AVS Forum banner
1 - 13 of 13 Posts

·
Registered
Joined
·
920 Posts
Discussion Starter · #1 ·
So I keep hearing from various posts here that if you clamp down the iris (assuming a manually adjustable model), you increase the native contrast ratio. 2 questions....


1. Why and how? If the brightest white and darkest black are proportional to each other, wouldn't they "follow" each other with overall brightness? (Hence the contrast ratio wouldn't change.


2. Let's say I bought a JVC or Sony, and ran it on high lamp, but closed the manual iris to a minimum, because there is plenty of light. Everything looks great. But, as time marches on, of course the lamp will start dimming.


After about 500-600 hours, I may have opened the iris up to 50-60% to compensate. I am still enjoying the EXACT same light output that I had settled on in the beginning, but did I loose native contrast? Or does the dimmer lamp 're-factor' into the equation and the CR is unchanged???


I've heard a few times that the picture can look muddy after a thousand or so hours, but is that just because it's dimmer to the user/what they were used to? I realize that a fresh new bulb is always very rich and bright, but if you can 'settle' on a certain brightness with the use of an iris and literally maintain that exact brightness with iris control, will the picture actually degrade by slowly loosing native contrast?
 

·
Registered
Joined
·
23,131 Posts

Quote:
Originally Posted by cbaseuser /forum/post/20884419


So I keep hearing from various posts here that if you clamp down the iris (assuming a manually adjustable model), you increase the native contrast ratio. 2 questions....


1. Why and how? If the brightest white and darkest black are proportional to each other, wouldn't they "follow" each other with overall brightness? (Hence the contrast ratio wouldn't change.

Because "desired" light isn't the only thing that comes out of the projector's lens, there's an amount of scattered light (from internal components) that comes out too. This amount of scattered light is essentially constant, so it is added to both black and white, and affects black relatively a lot more.


Part of projector design is a balance of light output vs reducing scattered/waste light. Closing down the iris reduces scattered light, thus improving CR (as it's a much larger proportion of black than white.

Quote:
2. Let's say I bought a JVC or Sony, and ran it on high lamp, but closed the manual iris to a minimum, because there is plenty of light. Everything looks great. But, as time marches on, of course the lamp will start dimming.


After about 500-600 hours, I may have opened the iris up to 50-60% to compensate. I am still enjoying the EXACT same light output that I had settled on in the beginning, but did I loose native contrast? Or does the dimmer lamp 're-factor' into the equation and the CR is unchanged???

Not really (I don't think), because you're opening the iris, thus allowing more light (both desired and scattered).


Everything is related proportionally to the amount of light coming out of the lamp. So while the lamp may dim, you're still going to have the same proportion of scattered light, so opening the iris will increase the amount of 'waste' light just like on a new lamp.


At least in general/in theory. Reality varies as how much effect the iris has on things. The Sharp Z20000 had two irises that could affect contrast by maybe a factor of 4-8x (min to max), where as other projectors it's maybe more like 30-40%.
 

·
Registered
Joined
·
1,915 Posts
So dlp's seem to benefit more from an iris when it comes to native contrast improvements.


On a related not, I've never seen a dlp with a dynamic iris. On a very dark scene, the iris would be closed down. Does the image not suffer from as much brightness compression in comparison to an lcd?
 

·
Registered
Joined
·
8,006 Posts
It doesn't have anything to do with the DMD technology. It is just that the Sharp DLP iris was the best in the business. I see no reason why the same iris technology couldn't be used for LCoS or LCD projectors.


The only limitation is that the projector must have high light output and reasonably good native CR without the iris (2-3000:1). Otherwise, the hit to light output is too great for the benefit received.

Quote:
Originally Posted by johnsmith808 /forum/post/20886160


I thought a while back it was accepted that dlp has the greatest native contrast improvement when it comes to fixed irises. Or was that just unique to the Sharp projectors?
 

·
Registered
Joined
·
4,880 Posts
DLPs gain the greatest static CR improvement from fixed iris because their design inherently produces the highest amount of light scatter. Unlike LCOS/LCD, DLPs do not absorb “off state” light instead it is directed into a light trap, leading to greater light scatter.


Conversely transmissive LCDs produce the least amount of light scatter which is why a lens iris is not used with LCDs and why we don’t see any significant increase in static CR from LCD devices. Sony had lens iris in their early LCDs but got only a 15% CR increase and so abandoned that approach.


The Sharp (later adopted by Marantz) dual iris is the most efficient design in terms of CR increase. Using a combination of cats eye shaped lamp and lens iris the Sharp was able to increase its static CR by 4X with a 1/3 reduction in light output (re gregr measurements).


SOTA SXRD designs yield a 2X CR increase paid for by a 1/3 light reduction. LCOS does not seem to get the same CR benefits from a lamp iris as does DLP.
 

·
Registered
Joined
·
23,188 Posts

Quote:
Originally Posted by HoustonHoyaFan /forum/post/20886687


DLPs gain the greatest static CR improvement from fixed iris because their design inherently produces the highest amount of light scatter. Unlike LCOS/LCD, DLPs do not absorb “off state” light instead it is directed into a light trap, leading to greater light scatter.

This has been my impression too. I haven't seen evidence that the dual iris thing works on non-DLP projectors like it does on those, where if done right it feels a little like magic to me.


It looks like it was over 6 years ago that I was able to get about 9k:1 native on/off CR at around 100 lumens on high bulb from an Optoma H79 using the dual iris idea (although I went with 4 irises where 2 sets were only for red to get close to D65 at the above).


When I would put a white piece of paper into the lens (after taking the front off) at the iris plane what I would see was a little bit like if the lens opening was a quarter and there were two dimed sized circles of light that overlapped a little bit (they would have overlapped more with 10 degree mirrors than 12 degree mirrors). One of those was white and the other basically magenta to my eyes. I believe that it was best to block the magenta one to get the best CR since it didn't contribute as much to white as the other one which was really more stray light.


You likely know this HHF, but expanding on what stanger89 said about desired light, it might be useful to consider a graph even though it isn't really accurate. It is more for illustration of some principles. There are some Gaussian curves here:

http://en.wikipedia.org/wiki/Gaussian_function


on the top right. For the moment we could consider the full -5 to +5 as the diameter of the lens and the red line as light for white passing through a plane in that lens where an iris could be placed. Let's also assume that the amount of light for black is uniform across that same plane, which has an area of pi * r^2, or about 78 units squared. If the lens is left completely open the amount of light for white will be the area under the red line and the amount of light for black will be the intensity of the black over the 78 units squared.


In order to increase the ratio between white and black a round iris could be put in the plane that only goes from -2 to +2. A little bit of white will be lost because some of the area under the red line is cut off. But a lot of black (at least as a percentage of its total) will be lost because now the area for the black to pass through is less than 13 units squared. In the idealized situation we are assuming for now the black will have dropped by about 84%, but looking at the Gaussian curve the white will have dropped probably 10% or less. Of course, this isn't completely reality and not many DLP projector manufacturers would leave the lens completely open. In this case the CR could be increase more and more by tightening the irises further and further, but in reality the curves aren't quite ideal like that and there is only so much tightening of the iris that works, especially without adding a 2nd iris.


To consider how a 2nd iris could help with a DLP let's go back to considering the round iris from -2 to +2 in the lens. Now an iris could be placed at a well chosen spot before the DMD to line up with that iris as perfectly as possible. When done right the white light that is controlled well would pass through the first iris, hit the DMD and for the mirrors that are on, mostly pass through the opening in the lens iris. Putting the lamp iris (although it might not be that close to the lamp) in has helped reduce light that would have lowered on/off CR.


Now starting from this case of 2 round irises that both go from -2 to +2 lets consider what happens if the right half of the lens iris was blocked. Since both halves of the iris basically have the same amount of light going through them for white and for black this would reduce both of those to half, which would keep the on/off CR the same. But now if we block half of the lamp iris in a way that perfectly lines up with the lens iris something interesting has happened. The white light pretty much makes it through keeping the white level about the same, but the amount of light getting to the screen for black has been reduced to almost half because the amount of light hitting the DMD and scattering has been reduced to half.


Looking at it again, blocking half the lens iris cut the total amount of light for black in half because it blocked half the area. Blocking half the lamp iris cut the amount of light for black in half because it reduced the amount of light per unit area to the iris plane in the lens to half. But it didn't do the same for white because white is the more controlled light (not random). In this ideal case there is no loss, but in the real world there would be some in my experience.


Although idealized I think the above explains somewhat why the Sharp 12k didn't lose much light for white when going from one iris being shut down (the medium contrast mode) to both irises being shut down (the high contrast mode), but did increase the on/off CR significantly. I think at least one other model didn't have their dual irises lined up well and so didn't get the effect that they could have gotten with better iris designs.


I hope the above makes sense even though I took ideal situations that only apply so far in the real world.


--Darin
 

·
Registered
Joined
·
4,880 Posts
Darin - great to see our resident CR master posting again. Your expertise has been missed around here.


If I read correctly, you used round instead of cat's eye shaped iris in your Optoma mod?


I am really surprised that you did not patent and market your mod, 9K:1 was unheard of when you first presented the mod back then.
 

·
Registered
Joined
·
23,188 Posts
Quote:
Originally Posted by HoustonHoyaFan
Darin - great to see our resident CR master posting again. Your expertise has been missed around here.
Thanks.
Quote:
Originally Posted by HoustonHoyaFan
If I read correctly, you used round instead of cat's eye shaped iris in your Optoma mod?
I ended up closer to cat's eye, but not exactly. There are probably some pictures around here somewhere from years ago. I basically had a clear iris and red iris next to each other in the lens and then the same to line up with those back between the lamp and the DMD. These where in the same places where Optoma had their irises, although they were pretty open.
Quote:
Originally Posted by HoustonHoyaFan
I am really surprised that you did not patent and market your mod, 9K:1 was unheard of when you first presented the mod back then.
It wasn't a lot of light output, but worked for me off a High Power screen. I tried to talk to somebody from Optoma, but they didn't seem to be willing to discuss it under NDA, so I didn't pursue it.


--Darin
 
1 - 13 of 13 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top