AVS Forum banner
Status
Not open for further replies.

Dynamic Iris VS Variable Iris

2.4K views 19 replies 9 participants last post by  TomCat  
#1 ·
I've been thinking over the whole issue of the upcoming Sony 1080p SXRD sets having a dynamic iris and would like to ask other forum members what they think.


Basically, the dynamic iris is a device that will adjust brightness in real time based on program material using a mechanical device similar to an aperature in a camera. For instance, to get darker blacks in a dark scenes, the iris will close giving you a darker black. This isn't an issue in bright scenes because the iris in your eyes close down giving you the impression that smaller black areas are dark. Also this problem of blacks appearing as dark gray only really shows up when viewed in a fairly dark room. The problem is when the dynamic iris sees a dark scene, it will darken the picture, but if you're in a well lit room, you've just crushed your blacks and have made your overall picture darker, basically for no reason. It would seem that a dynamic iris system would also require some kind of ability to measure the room's ambient light to know whether or not to adjust the iris. But I haven't really seen any comments about the "dynamic iris" sets having such an ability.


Does anyone have any insights as to how the dynamic iris system will work??
 
#2 ·
Quote:
Originally Posted by JimP
Does anyone have any insights as to how the dynamic iris system will work??
Great!!! Just go down and look at the Samsung 1080p DLP's; they use a dynamic iris and all of the other RPTV's look dull by comparison.
 
Save
#3 ·
I've been wondering about this some myself. I would expect it to increase the range of the tv by making a completely dark scene darker attacking the grey blacks when they are at their worst, but I also wonder if it isn't doing it at the expense of decreasing the constrast ratio and detail in any given scene.
 
#4 ·
Quote:
Originally Posted by UCSB
Great!!! Just go down and look at the Samsung 1080p DLP's; they use a dynamic iris and all of the other RPTV's look dull by comparison.


Has anyone used a signal generator and test the display under various lighting conditions?? I've looked through the Samsung 1080p thread and didn't see one, but maybe I just missed it. I'd like something more than "just blows me away" type comments. I've been looking for a Samsung 1080p set today at all the likely spots and didn't find one.
 
#5 ·
Quote:
Originally Posted by AkaStp
Do we know if the dynamic iris can be turned off in the new Samsung 1080p DLPs. This would at least enable us to judge the effect of the dynamic iris.
No ... it's operation is completely transparent and a core part of the operation of the set. You do not have any idea it is operating.
 
Save
#6 ·
Quote:
Originally Posted by JimP
Has anyone used a signal generator and test the display under various lighting conditions?? I've looked through the Samsung 1080p thread and didn't see one, but maybe I just missed it. I'd like something more than "just blows me away" type comments. I've been looking for a Samsung 1080p set today at all the likely spots and didn't find one.
I haven't heard of anything like that ... just the usual signal generators: DVD players and broadcast signals. Keep looking ... when you find one of these sets the benefits will be apparent. From my personal perspective, at this point in time there really isn't any reason to run a lot of tests because these sets outperform the competition by such a margin it would be a waste of time to measure. Perhaps later in the year when all of the 1080p sets are on the market, people will attempt to measure and understand the differences.
 
Save
#7 ·
Quote:
Originally Posted by JimP
...to get darker blacks in a dark scenes, the iris will close giving you a darker black...The problem is when the dynamic iris sees a dark scene, it will darken the picture, but if you're in a well lit room, you've just crushed your blacks and have made your overall picture darker...


Does anyone have any insights as to how the dynamic iris system will work??
It sounds like you understand the mechanics of how the D-iris works pretty well, except for its relationship to blacks. It is important to distinguish the two main issues surrounding black performance (and the differences between them) to understand this. Crushing blacks is something that happens for one of two reasons, either the signal is clipped somewhere in the aquisition/production/delivery chain, or the picture/brightness/contrast settings in the display are misadjusted, allowing the darkest blacks to dip below the absolute black performance of the display. IOW, performance exceeds the linear window of operation of the transfer curve of the video amplifying system.


Any modern display can be set to not crush blacks. It is typically a one-time setting that works for all content (except content like "Cold Case", where clipped blacks are a part of the production style).


The reason this becomes thorny in LCDs and DLPs is because light leak inside the RP cabinet makes the absolute black level higher than in a plasma or CRT or direct LCD, and it makes the other problem, non-linear response, more noticeable, and more critical to tweak for.


But D-iris should not contribute at all to crushing black levels. If you take any similar application, such as a projector or a camera, decreasing the total light throughput by decreasing the iris aperture drops the light exponentially more for bright areas than it does dark areas, and black detail, while physically harder to see because it is simply darker than it was before, does not clip or crush in the conventional sense, because there is no non-linear part of the operational window. In optics, for all practical purposes, it is infinitely linear, meaning blacks can't clip.

Quote:
Originally Posted by JimP
...This isn't an issue in bright scenes because the iris in your eyes close down giving you the impression that smaller black areas are dark. Also this problem of blacks appearing as dark gray only really shows up when viewed in a fairly dark room...
The reason D-iris can be effective is because it can actually lower the absolute black level for darker scenes without crushing the blacks. You struck here on exactly why this will work. When there is a predominance of light areas in the picture, our eyes (and perception) automatically compensate for the higher-than-normal absolute black level. Under such conditions a higher absolute black level doesn't seem that noticeable or out of place. Only when there is a dark picture does it seem higher than it should be. If a D-iris system can simply take an RMS value of the total brightness in a picture and use that voltage signal to close down or regulate the iris in front of the projector lamp (and also compensate by turning up the video brightness) during DARKER scenes the actual light LEAK level for dark scenes INSIDE THE RP CABINET will drop dynamically from its "always on full brightness" state, and absolute black levels will drop correspondingly for dark scenes, exactly where and when we need them to.

Quote:
Originally Posted by JimP
...to get darker blacks in a dark scenes, the iris will close giving you a darker black...The problem is when the dynamic iris sees a dark scene, it will darken the picture...
This looks like where the confusion lies. For a D-iris system to work, it will allow full light output for the brightest scenes, and lowest light output for the darkest scenes, and correspondingly more for mid-bright scenes, and not change the actual black levels themselves at all, just the ABSOLUTE black level. That will lower the absolute black only when needed, and allow full brightness during bright scenes when the absolute black level is less objectionable and less noticeable.


And it may not darken the "picture" at all. What changes is the amount of light that illuminates the inside of the cabinet. It is light leak from that lamp that is causing the problem of higher absolute black in the first place. Placing an iris in front of it will drop the light LEAK IN THE CABINET and lower the absolute black dynamically dependent on content at any given second.


True, this will also lower the light that is directed through the light engine, but if the S/N levels used there can support raising the video level correspondingly high enough to offset the lowered light input during the moments when the iris is closed down, the resultant output video level itself should not vary much, if at all. It's all about balancing the RATIO of light leak inside the cabinet with the actual output video level delivered directly through the light engine. You manipulate both by lowering both for dark scenes, and then change the ratio for dark scenes by dynamically raising the video level back up to compensate, sort of like the way analog audio compressors work. IOW, the control signal voltage that closes down the iris has a correspondingly opposite control signal voltage that raises the video level back to normal output levels, and that is after the fact and separate from the manipulation of absolute black, so does not raise it back up.


The only trade-off should be noise performance in the blacks (noise levels will increase as the D-iris darkens the light output while video levels are amplified to keep the actual output levels normalized). If they can keep that low, then it will work effectively.
 
Save
#8 ·
Tom Cat: so what you're saying is that there might be some difference in black noise level performance between various attempts at dynamic irises by different companies. Even though I'm a fan of great blacks I would hope that if forced to make a choic the companies designining irises would err on the side of less noise in blacks than blacker blacks. Noise to me is more annoying than less than perfect blacks--even though I'm a great fan of those. One last question: Does the use of a dynamic iris negate the need to ever use a neutral density filter with it? Could a filter that barely made blacks blacker enhance the iris or does an iris work better without out it or does it make any difference? Any other commentators besides TomCat are welcome to reply.
 
#9 ·
One would hope the 'iris' would eliminate the need for an ND filter. This is newer technology and deserves more time to prove itself.

To bring the best out of any Grand Wega display, we still need either a full calibration or those specialized UMR (et al) tweaks many of us hold dear to our hearts. Those proper tweaks in conjunction with the 'iris' should produce the best possible picture with the given technology. Whether an ND filter can improve on at that, time will tell.
 
Save
#10 ·
Quote:
Originally Posted by videoaddikt
One would hope the 'iris' would eliminate the need for an ND filter.


I would think that it would depend on how well a dynamic iris is implemented. As with anything, poor implementation could result with less than satisfactory results.


I look forward to seeing how the new Samsung 1080p owners see the sets once the newness wears off. Will they see the dynamic iris get confused and change during scenes that are an odd mixture of dark and lighter shades?
 
#11 ·
I am totally unclear why you think you need a room-light sensor?


First, the iris on the Sony -- at least the HS51 -- isn't willy nilly. It doesn't try to adapt to every single scene change instantly.


Secondly, once it adapts to a dark scene, it will modify the gamma so as to provide a series of gradiations that were simply crushed to nothing before on the base gamma.


The dynamic iris can not only lower the black levels luminance, it can also restore detail that was not reproducible in the "full open" state where a difference gamma had to be used.
 
Save
#12 ·
Quote:
Originally Posted by rogo
I am totally unclear why you think you need a room-light sensor?


First, the iris on the Sony -- at least the HS51 -- isn't willy nilly. It doesn't try to adapt to every single scene change instantly.


Secondly, once it adapts to a dark scene, it will modify the gamma so as to provide a series of gradiations that were simply crushed to nothing before on the base gamma.


The dynamic iris can not only lower the black levels luminance, it can also restore detail that was not reproducible in the "full open" state where a difference gamma had to be used.


Front projectors are used in total (or almost total) darkness. Rear projectors are used in dark, light and everything inbetween. In a room with bright lighting, it would seem that with a rear projector using a dynamic iris to darken an already dark scene would be counter productive.


I've read what TomCat posted and am still digesting it. Will probably make sense later.
 
#13 ·
Quote:
Originally Posted by JimP
Front projectors are used in total (or almost total) darkness. Rear projectors are used in dark, light and everything inbetween. In a room with bright lighting, it would seem that with a rear projector using a dynamic iris to darken an already dark scene would be counter productive.
Samsung 1080p sets have different operating modes: MOVIE, STANDARD, DYNAMIC ... to give you a quick way to compensate for large changes in room lighting / environment. For example, if your viewing in a darker environment, MOVIE will give you the best results. In extremely bright environments, DYNAMIC will increase lamp output and adjust settings. As an owner, you optimize these settings for your different viewing situations.
 
Save
#14 ·
Quote:
Originally Posted by JimP
Front projectors are used in total (or almost total) darkness. Rear projectors are used in dark, light and everything inbetween. In a room with bright lighting, it would seem that with a rear projector using a dynamic iris to darken an already dark scene would be counter productive.
No. It will still give you detail that would be lost otherwise.
 
Save
#15 ·
Crutchfield has a pic of the iris used on the Sony 50A10. It illustrates both the pen and closed positions. There's surprisingly not much to the device.


Here's a link to the pic: Link
 
Save
#16 ·
I'm not sure if this has any bearing on the discussion, but the Sony 42/50A10 sets' iris settings -replace- the Picture setting in the menu. There are only 4 settings. This would indicate to me (rather confusingly) that the iris and various electronic manipulations are affecting white level, not black level. Someone care to comment?
 
#19 ·
Quote:
Originally Posted by UUronl
Right, but from a calibration standpoint it seems that it is automatically keeping the whites from crushing or blooming, but we must calibrate Brightness the old-fashioned way.
The iris does present calibration challenges. But it can be done on the HS51. And once done, there are not PQ issues with the iris.


I think the RPTVs will be equally pleasing.
 
Save
#20 ·
Quote:
Originally Posted by rogo
The iris affects black and white level.


If you stop down the iris, peak white is inherently reduced.
Absolutely. That is why dynamically raising the video level to compensate would be necessary, and would need to work in concert with dynamically restricting the lamp output. Light leak is reduced, reducing absolute black, while net video output levels are compensated for and remain the same.
 
Save
Status
Not open for further replies.
You have insufficient privileges to reply here.