|
As this forum is rarely used anymore, we've locked it. Feel free to browse and read. Questions? Please reach out to us directly. Cheers! |
|
 |
|
|
| Author |
Message |
lyd
Joined: 15 Sep 2007 Posts: 390 Location: Lake Mills, Wi
|
| Posted: Sun Jan 13, 2008 10:17 pm Post subject: Low IRE detail lost as APL goes up -- anything to be done? |
|
|
I had been tweaking my brightness setting and the low-end gamma boost on the moome card using a couple of patterns comprised of nothing but sub-blacks to very dark grays. Like 2 to 24 in rgb, I think it is. Everything was looking great, since I got the tip to change that registry setting in PDVD.
Today I threw up a pattern that had a similar range of low values across the horizontal center, maybe about 50% of the height of the screen, and also had high spreads across the top and bottom 25%, intended for checking white clipping.
Looking at this pattern, all my low detail was gone. Black all the way through. If I adjusted brightness and/or gamma correction using this higher APL pattern, when I went back to look at the previous one with only very dark values I was seeing grays all the way down to like 4 RGB.
Is there something I can do to improve the shadow detail at higher average picture levels without destroying my inky blacks at lower ones?
lyd
|
|
| Back to top |
|
 |
garyfritz
Joined: 08 Apr 2006 Posts: 12088 Location: Fort Collins, CO
|
| Posted: Sun Jan 13, 2008 11:11 pm Post subject: |
|
|
That's the weakness of CRTs, especially if you have an AC model. That's called the ANSI contrast ratio -- the maximum difference between full-on and full-off on the same screen. ANSI contrast is tested with a checkerboard pattern, generally 4x4. In theory the black areas should be totally black, but because of light diffusion in the optics, light from the bright areas spills out into the black areas. If those black areas are dark shadow detail, the light from the bright areas can totally swamp them. That's what you're seeing.
Gamma boosting can help this, by raising the brightness level of areas that are greater than 0IRE. But in general you have to choose between "total fade-to-black, can't see my hand in front of my face, but I can't see any shadow detail when there are bright areas on the screen" and "I can see all the shadow detail, but the higher brightness setting means I only fade to gray," or somewhere in between.
|
|
| Back to top |
|
 |
perisoft
Joined: 29 Aug 2007 Posts: 2920 Location: Ithaca, NY
|
| Posted: Sun Jan 13, 2008 11:30 pm Post subject: |
|
|
You could make software that's the anti-iris - boost gamma during high brightness scenes, drop it back during low ones. Heh.
_________________
|
|
| Back to top |
|
 |
lyd
Joined: 15 Sep 2007 Posts: 390 Location: Lake Mills, Wi
|
| Posted: Sun Jan 13, 2008 11:44 pm Post subject: |
|
|
| perisoft wrote: | | You could make software that's the anti-iris - boost gamma during high brightness scenes, drop it back during low ones. Heh. |
Is that part of what the "CLEV-2" business is about in Powerdvd? It doesn't function with HD at this point anyway, so I haven't really worried about it, and anything that claims to dynamically alter the content to produce a "better image" tends to pin my snake-oil meters anyway, but in the blurbs it seems to claim to do something similar to what you are describing.
lyd
_________________ de gustibus non disputandum
|
|
| Back to top |
|
 |
perisoft
Joined: 29 Aug 2007 Posts: 2920 Location: Ithaca, NY
|
| Posted: Mon Jan 14, 2008 12:13 am Post subject: |
|
|
| lyd wrote: | | perisoft wrote: | | You could make software that's the anti-iris - boost gamma during high brightness scenes, drop it back during low ones. Heh. |
Is that part of what the "CLEV-2" business is about in Powerdvd? It doesn't function with HD at this point anyway, so I haven't really worried about it, and anything that claims to dynamically alter the content to produce a "better image" tends to pin my snake-oil meters anyway, but in the blurbs it seems to claim to do something similar to what you are describing.
lyd |
Damned if I know. As far as I can figure from the acronym it might be the second version of a method for levitating carrots.
I tend to think that boosting gamma per-frame dependent on brightness might actually work fairly well; certainly better than an iris which has physical limitations and thus has to act over time, meaning it will lag behind scene content and be obnoxious. If you managed to carefully match the perceived *drop off* in gamma and counteract it, *for each individual frame*, without substantially altering the high IRE content (which, again, irising DOES do the opposite way), I think you'd have a winner. Sheesh, sometimes I impress myself.
You could probably do it in Avisynth inside FFDshow pretty easily; the brute force way would be to throw a big-ass averaging filter on the whole image and alter gamma based on the average scene brightness. Should be pretty straight forward.
From watching a few movies on my AC Barco I'm not sure it's hugely necessary with good gamma in the first place; I suppose it's possible that I'm missing low IRE stuff in half-brightly-lit scenes but I haven't noticed anything like that so far. I should do some checkerboard tests...
I do know that I can see down to 3IRE on the nokia test gray-ramp screen, which has a substantial amount of brightness.
_________________
|
|
| Back to top |
|
 |
ecrabb Forum Moderator
Joined: 13 Mar 2006 Posts: 15909 Location: Utah
TV/Projector: JVC RS40, Epson 5010
|
| Posted: Mon Jan 14, 2008 12:16 am Post subject: |
|
|
Sort of. The CLEV stuff is just dynamic contrast. In low contrast scenes, it expands contrast to "punch" up the image. It would actually exacerbate the effect you're seeing... or, it would at least make the scenes that you don't have a problem with more like the scenes that you do have a problem with.
ANSI... or intra-scene contrast... Deep subject. It's also worth noting this effect (the net result) anyway can happen in both the projector and in your own eyes. In a fully light-controlled environment, during low APL scenes your pupils will be fully dilated and you'll be very sensitive to small changes in the shadows. But, during high APL scenes, your iris constricts the pupil, and shadow detail will be much more difficult (or impossible) to discern. Try looking into a window on the side of a white house on a sunny day.
Anyway, what Perisoft is talking about is dynamic gamma... You could dynamically adjust low-IRE gamma as APL increases to compensate for both the projection system's ANSI contrast AND the eye's light response. Sounds like fun, eh? There are (or have been) people screwing around with that - without a doubt. Applied very subtly, it could be an interesting effect. The net result - if not applied very subtly - would probably look somewhat "weird" or "artificial" - kind of like HDR (high dynamic range) photography - which is somewhat related.
SC
|
|
| Back to top |
|
 |
perisoft
Joined: 29 Aug 2007 Posts: 2920 Location: Ithaca, NY
|
| Posted: Mon Jan 14, 2008 1:04 am Post subject: |
|
|
| ecrabb wrote: | Sort of. The CLEV stuff is just dynamic contrast. In low contrast scenes, it expands contrast to "punch" up the image. It would actually exacerbate the effect you're seeing... or, it would at least make the scenes that you don't have a problem with more like the scenes that you do have a problem with.
ANSI... or intra-scene contrast... Deep subject. It's also worth noting this effect (the net result) anyway can happen in both the projector and in your own eyes. In a fully light-controlled environment, during low APL scenes your pupils will be fully dilated and you'll be very sensitive to small changes in the shadows. But, during high APL scenes, your iris constricts the pupil, and shadow detail will be much more difficult (or impossible) to discern. Try looking into a window on the side of a white house on a sunny day.
Anyway, what Perisoft is talking about is dynamic gamma... You could dynamically adjust low-IRE gamma as APL increases to compensate for both the projection system's ANSI contrast AND the eye's light response. Sounds like fun, eh? There are (or have been) people screwing around with that - without a doubt. Applied very subtly, it could be an interesting effect. The net result - if not applied very subtly - would probably look somewhat "weird" or "artificial" - kind of like HDR (high dynamic range) photography - which is somewhat related.
SC |
Hmm, HDR processing of filmed material could be pretty cool now that you mention it. But yeah, HDR is fantastic if used in a way that you can't tell it's HDR. Most of the HDR stuff on flickr, though, consists of deep blue skies with glowing trees...
It's kind of like when people found the photoshop lens flare filter. Just because you can doesn't mean you should!
Anyway, yeah, I think you COULD do a good job of dynamic gamma. Your eyes screw with the images too, don't forget; I think the reason CRT projectors are so effective for movie watching is that your eyes don't tend to be very good at detecting low brightness detail when other bits of a scene are high brightness *either*. But they're damned good at detecting overall brightness vs. a true black. CRT projectors are bad in the same way your eyes are; digitals are the opposite.
_________________
|
|
| Back to top |
|
 |
lyd
Joined: 15 Sep 2007 Posts: 390 Location: Lake Mills, Wi
|
| Posted: Mon Jan 14, 2008 1:06 am Post subject: |
|
|
| ecrabb wrote: | | It's also worth noting this effect (the net result) anyway can happen in both the projector and in your own eyes. In a fully light-controlled environment, during low APL scenes your pupils will be fully dilated and you'll be very sensitive to small changes in the shadows. But, during high APL scenes, your iris constricts the pupil, and shadow detail will be much more difficult (or impossible) to discern. Try looking into a window on the side of a white house on a sunny day. |
This is the first thing that occurred to me when I read Gary's response. I'm getting a little older, and I can't see into the shadows in high contrast situations in real life as well as I used to. I figured based on that that I would optimize for FFB and good detail at low average APL, because losing that shadow detail in higher contrast scenes won't seem terribly unnatural anyway.
I certainly wasn't really missing it, before I noticed the issue in that pattern. After getting the moome card and making that registry change for PDVD, I've been really happy overall.
lyd
_________________ de gustibus non disputandum
|
|
| Back to top |
|
 |
lyd
Joined: 15 Sep 2007 Posts: 390 Location: Lake Mills, Wi
|
| Posted: Mon Jan 14, 2008 1:10 am Post subject: |
|
|
| perisoft wrote: | | I do know that I can see down to 3IRE on the nokia test gray-ramp screen, which has a substantial amount of brightness. |
Isn't that a bad thing? Don't you want a 7.5 IRE black pedestal?
lyd
_________________ de gustibus non disputandum
|
|
| Back to top |
|
 |
perisoft
Joined: 29 Aug 2007 Posts: 2920 Location: Ithaca, NY
|
| Posted: Mon Jan 14, 2008 1:54 am Post subject: |
|
|
| lyd wrote: | | perisoft wrote: | | I do know that I can see down to 3IRE on the nokia test gray-ramp screen, which has a substantial amount of brightness. |
Isn't that a bad thing? Don't you want a 7.5 IRE black pedestal?
lyd |
I'm no expert - maybe so. I heard 3 somewhere, so 3 it was. I'll have to try setting to 7.5 and see how it looks. Not that the thing is really calibrated well anyway!
_________________
|
|
| Back to top |
|
 |
MikeEby
Joined: 24 Jun 2007 Posts: 5237 Location: Osceola, Indiana
|
| Posted: Mon Jan 14, 2008 3:33 am Post subject: |
|
|
| lyd wrote: | After getting the moome card and making that registry change for PDVD, I've been really happy overall.
lyd |
Just curious...Why are you even using a Moome card with an HTPC? I run RBGHV direct to my projector. When I was using the Moome card I saw things just like you are talking about. My Moome is a Gen 1 XG internal that has known black crush issues. I don't know about the Marquee card you are using maybe its ok.
Mike
_________________ Doing HD since the last century!
|
|
| Back to top |
|
 |
jask
Joined: 17 Mar 2006 Posts: 10187 Location: kamloops BC
|
| Posted: Mon Jan 14, 2008 4:30 am Post subject: |
|
|
|
I was thinking the same thing! I would guess that you will be able to set up a better image from the HTPC if you pull the Moome out of the chain and redo everything from a HD test pattern.
|
|
| Back to top |
|
 |
lyd
Joined: 15 Sep 2007 Posts: 390 Location: Lake Mills, Wi
|
| Posted: Mon Jan 14, 2008 11:22 am Post subject: |
|
|
| MikeEby wrote: | | Why are you even using a Moome card with an HTPC? I run RBGHV direct to my projector. |
Actually, I was asking this very same question in my first posts here, and maintaining that with a PC to manage gamma curves there was no value to a moome card. I wound up changing my mind, for several reasons.
Even after I decided to get the VIM-HD, I was still planning on leaving the gamma circuit disabled, or even removing the board, and tweaking the curve in my video drivers, but after playing with it I was quite happy with the results so I left well enough alone.
When I was using RGBHV and tweaking gamma in the driver the shadow detail was no worse, but it was no better either. And since using the video drivers to make those adjustments is kind of an inaccurate (due to the interface -- you just manually drag a tiny little representation of the curve around) PITA, I am pretty content with it the way it is, at least for now.
Also, while better analog cables might have helped this (I was using good quad-RG6, but with cheapo F to BNC adapters on the ends) input through the moome card is noticeably sharper than I was getting with RGBHV.
Feel free to apply a grain of salt if you wish; I've got this $400+ thing I need to rationalize having purchased, after all, but I'm happy.
lyd
_________________ de gustibus non disputandum
|
|
| Back to top |
|
 |
MikeEby
Joined: 24 Jun 2007 Posts: 5237 Location: Osceola, Indiana
|
| Posted: Mon Jan 14, 2008 3:21 pm Post subject: |
|
|
| lyd wrote: |
Feel free to apply a grain of salt if you wish; I've got this $400+ thing I need to rationalize having purchased, after all, but I'm happy.
lyd |
Been there, done that! My Moome was only $225 so it wasn’t as hard to swallow. And to be fair to Moome the component input is fine, since my XG 1351 does not have that type of input it will be used with cable boxes or satellite receivers so it’s not a total loss.
Mike
_________________ Doing HD since the last century!
|
|
| Back to top |
|
 |
|
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum You cannot attach files in this forum You can download files in this forum
|
Forum powered by phpBB © phpBB Group
|
|