It's official. I see a rainbow-like effect on every plasma. - Page 13

I can promise you that it's not the sub-pixel controller. The effect can be seen even on my HTPC feed which is 1:1 pixel-mapped and would therefore disable the controller.

Isn't the floating black level issue fixed on the newer panels? Or is that Maxent an older unit?

If this effect is caused by rapid eye movement has any one tested to see if viewing distance is one cause. When close you have more eye movement and when way back you can take the whole screen in without much movement.
Just a thought.

Roy

Interestingly, I notice more rainbows from 8' away than 1'.

I noticed yellow trail while watching Sin City on my Sharp LC-40C45U (45D40U equivalent) LCD TV. I also can see it when moving white mouse curser across black background.

which are quite overrated I hear.

who needs "reality" when you have A/C, a recliner, a plasma TV and a remote?

"real life" is for the birds. if it doesn't come in through the wall via a co-ax cable I can't be bothered by it.

Its official this is still a silly thread title.

So is this still a legit problem with the latest Plasmas? I see that this thread has an incredibly thin post count when compared to other threads and especially considering how long the thread has been active... I'm really trying to decide between the Panny 42 600u and the Pioneer 4270... But knowing I'm about as picky and sensitive as anyone can be when it comes to image quality I'm worried that this kind of thing will bug the hell out of me it it's a real problem. I haven't seen it at all in-store, but that doesn't count for much of anything.

Anyone have any info on whether the Pio suffers from this problem? Is it a thing of the past? Thanks in advance...

Yes it's still a problem if you are susceptible to the effect - I see rainbows on all plasma's including my 4 week old Pioneer 50".
It's something that you sort of learn to live with to a large extent - when I first started watching my 3yr old Hitachi (ALIS - they also produce rainbows) it drove me crazy but now I have learnt to ignore most of these effects...
I think it's something to do with persistance of vision - those of us with a very short POV have the problem, most people with normal POV don't see rainbows

Anyone can see the rainbow effect on a plasma screen if you sit right in front of the screen and wave your hand rapidly between the screen and your eyes with open fingers (i.e., the same way you would see 60Hz flicker from a CRT TV). You will see faint red and green fringes around your fingers. I don't know what causes the effect - obviously some sort of interaction between persistence of vision in your eyes and the way the plasma screen displays the picture. Maybe the red/green/blue phosphors light up and decay at a slightly different rate. I don't see anything in normal viewing, and I do see rainbow effect on DLPs.

Do any of you guys who see rainbows on plasma and DLP also get bothered by the new LED red braking lights on cars?

The darned LED lights "jump" in front of my eyes - maybe I really need to see an optometrist!!!

PS - I readily see DLP rainbows even on static DLP images (such as business Powerpoint presentations from a DLP projector), and see plasma rainbows on high-contrast material only (ie rolling credits, dark scenes with white areas, etc).

I see the rainbow effect on my DLP projector (Infocus 4805) as well as my brand new Panasonic 42PX60U plasma. Although, in both cases I usually only see it when it's high contrast material with little color in the picture. So, Sin City is a perfect example being that it's a high contrast black and white movie. The whole time I was watching that movie I was seeing colors flashing all over the place on my DLP projector. With the plasma I do see different color flashes some times (again, typically on black and white high contrast material) but usually it's just flashes of yellow. And if I blink a lot or move my head I'll notice them even more. It's not so bad that I can't live with it, however. Luckily when watching most content I never notice it. It's just certain content (like Sin City) where it really shows up.

BTW, in response to the post above, I do not have an issue with the LED brake lights. They cause no visual issues for me.

Yes, LED brake lights do bother me. You're not crazy.

YES!

I've remarked on this before. Those new brake lights have been driving me insane - they cause strobing trails as my eyes scan the traffic ahead of me! I can't imagine what it would be like if every car used those.

(And yes, I am very sensitive to rainbows on DLP).

I was reading through some papers that described Panasonic's "sub-pixel controller" and how it works. It is essentially the same as:

http://www.grc.com/ctwhat.htm

and will definitely cause color fringing on high contrast scenes but not the color trailing reported here.

As for phosphor decay the average decay time for PDP phosphors is <2ms to 10% luminence so they should not cause any color seperation effects that are different than those on CRTs

This leaves PWM (pulse width modulation) as the culprit. This is the best fit as it is the greatest source of color seperation and will cause color trailing in human perception.

But the point is that CRTs reveal similar effects. It's not different.

Why oh why wasn't this thread nearer the top of the forum more often? I just settled on a plasma (Panasonic TH42-PX60U) and got it this morning. I really wish I had known about this issue before buying.

I hooked up some rabbit ears and pulled in our local PBS in 1080i at ~90% signal strength, made minor adjustments to the picture (contrast way down, color temp to "warm," all image enhancement stuff off). Beautiful picture of the Mormon Tabernacle Choir performing, but it's a historical documentary so the view soon switches to panning across a black and white photo and I see the most horrible RBE I've ever seen. Flashes of color at the smallest movement of my eyes; I wasn't darting my eyes back and forth "looking" for them, just regular viewing. This is much worse than any DLP rainbows I've ever experienced.

I am susceptible to DLP rainbows but have trouble seeing them on the 3-chip displays. I also see green ghosting on RP CRTs and usually need to set PC CRT displays to the highest available refresh rate to avoid the annoyance of flickering.

I tried several other sources and channels and experienced this effect on every one-- of course, high-contrast scenes really bring them out. Even after turning the set off I had some visual "aftershocks." These rainbows sort of "stab" your eyes and it's very disconcerting to relive that effect.

So far the TV is nearly unwatchable for me. I'm still trying to recover from the headache induced by about 20 minutes of viewing (guess I am a glutton for punishment) and that was almost two hours ago! Is there something seriously wrong with my TV? This is the first big TV I've ever had and it looks beautiful other than these horrible flashes. Very sad...

I think we're mixing up some different effects here.

Regarding LEDs: LEDs are a continuous monochromatic light source - they don't strobe or flicker. You could see them leave a strobing trail as your eyes move, but this is because your eyes move in microjumps and the LED brake lights are very bright.

Regarding seeing rainbow effect with 3-chip DLPs: whatever you're seeing, it's not rainbow effect. 3-chip DLP systems don't have rainbow effect.

If you see some sort of horrible rainbow effect on your new plasma TV which makes it nearly unwatchable and gives you a headache, I suggest that you return it. Maybe your eyes are somehow different from the thousands of other owners who don't see this, or perhaps the effect is more psychological than physiological, but it's real to you.

It is my understanding that to dim the LEDs they are strobed. You can see a broken trail if you move past such an LED even without moving your eyes.

I am going to play with settings more tonight and also see if my wife sees this effect (she has pointed out rainbows in DLP and green flashes in RP CRT to me when I haven't noticed). If we can't get it to improve it will be going right back. At that point it will be LCoS or LCD for us, I'm afraid...I really do love the plasma picture other than this.


Good grief, it's not something I was looking for, or even knew about until I viewed the set at home--I was flabbergasted and came here searching for information. Of course in a perceptive sense it is "real to me" (which carries the connotation that I'm imagining this) but there was a fellow just a while back in the thread who posted high-shutter-speed proof of exactly this effect. Just because some people can't perceive it doesn't mean it is nonexistent in a physical (well, as physical as photons can be) sense.

There are dozens of people who see this and who have posted about it on this thread and others. I just wish I'd read about it beforehand.

The led brake lights on cars don't appear to me much brighter at all than the other cars.
Further, I can look at very bright light sources, sweep my eyes past them and experience nothing like I experience with the led lights. (I just tried this, walking right up to a bright light in our hallway - so bright it was hard to stair at directly, and I swept my eyes back and forth. Nothing like the led effect - a strobing, breaking up of the image - happened at all).

Agreed--our individual perceptions are our reality and that's all that matters. I fortunately am not able to perceive the plasma strobing phenomenon. Have you considered direct-view LCD (similar form-factor but perhaps more palatable image generation mechanism) or LCoS-based rear projection?

I also believe they do this to consume less power (and probably money) without diminishing their perceived brightness.

Sounds like the same people did the ergonomic studies...(or didn't).

I got the Panny on a nice "Black Friday" deal so moving to either 42" LCD or 50" SXRD (currently researching Sony vs. JVC LCoS) more than doubles the price. Sucks to be me (or at least to have my eyes/brain/whatever wiring), absent the rainbow flashes this set looks awesome.

Here's a technical paper on how eye tracking interacts with pulse-width-modulated displays (including all DLP, plasma and LCD displays): http://www.poynton.com/papers/Motion...yal/index.html

Their main point is that CRTs escaped this effect because the screens were small, resulting in small angle change for moving objects in the picture, and the persistence of CRT screen phosphors is very short.

It seems to indicate the possibility of seeing rainbow effect on a plasma or LCD (equally, by the way) if:
a) your eye is tracking a moving object in the picture, and your screen is big enough and you are sitting close enough that the movement covers a significantly larger angle of view than watching a convential CRT (or equivalently that you sweep your eyes across a wide angle of view deliberately)
c) the color balance of the object is such that the pulse width ("on" duty state) is significantly different for the 3 colors of pixels

The effect will be worse for objects of high brightness on a dark background (high contrast ratio), and worse if you are sitting close to a big screen.

Possibly some people are seeing more rainbow effect because their screens are set brighter, and they're sitting closer to a bigger screen than others.

(and by psychological factor, I meant the "first dent in the new car" effect - once you see something that really bothers you, it will loom huge in your perception while other people see it but dismiss it as trivial)

Maybe you meant DLP? As sensitive (picky?) as I am I've never seen anything like rainbows on an LCD of any size.


I understand what you're talking about. Now that I've seen this flaw, it will be difficult for me to live with, but I will still try some different lighting and a wider variety of source material. I lived through the "first dent in a new car" several years ago-- still have the car, though! (and the dent)

The pulse-width-modulation effect discussed in the above-referenced paper, which we are assuming is the cause of the rainbow effect on a plasma screen, affects LCDs equally. The paper discusses the effect mainly on LCDs, with a passing mention that it affects plasma screens too. It's far worse on single-chip DLPs because of the sequential color projection.

On an LCD or plasma, what you're seeing from one RGB pixel in a 1/60th second frame interval might be something like:

RRRRR
GGGGGGG
BB

Mostly 3 colors overlapping, but resulting in a fleeting impression of yellowness for part of the interval, and greenness for another part of the interval. That interval *should* be too short for your eye to perceive separately, unless your eye is sweeping rapidly across the display and pauses there on one of its microjumps for less than 1/60th sec.

On a single-chip DLP, the above pixel will be displayed in sequential colors as:

RRRRR GGGGGGGBB

but a little quicker so that it's still 1/60th sec. Now you have much more opportunity to see the separate red, green and blue phases. But what DLPs actually do now is to have multiple repetitions on a faster-spinning color wheel, so that what you actually get in 1/60 sec is something like:

RRRRR GGGGGGGBB RRRRR GGGGGGGBB RRRRR GGGGGGGBB

That reduces, but not eliminates the opportunity to see single colors (rainbow effect).

I cannot stand those LED breaking lights. They are as bright as break lights when on and practically blind me when they break. I feel as if it is dangerous to drive with them on the road. The only things that piss me off more are those super super bright headlights that are like a normal cars brights being on, and those 15 mph speed bumps they keep putting into neighborhoods.

No, again, you have a limited understanding of the technologies involved. 3-chip DLP uses sequential color too. 3-chip DLPs can only create 8 colors at any given instant in time, all others are dithered representations. 1-chip DLPs can only create 3 colors, and because white is not included in that the visibility of this temporal artifacting can be quite significant. 3-chips can create white which really reduces or eliminates the visibility of temporally created color obviously on B&W material, but it is not eliminated.


I don't see where you're lifting that from the article at all, LCDs do not operate by PWM/PDM. The article discusses blurring or latency in the display, not any kind of rainbow artifact visibility. LCDs are not binary, they are not creating color via temporal means like DMDs are at all. They are very much like CRTs in this regard.

Actually no, it doesn't. See http://en.wikipedia.org/wiki/Dlp: "The "Rainbow Effect" is unique to single-chip DLP projectors."

I think you're getting confused about terminology. A 3-chip DLP still uses time-based modulation as a single-chip DLP - it just does it separately for each primary color at the same time, rather than projecting each color sequentially. I agree that it's not quite right to call it pulse-width modulation - but the principle is the same as the other mechanisms for modulating color by the length of time that each color is projected within a small time window.

I've actually been bothered by rainbow-like strobing on 3 chip DLPs. So have a few other people in the projector forum. The effect was exactly like the normal rainbow artifacts, excepting the strobed
high contrast trails didn't break up into more than one color.