[jtiner]

I've been searching threads discussing older panels that exhibit problems with false contouring, but I'm wondering what experiences users are having now with newer panels (2008/2009 Panasonic/Samsung/LG, say).

I've seen this term used to describe different things, but I'm referring specifically to the problem of displaying what should be a smooth gradation in color/brightness (such as the glow around a sunset or the glow around a lamp in a dark scene) displayed as discreet bands/steps with sharply defined boundaries between each step.

I haven't seen this as much as I used to, and I don't see it mentioned very frequently when searching through threads, but a few people do seem to be experiencing it on new plasma and LCD panels. Has the current processing in panels reached a point where this isn't a problem now?

 

[Gary McCoy]

The short answer is YES, most HDTVs utilize at least 10-bit engines in the video processors inside the sets.


However the problem you are describing (the real term for which is "posterization") can still be seen with certain older console video games or older computer video boards. Anywhere one type of electronics passes information to another, less than optimal design compromises may have been made for the sake of faster performance, which truncate the least significant bits of pixel colors and produce the effect you are describing.


The difference is that nowadys this is not the fault of the display, but the video source component, if that component was originally designed for an analog video interface such as VGA or component video, and is now being used on a modern digital display with HDMI or DVI. (Example - an old PS2 video game running in PS2 emulation on a newer PS3.)


But as long as everything you are using is HDMI connected, no problems will be seen.

 

[jtiner]

I've been noticing this effect this evening for off-air sources on the small set in the kitchen (Samsung LN22B650).

One CBS affiliate is upconverting everything local, so I see quite a few processing artifacts there, as well as false contouring/posterization.

I've also noticed it on the other CBS affiliate (which is HD for local and network) and the NBC affiliate. The effect was apparent on several commercials during the local news.......... I'll have to watch for it during network feeds in the evening. Perhaps it's an artifact of the local commercial playout systems being used. I would hope it's not the panel.

 

[irfan]

even if its a 10bit display, would that matter? isnt much of the problem in the signal itself?

 

[jtiner]



Quote:
Originally Posted by irfan /forum/post/18164430


even if its a 10bit display, would that matter? isnt much of the problem in the signal itself?

It can be the source, as I've seen material exhibit this problem on my CRT, but I think it's more likely the panel.

I was recently looking at the Samsung LN46B650 LCD and the PN50B650 plasma side by side at a retailer. The source material was horrible, but there was a short sequence at one point with a color/brightnes fade (like a sunset) and there was noticeable banding on the LCD but not on the plasma. That concerned me, since I was considering the LCD.

And that's really the problem; you can't really compare in-store; it'd be great if you could play out your own material and test patterns........

 

[Gary McCoy]

I have seen examples in stores where they were attempting to "compare" video displays by hooking them to the "same" video source.


However the basic digital copy protection scheme defeats this, there can be only one signal "source" and one signal "sink" when a transfer is occurring over a digital HDMI interface. Thus the second display gets attached to the analog component video or even the S-video output of the Blu-Ray player or scaling DVD player. The analog outputs are displaying 480i (S-Video) or 480p (component video) versions of the HD signal which simply cannot be compared with the 1080p original, both posterization and resolution scaling artifacts exist in the analog versions.


Keep this in mind when assessing displays in stores - you CAN'T show two side-by-side displays with the same Blu-Ray image because of copy protection. Therefore if you actually see this you know they are using some other source that is NOT copy-protected, usually a DVR downloaded from the Internet or a satellite receiver. Even then, they may be sharing this source over both HDMI and analog outputs from a modern HDMI A/V receiver.

 

[HogPilot]

My Lumagen Radiance lets me pipe the same 1080p signal via HDMI to two separate displays simultaneously, and there are HDMI splitters out there that do this as well. Receivers are becoming more common which have multiple HDMI outs that will let you do the exact same thing.


That being said, I'd still check to make sure the same signal is being piped over HDMI and that one of the displays isn't being fed 1080i over component. If it was fed with S-Video or composite, you'd see a major degradation in PQ that would be immediately evident.

 

[jtiner]

Regarding the source for multiple displays at retailers, I wasn't able to discern what they were doing for distribution (the displays were mounted to the wall with items stacked below them on the floor). The LCD exhibited the banding while the plasma exhibited lag (clearly behind the other displays on quick edits). It's hard to trust what you see at retail locations.

I saw some kind of server appliance at Sears but it was only storing content and was connected only by ethernet. I didn't see the media appliance it fed.

I have seen component video distribution in use at Sams Club, and coaxial distribution at Best Buy. My guess on the coax is a QAM modulator fed by analog component video. That was some time ago.