View Full Version : Is it true that the dynamic iris doesn't work well in letterbox ?
Jones_Rush 02-14-07, 05:49 PM I've read that when you project a 2.35:1 letterboxed image, the DI doesn't understand that the top and bottom bars are not part of the movie, and hence the picture becomes in many scenes dimmer than what it should have been (if all the panel was used).
This is unbelievable if correct. Yes, I heard all sorts of reasons of why to go with an anamorphic lens, but this is definitely the most bizarre.
stanger89 02-14-07, 06:05 PM I'm not sure how I'd word it exactly. But I'll put it this way.
Most DI implimentations (at least the good ones) analyze the APL and distribution of luma to determine what iris and gamma setting to use for a particular scene.
Letterbox bars will bias the image toward low APL and will weight the distribution likewise.
Thus the DI algorithm will choose an iris/gamma setting lower than it would if it were to analyze only the picture area.
As an illustration, average luminance for a full-white 2.35:1 image in a 16:9 frame would be approximately 67%, vs 100% for just the active picture area. However even for this case it's hard to say what's "right", given that you are projecting bars as part of the image. Should they be ignored, and display potentially far brighter than necessary, or should they be included and "drag down" bright scenes.
Now, considering real-world scenes are far, far less than 100% average luma, more like 15-20% average luma, and the distribution is far greater than the 100% case. Thus it is even harder to say if it's really "wrong" or "bad" for DI to be affected by the bars. The bars are often one of the key places one notices an elevated black level.
I guess I'd chalk it up as another positive for for CIH, but I probably wouldn't really consider it a negative against non-CIH.
J.Mike Ferrara 02-14-07, 06:10 PM First, TI gives us dancing mirrors and spinning wheels, then Sony/Panny gives us flexing, breathing lenses. Woosh! All this extra motion is making me dizzy.
No thank you. I'll take my video neat. ;)
Jones_Rush 02-14-07, 06:11 PM Well, the black bars are not really an issue for many people, who mask with black velvet around their screen, or around the 2.35:1 window on the projector's lens. For those people, who do not go with an anamorphic lens, inclusion of the black bars in the DI algorithm is bad news.
I guess I'd chalk it up as another positive for for CIH
Some people are doing CIH without an anmorphic lens (like with zoom), and it's bad news for them too.
Actually, using a 1080p projector, I thought of doing CIH without an anamorphic lens nor zoom, by going with 1446x817 for 16:9 movies. Now this DI problem might have shot down my entire plan, because at 1446x817 mode I'm using just 57% of the panel, which will make the DI algorithm go nuts.
My Sony Pearl has two auto-iris modes. If you find that Auto1 is closing down too tightly on 2.35:1 content, you could always switch to the less aggressive Auto2.
I can honestly say that I have never found there to be any problem with 2.35:1 material in either iris mode.
Jerry Gardner 02-14-07, 06:28 PM I can honestly say that I have never found there to be any problem with 2.35:1 material in either iris mode.
Gremmy, haven't you realized by now that real-world experience is not what we want to hear? We'd all much rather speculate and conjecture on the effects of letterbars on the iris while at the same time get on our high-horse and proclaim that the RS-1 won't have this problem. :D
Jones_Rush 02-14-07, 06:34 PM That's true, the RS1 won't have this annoying problem because there's no DI...
Current C2fine LCD projectors are only about 1000:1 without the dynamic iris, so defeating it is a no go.
Gremmy, haven't you realized by now that real-world experience is not what we want to hear? We'd all much rather speculate and conjecture on the effects of letterbars on the iris while at the same time get on our high-horse and proclaim that the RS-1 won't have this problem. :D
True dat. :D
Hey, even though this isn't an issue on my projector, at least this thread is dealing with a topic I haven't seen discussed in a while. There's a fair bit of ADD around here. :D
HoustonHoyaFan 02-14-07, 08:28 PM I've read that when you project a 2.35:1 letterboxed image, the DI doesn't understand that the top and bottom bars are not part of the movie, and hence the picture becomes in many scenes dimmer than what it should have been (if all the panel was used).
Where did you read that. I own 2 DI pjs and that situation never happens.
Jones_Rush 02-15-07, 02:46 AM Where did you read that. I own 2 DI pjs and that situation never happens.
I've read it in user experience with the pearl.
Gremmy, haven't you realized by now that real-world experience is not what we want to hear? We'd all much rather speculate and conjecture on the effects of letterbars on the iris while at the same time get on our high-horse and proclaim that the RS-1 won't have this problem. :D
As the self-appointed RS1 cheerleader, I'd like to thank you for being the first in this thread to point out that the RS1 won't have this, or any other, DI related problems.... :p
scaesare 02-15-07, 09:09 AM First, TI gives us dancing mirrors and spinning wheels, then Sony/Panny gives us flexing, breathing lenses. Woosh! All this extra motion is making me dizzy.
No thank you. I'll take my video neat. ;)
So you are selling your 3K:1 Qualia for a 15K:1 RS-1?
Or do you prefer the gray bars lumens?
(Brought to you by the "I Actually Like The Qualia But Hate The Disguised Bashing" committee)
J.Mike Ferrara 02-15-07, 11:11 AM So you are selling your 3K:1 Qualia for a 15K:1 RS-1?
Or do you prefer the gray bars lumens?
(Brought to you by the "I Actually Like The Qualia But Hate The Disguised Bashing" committee)
"I don't have no stinkin' gray bars"
New room treatment and careful calibration helps considerably.
And.... I've got lumens to spare! Can the RS-1 crowd say that? :D
|
|