AVS Forum banner
1 - 20 of 22 Posts

·
Registered
Joined
·
9,959 Posts
Discussion Starter · #1 ·
Hello,


I'm curious about how a manual iris affects calibration.


I have use dynamic iris in the past and it certainly affected the brightness, contrast, and gamma. In fact it gave the gamma a funky shape so I stopped using the dynamic iris.


However,... I never tried the manual or static iris. I guess that's because my pj doesn't put out enough light. However, having recently put a new bulb in I'm hoping I will have enough output to try to use the manual iris in my next calibration.


-Brian
 

·
Banned
Joined
·
3,587 Posts
The iris does NOT give gamma a "funny shape" - the problem is the way you measure gamma NOT the way the iris works. To judge the effect of the iris you must measure ALL STEPS WITHIN 1 FRAME OF VIDEO in order to understand that the iris only makes images brighter or darker and does NOT really change gamma. YOU CANNOT measure different grayscale steps with different iris openings... that is completely wrong and it does not represent what you see in projected images. You must consider projected images as a series of still images... like many individual photographs. You see each "photograph" (or frame of video) with ONE iris opening. If you measure window or full-screen test patterns with Auto Iris active you will measure the images with many different IRIS openings but YOU WILL NEVER SEE THAT HAPPEN WHILE YOU ARE WATCHING VIDEO so you cannot measure the grayscale that way and get anything that makes sense.


That said, gamma will change a LITTLE when you compare fully closed measurements to fully open measurements just because gamma is calculated based on a value of white close to 100% (some software will subtract a small number from the 100% white measurement... if the measurement was 16 fL for 100% white, the software might use 15.9 fL and call that 99% white. Gamma does not exist for 100% or 0%... it only exists for luminance values above 0 and below 100. The change in gamma between iris open and iris closed is small and not really possible to detect in video images.


So you must turn Auto Iris off for calibration measurements. You might want to make a manual iris setting in the middle of the adjustment range so when you turn Auto Iris on when you are done, the "distance" from middle point to fully closed or fully open is as small as you can make it.


The real purpose of the Auto Iris is to improve the perceived black level in dark scenes, so you could calibrate with Auto Iris open (fully) and just let the iris make the image darker... but that doesn't leave you any opportunity to open the iris as the lamp ages to get more light.


These are all trade-offs and how you deal with them is part of the art of calibration. This particular question comes up a lot because people do not understand what is happening with the iris and how it relates to calibration. What I have said here is 100% accurate... if you do not think that is correct, then you do not understand the issue yet. You have to sometimes think about this for a while before you understand what I am saying here. I will not argue with anyone about this description of why what you see when you make measurements with Auto Iris turned on is wrong---if you think what I said here is wrong, you just haven't grasped the concept yet and I cannot magically make that happen for you. I am saying this now because it always seems that SOMEBODY wants to argue about this with me when the point is not arguable...


Consider taking a photograph and putting an adjustable light on it. Make the light very bright ... then make the light dimmer... does the relationship of light to dark change in the 2 cases? No. (therefore the gamma of that image is not changed in any meaningful way) All that changes is how bright or dark the image is. Gamma has to do with the relationship between each shade of gray and the next shade darker or lighter... that cannot happen in video when you change the iris - and that is analogous to making a light illuminating a photograph lighter or darker..
 
  • Like
Reactions: h2c

·
Registered
Joined
·
9,959 Posts
Discussion Starter · #3 ·
Oh ok.


I may have time to do some measurements today.


I'm sure most people here have a different level of understanding and I come here to learn.


I will be the first to admit I don't understand fully the auto iris and how to calibrate with it. I don't use the auto iris in my projector. I do understand your way of working with the auto iris. However, I'm not interested in using the auto iris. That's like the video equivalent of holding the volume knob during an action movie to turn down the explosions. (A.K.A Dolby night mode)


I am interested to see if I can find the manual or static iris useful. So far, I haven't used the static iris either....


I have experimented with the auto iris previously and my thoery was to calibrate with the settings that would be in use therefore that's how I arrived at the gamma with the funny shape. I appreciate and understand what you say about it.


Thanks again,


Brian
 

·
Registered
Joined
·
2,760 Posts
Opening or closing the manual iris does often change white balance. It shouldn't impact the correct brightness or contrast settings in most situations. So basically all you should have to worry about if you make a big fixed iris change would be redoing the white balance.
 

·
Registered
Joined
·
9,959 Posts
Discussion Starter · #5 ·
Hi,


Spent hours calibrating today but even with a 30 hour bulb I don't get enough light output on my five year old pj to let me consider using the manual iris.


So I needed have asked for now.


Anyways,... Thanks again.
 

·
Banned
Joined
·
3,587 Posts
Your assement of the value of an Auto-Iris feature is true for some projectors with poor auto-iris implementations, but for projectors where the Auto-Iris feature works WELL and by working well I mean that it does not make noise above the level of the lamp cooling fan and that the changes in the opening are not perceptible during normal video content (forget test patterns, you could pretty easily devise a test pattern sequence to trip-up ANY auto-iris feature.


It just so happens that Sony's Auto-Iris feature is exceptional in it's operation and you are unlikely to notice it in operation for more than 1 or 2 seconds over 10s or 100s of hours of viewing movies or TV programming. Some other companies, such as one that used to be the premier computer printer manufacturer back in the dot matrix printer days, have Auto-Iris systems that make too much noise AND the "gain riding" on luminance is visible in video content with some regularity. So just writing-off the Auto Iris is a big mistake if you happen to have one of the brands/models where it is very well-implemented.


You don't "work with" Auto Iris during calibration... you just turn it off while you are calibrating. Usually I would calibrate the projector with the manual iris in the most open setting used by Auto Iris and set that to get 16 fL. If the Contrast control does not permit that to happen (operates funny or does not have enough adjustment range), you may have to close the iris somewhat and see if there is an Auto Iris mode that also "stops" at that less than fully open setting - not usually too likely unless there is more than 1 Auto Iris mode. Some Auto Iris systems have only an on or off option


When the calibration is done and everything looks right, you turn the Auto-Iris back on. The purpose is not to make images brighter than the nominal 16 fL SMPTE standard (the standard allows for 12-20 fL from a projection screen). The real purpose of the Auto Iris is to mask the fact that the projector's black level is not as impressive as you might wish. If you are blessed with an Auto Iris control that works well, believe me, it will make images look far more impressive than they can look in dark scenes with milky/foggy blacks.


In a theoretically perfect optical system, opening and closing the iris won't change anything but luminance. In the real world where optical systems are NEVER 100% perfect, small shifts here and there MIGHT happen, but they are generally so small that they will not be visible... for example, you'll never see STOP sign red shift to a shade of reddish-orange just because the Auto-Iris opens or closes, but you might measure a small shift in red in one direction or another. Is it anything to worry about? Not likely, but that's why you have a meter and calibration software, you can measure the differences and see if the errors are large enough to be worrisome.
 

·
Guest
Joined
·
867 Posts
Does not an iris alter contrast ratio if it is in the lens as the smaller the iris in the lens the less unwanted stray light that gets through and onto the screen?


Does not a dynamic iris also alter contrast ratio by matching the brightest the chip can go to the brightest thing to be displayed on screen so enabling the full range of the chip to be used as far as going down to black?


Does not the above changes in the contrast ratio being displayed mean the contrast between different black-grey-white steps in the grayscale are different, that gamma in the display is changed as the iris and black level change?


For example with a dynamic iris an image with both 100% white and 0% black opens the iris so max chip brightness matches 100% white on screen luminance and can use the full range of the chip down to 0% black. Another image has only 60% white and 0% black so the iris is closed down so max chip brightness matches 60% white on screen luminance and the full range of the chip can be used down to a lower luminance 0% black. The contrast ratio in the second image 60% white to 0% black is higher than it is in the first image 60% white to 0% black. If the contrast ratio between steps in the greyscale is different then the gamma is different.


Or is the only step that changes the one to black, only 0% black changes, so gamma does not change as 0% black is supposed to be the absence of light. In which case any image without 0% black in it would not benefit from a dynamic iris.
 

·
Registered
Joined
·
12,417 Posts
Does not an iris alter contrast ratio if it is in the lens as the smaller the iris in the lens the less unwanted stray light that gets through and onto the screen?
Too bad this thread went cold. I was keen to see the answer to this question, as I have seen folks reducing the aperture setting to improve contrast. E.g.:
I go put on Pacific Rim and drop my High lamp iris to -10, -12 and am seeing for the first time those higher contrast ratio blacks with still bright for my happy taste images in 3D.
How is high lamp with reduced aperture different than low lamp with open aperture regarding contrast ratio?
 

·
Registered
Joined
·
11,482 Posts
Too bad this thread went cold. I was keen to see the answer to this question, as I have seen folks reducing the aperture setting to improve contrast. E.g.:
How is high lamp with reduced aperture different than low lamp with open aperture regarding contrast ratio?
Yes, closing down the aperture definitely improves the contrast; i.e., the black level drops proportionately more than the white level.
 

·
Registered
Joined
·
12,417 Posts
Yes, closing down the aperture definitely improves the contrast; i.e., the black level drops proportionately more than the white level.
Thanks. Is the mechanism as Dovercat asks -- stray light scatter in the lens?

I've read that placing the PJ as close to the screen as possible will increase contrast. Does that not also use more of the lens surface?

This is not very intuitive. :(
 

·
Registered
Joined
·
11,482 Posts
Thanks. Is the mechanism as Dovercat asks -- stray light scatter in the lens?

I've read that placing the PJ as close to the screen as possible will increase contrast. Does that not also use more of the lens surface?

This is not very intuitive. :(
You piqued my curiosity so I did some measurements on my x570, in Cinema Mode, at low/high power and min/max zoom. Closing the Iris seems to help only for Max Zoom. Here are the results:



LP
Cinema MinZ Auto2 83000:1
Cinema MinZ Iris 0 12800:1
Cinema MinZ Iris -7 13700:1
Cinema MinZ Iris -15 12700:1

HP
Cinema MinZ Auto2 99000:1
Cinema MinZ Iris 0 12900:1
Cinema MinZ Iris -7 13900:1
Cinema MinZ Iris -15 12900:1

LP
Cinema MaxZ Auto2 71000:1
Cinema MaxZ Iris 0 9200:1
Cinema MaxZ Iris -7 11000:1
Cinema MaxZ Iris -15 11500:1

HP
Cinema MaxZ Auto2 93000:1
Cinema MaxZ Iris 0 10500:1
Cinema MaxZ Iris -7 12100:1
Cinema MaxZ Iris -15 12700:1
 

·
Registered
Joined
·
1,340 Posts
Thanks. Is the mechanism as Dovercat asks -- stray light scatter in the lens?

I've read that placing the PJ as close to the screen as possible will increase contrast. Does that not also use more of the lens surface?

This is not very intuitive. :(
In general, closer to the screen increases brightness and lowers contrast, the opposite for farther from the screen. Its because of the "aperture" reducing light transmission. Imagine a room(lens tube)with a window(front lens element) and you are the internal lens that moves when you zoom. Standing right next to the window(zoom in for close to the screen)its very bright and not shadowy. Now, walk to the other side of the room(zoom out for far from the screen)its much less bright and more shadow. The size of the window(aperature)didnt change, but the added distance and internal volume decreases the direct light(lowering brightness) and the reflected light(increasing contrast). Hope this helps your understanding.
 

·
Registered
Joined
·
12,417 Posts
You piqued my curiosity so I did some measurements on my x570, in Cinema Mode, at low/high power and min/max zoom. Closing the Iris seems to help only for Max Zoom. Here are the results:
Thanks much for taking the time for measurements. They are interesting and useful.

I take it the Auto2 mode is dynamic iris mode? And the contrast test is performed with a 2-step measurement, 100 IRE test pattern ("full on"), and 0 IRE test pattern ("full off"), then compute the ratio? As opposed to contrast within a given scene (aka ANSI contrast)?

Does this imply this whole discussion of aperture is mainly a matter of how black is black when all is black (or a very dark scene), or does it also impact the contrast within a static image? I would think that stray light scatter inside the lens or light engine would primarily affect the latter.
 

·
Registered
Joined
·
12,417 Posts
In general, closer to the screen increases brightness and lowers contrast, the opposite for farther from the screen. Its because of the "aperture" reducing light transmission. Imagine a room(lens tube)with a window(front lens element) and you are the internal lens that moves when you zoom. Standing right next to the window(zoom in for close to the screen)its very bright and not shadowy. Now, walk to the other side of the room(zoom out for far from the screen)its much less bright and more shadow. The size of the window(aperature)didnt change, but the added distance and internal volume decreases the direct light(lowering brightness) and the reflected light(increasing contrast). Hope this helps your understanding.
Thanks. This is the post that threw me off: Link
 

·
Registered
Joined
·
11,482 Posts
I take it the Auto2 mode is dynamic iris mode? And the contrast test is performed with a 2-step measurement, 100 IRE test pattern ("full on"), and 0 IRE test pattern ("full off"), then compute the ratio? As opposed to contrast within a given scene (aka ANSI contrast)?
Yes, those measurements are for sequential (inter-scene) full on/off contrast. The intra-scene contrast will be far lower, in the order of 1,000 rather than 10,000 or 100,000.
Does this imply this whole discussion of aperture is mainly a matter of how black is black when all is black (or a very dark scene), or does it also impact the contrast within a static image? I would think that stray light scatter inside the lens or light engine would primarily affect the latter.
Yes, dynamic iris affects primarily the full on/off contrast, unlike local-dimming TVs that improve the ANSI contrast.
 

·
aka jfinnie
Joined
·
6,078 Posts
You piqued my curiosity so I did some measurements on my x570, in Cinema Mode, at low/high power and min/max zoom. Closing the Iris seems to help only for Max Zoom. Here are the results:

LP
Cinema MinZ Auto2 83000:1
Cinema MinZ Iris 0 12800:1
Cinema MinZ Iris -7 13700:1
Cinema MinZ Iris -15 12700:1

HP
Cinema MinZ Auto2 99000:1
Cinema MinZ Iris 0 12900:1
Cinema MinZ Iris -7 13900:1
Cinema MinZ Iris -15 12900:1

LP
Cinema MaxZ Auto2 71000:1
Cinema MaxZ Iris 0 9200:1
Cinema MaxZ Iris -7 11000:1
Cinema MaxZ Iris -15 11500:1

HP
Cinema MaxZ Auto2 93000:1
Cinema MaxZ Iris 0 10500:1
Cinema MaxZ Iris -7 12100:1
Cinema MaxZ Iris -15 12700:1
They're surprisingly low numbers - I usually see over 20k:1 on my old X30 (with no auto-iris). Is that off screen or lens?
 

·
Registered
Joined
·
11,482 Posts
They're surprisingly low numbers - I usually see over 20k:1 on my old X30 (with no auto-iris). Is that off screen or lens?
The newer low end JVCs (RS4x0) have lower native contrast than previous models. I get between 12,000 and 20,000 mid zoom and low power and iris 0, depending on the unit.
Only a small difference measured on screen or off lens.
 

·
aka jfinnie
Joined
·
6,078 Posts
The newer low end JVCs (RS4x0) have lower native contrast than previous models. I get between 12,000 and 20,000 mid zoom and low power and iris 0, depending on the unit.
Only a small difference measured on screen or off lens.
Glad I'm going from X30 to X7900 next then, that would have been a let down... :(
 

·
Premium Member
Joined
·
12,752 Posts
Yes, those measurements are for sequential (inter-scene) full on/off contrast. The intra-scene contrast will be far lower, in the order of 1,000 rather than 10,000 or 100,000.

Yes, dynamic iris affects primarily the full on/off contrast, unlike local-dimming TVs that improve the ANSI contrast.
Local dimming is not designed to improve ANSI contrast, it is designed to improve on/off contrast. Local dimming is done so that you don't have to compromise gamma as much as doing a full screen dim.
 

·
Registered
Joined
·
12,417 Posts
Local dimming is not designed to improve ANSI contrast, it is designed to improve on/off contrast. Local dimming is done so that you don't have to compromise gamma as much as doing a full screen dim.
:confused: Local dimming was introduced back when TV's used LCD panels which could not completely shut down the backlight in dark areas of the screen when other areas needed high brightness. Modulating the entire backlight (akin to dynamic iris) would limit the brightness which was unacceptable. Local dimming divides the backlight into smaller zones -- it looks like a low-res version of the video converted to black and white. That lets the bright areas pop unimpeded and the dark areas to go black.

Isn't that what ANSI contrast refers to -- the contrast within a B/W checkerboard -- a single image presenting bright and dark areas?
 
1 - 20 of 22 Posts
Top