Image Contrast of the DI (which is 4500:1) we can see that the native 14000:1 on/off projector will have nearly twice the contrast in this scene.

For the space scene (ie low APL, max lum 81%):
Projector with Iris off - ~1857:1
Projector with DI on - ~4500:1
14000:1 on/off Native projector - ~8000:1

This is a glaring example of why the on/off CR metric alone is insufficient for characterizing or extrapolating the performance of a DI equipped projector. In fact the Intrascene calculator should be changed so that on/off alone is not the only predictor of contrast for DI projectors.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
Mark Petersen 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