It's official. I see a rainbow-like effect on every plasma. - Page 10 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #271 of 883 Old 05-16-2006, 01:12 PM
Senior Member
 
mule65's Avatar
 
Join Date: Apr 2006
Location: Minneapolis
Posts: 435
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by Felgar View Post

So am I. Isn't that a live video of an actual plasma screen showing the game?

Here's the exact post

As far as I can tell, he's watching some video, freezing it, and seeing these trailing yellow blurs while paused. Maybe I'm a confused noo-B but if you can see them when paused then it's in the source material.
mule65 is offline  
Sponsored Links
Advertisement
 
post #272 of 883 Old 05-16-2006, 01:22 PM
 
NathanC's Avatar
 
Join Date: Nov 2005
Posts: 469
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by mule65 View Post

Here's the exact post

As far as I can tell, he's watching some video, freezing it, and seeing these trailing yellow blurs while paused. Maybe I'm a confused noo-B but if you can see them when paused then it's in the source material.

But we're not pausing the game, we're pausing a video of the game on a plasma. Therefore it has nothing to do with the source game, it has to do with the plasma showing the game. Can you not understand?
NathanC is offline  
post #273 of 883 Old 05-16-2006, 01:23 PM
AVS Special Member
 
Felgar's Avatar
 
Join Date: Jan 2002
Location: Calgary, Canada
Posts: 2,502
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I'm referring to the same post. He says "I came across a video of Rainbow Six being filmed off a plasma" which means that the video is actually of a plasma screen showing the game. The camera is catching the same artifact off the screen that their eyes are catching. If somone stood in front of the camera he would block the screen.

In this case the effect 'could' be the camera or it could be the screen, but being able to pause it has no bearing on whether the issue is source-material related or not in this case, because the original source material is the TV showign the game, not the game itself.
Felgar is offline  
post #274 of 883 Old 05-16-2006, 01:27 PM
 
NathanC's Avatar
 
Join Date: Nov 2005
Posts: 469
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Here's a video i just took. It's not good quality because i forgot to turn a light on and i'm using a digital camera, but even then you can see how the edges of the tree and mountain kind of glow yellow when panning, especially to the left. http://www.supload.com/vid/100_0160/362696214/mov/ It's a big file, sorry, it's only about 8 seconds long, but the yellow shown is exactly what i see.
NathanC is offline  
post #275 of 883 Old 05-16-2006, 01:30 PM
Senior Member
 
mule65's Avatar
 
Join Date: Apr 2006
Location: Minneapolis
Posts: 435
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by Felgar View Post

In this case the effect 'could' be the camera or it could be the screen, but being able to pause it has no bearing on whether the issue is source-material related or not in this case, because the original source material is the TV showign the game, not the game itself.

I see. Still, a good test would be to find a suspect scene on a DVD and slo-mo or pause it to verify its not in the source. Has this been done? I'm not disputing any claims. Obviously, some people can process visual information faster than others. Race car drivers can read the label on a record album while spinning 78 rpm.
mule65 is offline  
post #276 of 883 Old 05-16-2006, 01:36 PM - Thread Starter
AVS Special Member
 
Yoda1's Avatar
 
Join Date: May 2005
Posts: 1,123
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by NathanC View Post

Here's a video i just took. It's not good quality because i forgot to turn a light on and i'm using a digital camera, but even then you can see how the edges of the tree and mountain kind of glow yellow when panning, especially to the left. http://www.supload.com/vid/100_0160/362696214/mov/ It's a big file, sorry, it's only about 8 seconds long, but the yellow shown is exactly what i see.


I'll be honest, I can't see the yellow flashing/pulsing on that video at all.

I do notice a lot of screen tear, though!

What is up with every 360 game having screen tear?
Yoda1 is offline  
post #277 of 883 Old 05-16-2006, 01:47 PM
AVS Addicted Member
 
Ken Ross's Avatar
 
Join Date: Nov 2000
Location: N.Y.
Posts: 23,765
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 834 Post(s)
Liked: 907
Quote:
Originally Posted by Felgar View Post

My understanding is that the video being referenced is actually of a plasma screen showing the game... it's not source game material.

Felgar, I think his point is that if the issue shows in a freeze frame, it's obviously not related to a rise/decay issue of phosphors...it's there in the source. I would tend to agree if it shows in a freeze frame of the actual source. But this is a video of the screen, so it's not really a freeze of the source.
Ken Ross is offline  
post #278 of 883 Old 05-16-2006, 01:52 PM
AVS Special Member
 
1920x1080's Avatar
 
Join Date: Oct 2004
Location: SoCal
Posts: 1,268
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Yoda1 View Post

Well, like many people this week, I've been all about E3 and just this morning I came across a video of Rainbow Six being filmed off a plasma. Now, I know many of you will continue to say that I'm crazy, seeing things, need my eyes fixed, etc., but I think this video shows some of the things that dalandis and I are complaining about.

If you have the time, download this video -- http://www.gametrailers.com/player.p...e=mov&id=10717

About midway through the video, once the SWAT team gets inside the Casino, you see them taking cover and running past big white flourescent lights in the background. If you look closely as the camera pans past those big lights from left and right, you can see yellow blurring or trailing. It also happens on many other scenes but it's a bit more subtle. Also, if you freeze frame some of these scenes you can clearly make out a yellow outling around the player models and other objects onscreen. If I could, I would take stills and post them here, but i have no idea how to do something like that. If someone here sees what I'm talking about, and could take some screen caps of the effect as it's happening, that would be awesome.

P.S. - The more I watch this video, the more clearly I see this. How anyone with a discerning eye couldn't see what I'm talking about in this video, is absolutely beyond me.

Here's a screen shot that clearly shows the effect. Yoda1, I couldn't find any reference to how the video was created on that site. Was it something you saw being done at E3?
LL

"In theory, there is no difference between theory and practice. But in practice, there is." - Yogi Berra
1920x1080 is offline  
post #279 of 883 Old 05-16-2006, 02:04 PM
AVS Special Member
 
conan48's Avatar
 
Join Date: Jun 2003
Location: Igloo
Posts: 2,159
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 48 Post(s)
Liked: 155
I noticed it in the file. You can see a yellow smear where the trees are. If you can upload a higher quality version I'm sure it would be easier for people to see. Also, watching the video in slow motion, you can see that there is an afterimage when the camera pans. The image is leaving a ghost then when the ghost image catches up it leaves the trail.

I think this may be caused by image retention. Sometimes when you quickly switch channels and it's on a dark scene you can see IR from the pervious channel. This may be a form of IR to to milisecond and nothing to do with phosphor decay. Do you guys remember when on a CRT there was a solid black background and a bright object moved across, you would see a trail. Now the CRT trail was alot longer in duration but it may have something to do with it.(only noticed it on black with very bright objects) Also I have seen green trails on my Sony RPTV on COD2 (sold TV, didn't have time to test other games) All these TVs can potentially suffer from burn in or Image retention. Now we know that direct view Crts have the least likely to burn, and trails are least likely to be noticed (some people do see trails on crt). CRT RPTV is more likely to burn but the trails are very minimal. Plasma is the most likely to get burn and has the worst trails.

LCD, LCOS, DLP have no trails whatsoever and they have no chance of getting burn it or image retention. Yes DLP has rainbows, but thats because of the color wheel.

I think I may be on to something.
conan48 is offline  
post #280 of 883 Old 05-16-2006, 02:33 PM - Thread Starter
AVS Special Member
 
Yoda1's Avatar
 
Join Date: May 2005
Posts: 1,123
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by 1920x1080 View Post

Here's a screen shot that clearly shows the effect. Yoda1, I couldn't find any reference to how the video was created on that site. Was it something you saw being done at E3?

No, I wasn't there. I don't know what kind of camera they used, I just stumbled upon this video through GameTrailers cause I was interested in seeing footage of the game. It was about midway through that I started seeing yellow/pulsing flashes and I knew right then and there that this game was being demoed on a plasma.
Yoda1 is offline  
post #281 of 883 Old 05-16-2006, 03:31 PM
 
ChrisWiggles's Avatar
 
Join Date: Nov 2002
Location: Seattle
Posts: 20,730
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Guys, I just wanted to pop in and clarify some of the confusion about capturing this artifact.

There are two distinct *different* ways this arises due to phosphor decay times.

First, what is captured in the video and the still shots is what I would characterize more as phosphor lag and trails. This is seen during rapid motion, and as a bright portion of an image is driven very bright, and then it passes by, there is a decay time (especially with green phosphor) that does not immediately return to black. Thus, you get a kind of greenish decay trail with motion most easily visible during high contrast scenes with fast motion. This is also seen on many CRTs, again depending on phosphor decay time.

Second, is that because the phosphors decay at different rates and there is refreshing going on due to the PDM/PWM nature of how plasmas work, there is also decays going on constantly within any image, and this includes still images. This is what is most bothersome because it's a fatiguing and distracting "rainbow"-like artifact that sort of jumps out at you as you move your eyes about. You can isolate this issue as distinct from lagging trails by pausing an image. If you hold your eyes still, or take a screenshot with the camera still, you will not see anything abnormal, and will just see the image as intended. However, if you move your eyes about rapidly, or take a screenshot while moving a camera rapidly, you may see these decay/refresh type artifacts that go into making the image "jump" out.

This is not the same as sequential color-rainbows on a 1-chip DLP, but what you see is not too dissimilar. On a color-wheel system, if you have a white object, it's actually R,G,B flashing very fast that you see as white. If you move your eyes rapidly, you can see this separate temporally into the RGB components that are flashing sequentially to create "white," and that is a DLP rainbow. Here, there is no sequential color going on, in essence white is created and is actually white (it's spatially converged to white essentially), but that white also decays and when it does so it doesn't decay from white to gray to black because the phosphors decay at different rates. So it's white, and then it starts to decay but the green decays slower so it changes between white and a decaying greenish yellowish color until the phosphor are excited again to create white. Again, if you dash your eyes around or take a screenshot while rapidly moving the camera, you may be able to sense this occurring.

In addition to this, there is also the fact that plasmas operate with PDM/PWM principles in the binary cells to excite the phosphors. This may also be a contributing factor for this temporal color "rainbow" artifact.

http://www.avsforum.com/avs-vb/showt...+rainbows+cell

In that thread these kinds of things were discussed, and I do agree with differentiating these "rainbow" artifacts from DLP sequential-color rainbows which actually are R,G,B temporally displayed. These artifacts on Plasmas and on CRTs are different than that, yet they have a similar kind of effect of revealing colors that are not part of the image that are temporal in nature. High-contrast black and white images will bring these problems out the worst, again not unlike DLP rainbows, but the nature of the artifacting is still distinct. If there were another term that differentiates this artifact on plasmas an CRTs from colorwheel-rainbows I would use it, but I have not encountered an appropriate term for this and I think inventing a new one might increase confusion. I just want to make clear that this is perceptually similar to DLP rainbows, but is not the same thing, and I don't want the use of the term "rainbows" to be interpreted in a way such that I seem to be claiming that plasmas or CRTs operate with sequential color and that these artifacts are identical, because they are not.
ChrisWiggles is offline  
post #282 of 883 Old 05-16-2006, 04:25 PM
 
NathanC's Avatar
 
Join Date: Nov 2005
Posts: 469
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by ChrisWiggles View Post

Guys, I just wanted to pop in and clarify some of the confusion about capturing this artifact.

Excellent post. Thanks. Clarifies everything nicely.
NathanC is offline  
post #283 of 883 Old 05-16-2006, 05:29 PM
AVS Special Member
 
Zues's Avatar
 
Join Date: Jul 2004
Posts: 3,835
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
The plasma rainbows are similar to dlp effect on bright scenes or dark scenes with white in them. Crt rainbows? Lets just say the plasma rainbows are more problematic than any crt, rear or direc-view ive had. This effect is not on crt imo.
Zues is offline  
post #284 of 883 Old 05-16-2006, 05:55 PM
Member
 
74Capri's Avatar
 
Join Date: Dec 2005
Posts: 136
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
You will find no rainbow with the dell.

Ain't no pot of gold, owning this set.
74Capri is offline  
post #285 of 883 Old 05-16-2006, 10:20 PM
Member
 
neomoz's Avatar
 
Join Date: Aug 2002
Posts: 177
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I've owned my plasma for 3 1/2 years now. My first one was an alis panel which was replaced by sony when it died for a non alis panel. Both of these panels exhibit that yellow(most of the time) motion trail on certain colours. at times it very noticable or very slight. Like someone else said it appears to be related to phosphor decay and there isn't much you can do about it. Just a limitation to the technology.
neomoz is offline  
post #286 of 883 Old 05-17-2006, 07:07 AM
AVS Special Member
 
conan48's Avatar
 
Join Date: Jun 2003
Location: Igloo
Posts: 2,159
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 48 Post(s)
Liked: 155
So it seems that people do see the problem but have learned to deal with it. I guess it's not a deal breaker. I'll ask anyway. How many people see trails and but still stick to plasma vs. how many people have returned there sets for something else.
conan48 is offline  
post #287 of 883 Old 05-17-2006, 08:10 AM
AVS Special Member
 
scherer326's Avatar
 
Join Date: Jun 2005
Posts: 1,031
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I have had my panasonic th-42px60u plasma since march of this year. The HD picture quality in my opinion is great on this tv. But for some reason, starting in late april to present, the "rainbow-;ike" effect seems to be getting worse. Is this just my eyes deceiving me. I have over 100 hrs on the tv.
scherer326 is offline  
post #288 of 883 Old 05-17-2006, 08:34 AM
AVS Special Member
 
Felgar's Avatar
 
Join Date: Jan 2002
Location: Calgary, Canada
Posts: 2,502
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by ChrisWiggles View Post

First, what is captured in the video and the still shots is what I would characterize more as phosphor lag and trails. This is seen during rapid motion, and as a bright portion of an image is driven very bright, and then it passes by, there is a decay time (especially with green phosphor) that does not immediately return to black.

I do not believe this is the cause of the 'green trail' effect. The phosphors just don't lag that much; PWM could not work if the phosphours decayed that slowly. At 60 Hz, each frame is about 17 ms. Compare that with the response time necessary to accomplish intensity variation with PWM... Honestly I'm not sure what the number would need to be but it would need to be a mere fraction of that 17 MS. Can anyone find documentation on that?

Quote:
Originally Posted by ChrisWiggles View Post

Here, there is no sequential color going on, in essence white is created and is actually white (it's spatially converged to white essentially), but that white also decays and when it does so it doesn't decay from white to gray to black because the phosphors decay at different rates. So it's white, and then it starts to decay but the green decays slower so it changes between white and a decaying greenish yellowish color until the phosphor are excited again to create white.

I also question this description. If the image is paused then the white is not decaying because a white pixel stays white. And if it's not paused, then the effect is actually what you described earlier.

And also, if the 'rainbow' you're seeing when paused is a result of being able to discern PWM on a single-pixel level, then it doesn't make sense to me that you see it worst on a white pixel. A white pixel is really just 'all on', as the plasma would be keeping the pixel on 100% of the time in order to make its reference 'white' color. Like in a drawing program, RGB values is 255, 255, 255 for white meaning 'full' values for each color. PWM would come into play for say a 50% gray which would be RGB 130, 130, 130. For a grey pixel I could possibly accept PWM as a cause, because then each sub-pixel would be 50% on, 50% off. In fact a dark grey at say 20% on time should make this effect most noticeable, if indeed the cause you attribute to the effect is what you think it is.

So IMO this leaves us with 2 questions I think still need to be answered. 1) What is the cause of motion-related color trails, and 2) is seeing a 'rainbow' effect while darting your eyes an actual display flaw or possibly just an effect of the physiology of human eyes.

Honestly I'm not trying to be difficult. I just feel that quite a few might be on the wrong track, and until we can come up with an explanation that makes sense, we need to continue to look for a cause of these issues. Until we know the true cause, finding a solution would be near-impossible.
Felgar is offline  
post #289 of 883 Old 05-17-2006, 08:50 AM
Ein
Advanced Member
 
Ein's Avatar
 
Join Date: Jul 2003
Location: Wisconsin
Posts: 813
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I only see the greenish comet tail when the end credits are in white and scrolling very fast. It's mostly in the PBS HD programs


Why called it "Rainbow", when you only see one color?

Should just called it the comet tail effect.
Ein is offline  
post #290 of 883 Old 05-17-2006, 09:06 AM
AVS Special Member
 
scherer326's Avatar
 
Join Date: Jun 2005
Posts: 1,031
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
sorry will now call it the "green trail effect"
scherer326 is offline  
post #291 of 883 Old 05-17-2006, 09:47 AM
AVS Special Member
 
xrox's Avatar
 
Join Date: Feb 2003
Posts: 3,169
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 50
Quote:
Originally Posted by Felgar View Post

I do not believe this is the cause of the 'green trail' effect. The phosphors just don't lag that much; PWM could not work if the phosphours decayed that slowly. At 60 Hz, each frame is about 17 ms. Compare that with the response time necessary to accomplish intensity variation with PWM... Honestly I'm not sure what the number would need to be but it would need to be a mere fraction of that 17 MS. Can anyone find documentation on that?


I also question this description. If the image is paused then the white is not decaying because a white pixel stays white. And if it's not paused, then the effect is actually what you described earlier.

And also, if the 'rainbow' you're seeing when paused is a result of being able to discern PWM on a single-pixel level, then it doesn't make sense to me that you see it worst on a white pixel. A white pixel is really just 'all on', as the plasma would be keeping the pixel on 100% of the time in order to make its reference 'white' color. Like in a drawing program, RGB values is 255, 255, 255 for white meaning 'full' values for each color. PWM would come into play for say a 50% gray which would be RGB 130, 130, 130. For a grey pixel I could possibly accept PWM as a cause, because then each sub-pixel would be 50% on, 50% off. In fact a dark grey at say 20% on time should make this effect most noticeable, if indeed the cause you attribute to the effect is what you think it is.

So IMO this leaves us with 2 questions I think still need to be answered. 1) What is the cause of motion-related color trails, and 2) is seeing a 'rainbow' effect while darting your eyes an actual display flaw or possibly just an effect of the physiology of human eyes.

Honestly I'm not trying to be difficult. I just feel that quite a few might be on the wrong track, and until we can come up with an explanation that makes sense, we need to continue to look for a cause of these issues. Until we know the true cause, finding a solution would be near-impossible.

8-13 pulses per frame with a combined time of about 4-6ms means the shortest individual pulse would be well below 1ms in duration.

Note: Plasmas do not use pure Pulse Width Modulation (PWM). This would mean they would need 256 bit processing. Since they are limited to 8-13bits per sub-pixel they must use temporally seperated PWM, where binary combinations are used to obtain the desired luminence.

1, 2, 4, 8, 16, 32, 64, 128 in time domain, each number represents pulse duration.

For example in order to obtain a luminence of 128:

-, - , - , - , - , - , - , 128 ( cell is off until very end of frame period)

in order to obtain a luninence of 5

1, - , 4 , - , - , - , - (cell on near begining of frame period)

Over thinking, over analyzing separates the body from the mind
xrox is offline  
post #292 of 883 Old 05-17-2006, 10:22 AM
AVS Special Member
 
Felgar's Avatar
 
Join Date: Jan 2002
Location: Calgary, Canada
Posts: 2,502
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
8-13 pulses... Where do you get that number xrox? And wouldn't the combined time be the full frame on a 100% white pixel? (17 ms?)
Felgar is offline  
post #293 of 883 Old 05-17-2006, 10:33 AM
AVS Special Member
 
xrox's Avatar
 
Join Date: Feb 2003
Posts: 3,169
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 50
Quote:
Originally Posted by Felgar View Post

8-13 pulses... Where do you get that number xrox? And wouldn't the combined time be the full frame on a 100% white pixel? (17 ms?)

Plasmas (depending on which model) will have 8-13 bit processing meaning they will have 8-13 subfields per frame per subpixel where each subfield is a unit in the Binary Code PWM. Furthermore, each subfield will have a write pulse, multiple sustain pulses, and an erase pulse. This means most of the frame time is used up writing and erasing and only 4-6ms is left over for sustaining (ie - emitting light).

Over thinking, over analyzing separates the body from the mind
xrox is offline  
post #294 of 883 Old 05-17-2006, 11:22 AM
 
ChrisWiggles's Avatar
 
Join Date: Nov 2002
Location: Seattle
Posts: 20,730
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by Felgar View Post

I do not believe this is the cause of the 'green trail' effect. The phosphors just don't lag that much; PWM could not work if the phosphours decayed that slowly. At 60 Hz, each frame is about 17 ms. Compare that with the response time necessary to accomplish intensity variation with PWM... Honestly I'm not sure what the number would need to be but it would need to be a mere fraction of that 17 MS. Can anyone find documentation on that?

This is something that I thought about, and my familiarity directly with plasmas is not as great as my familiarity with projection displays, but the plasma decay time still remains an issue, regardlesss of whether the excitation of the plasma is PDM/PWM or not. That is to say, the cells operate this way to excite the phosphor, and the cells have a much higher "speed" than the phosphor, but the phosphor still has its decay time regardless. The other thing that I have not adequately considered because I am not as familiar with it now, is how fast the cells are, because older plasmas did have problems with image retention.

Quote:


I also question this description. If the image is paused then the white is not decaying because a white pixel stays white. And if it's not paused, then the effect is actually what you described earlier.

Yes, but if you pause the image, you can still see the rainbow-like artifacts, very similarly to as if you pause the image on a DLP or a CRT that's refreshing more slowly etc.

Quote:


So IMO this leaves us with 2 questions I think still need to be answered. 1) What is the cause of motion-related color trails, and 2) is seeing a 'rainbow' effect while darting your eyes an actual display flaw or possibly just an effect of the physiology of human eyes.

Well, again the trails are the simplest to explain because it is just decay time, though it may be aggravated by the decay times of the cells though I do not know if that is a problem anymore. For 2) it isn't a physiological artifact because it is capturable on camera and is not visible on some other displays or othe high-contrast images and things like that. Though, it is in some sense physiological because some people see this more than others.
ChrisWiggles is offline  
post #295 of 883 Old 05-17-2006, 11:36 AM
Member
 
kmramos's Avatar
 
Join Date: May 2001
Posts: 99
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I've had my Fujitsu plasma for about 7 months now. I used to see the yellowish green flashes a lot, but now I hardly notice them. When you first see it, your eyes will be darting across the screen constantly, and this makes it worse. In time, your eyes will stop darting so much, and the flashes will mostly stop. I think being content with your setting also helps, as you will start watching the content and stop watching your display.
kmramos is offline  
post #296 of 883 Old 05-17-2006, 01:13 PM
AVS Special Member
 
Felgar's Avatar
 
Join Date: Jan 2002
Location: Calgary, Canada
Posts: 2,502
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by xrox View Post

Plasmas (depending on which model) will have 8-13 bit processing meaning they will have 8-13 subfields per frame per subpixel where each subfield is a unit in the Binary Code PWM. Furthermore, each subfield will have a write pulse, multiple sustain pulses, and an erase pulse. This means most of the frame time is used up writing and erasing and only 4-6ms is left over for sustaining (ie - emitting light).

I see what you're saying about the 13 subfields, and after a lot of thought and deliberation I agree that it makes sense. But, how can you be sure that each picture field is not sub-divided into many groups? Like in your example with value of 128 in an 8-bit system, that would correspond to about half time on. But the cell itself could pulse many times per frame rather than just once per frame. The signal processing would not change in doing that, because it would be same sub-pixel value, just repeated in much smaller time slices. It would all depend on how quickly the sub-pixel can respond.

Incidently the more I think about it, the more I'm lead to believe that any rainbow effect is the direct result of PWM and not phosphour decay. If the on-off periods are not in sync between sub-pixels, the result when darting eyes and picking up just a fraction of the image would be the rainbow because the image could be caught with 1 sub-pixel on and the other off.

Incidently this artifact would appear on a 3-chip DLP for exactly the same reason. Didn't someone mention earlier that he could see it on a 3-chip DLP? To me this would verify that the effect is solely a result of PWM and not phosphour decay at all, as DLP doesn't experience decay and has on-off cylce times of microseconds.
Felgar is offline  
post #297 of 883 Old 05-17-2006, 02:03 PM
 
ChrisWiggles's Avatar
 
Join Date: Nov 2002
Location: Seattle
Posts: 20,730
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Incidently this artifact would appear on a 3-chip DLP for exactly the same reason. Didn't someone mention earlier that he could see it on a 3-chip DLP? To me this would verify that the effect is solely a result of PWM and not phosphour decay at all, as DLP doesn't experience decay and has on-off cylce times of microseconds.

This has been brought up in the >3.5K forum and yes even 3-chippers would theoretically be susceptible to rainbow artifacts since they can produce only these colors: R,G,B,C,M,Y,White,black at any one instant and all other colors are produced by mixing these temporally. I have never seen rainbows an a 3-chip DLP.

The other thing to mention is that remember that CRTs do not use any type of PWM/PDM methods and actually are producing the colors that you see, yet rainbow-like artifacts are still visible to many viewers depending on the phosphors in use and the refresh rates etc.
ChrisWiggles is offline  
post #298 of 883 Old 05-17-2006, 02:36 PM
Member
 
RBurgundy's Avatar
 
Join Date: Mar 2006
Location: MA
Posts: 180
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Here is the best explaination of what I see. Anyone see the same?

Any questions?
RBurgundy is offline  
post #299 of 883 Old 05-17-2006, 02:40 PM
AVS Special Member
 
xrox's Avatar
 
Join Date: Feb 2003
Posts: 3,169
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 50
Quote:
Originally Posted by Felgar View Post

I see what you're saying about the 13 subfields, and after a lot of thought and deliberation I agree that it makes sense. But, how can you be sure that each picture field is not sub-divided into many groups? Like in your example with value of 128 in an 8-bit system, that would correspond to about half time on. But the cell itself could pulse many times per frame rather than just once per frame. The signal processing would not change in doing that, because it would be same sub-pixel value, just repeated in much smaller time slices. It would all depend on how quickly the sub-pixel can respond.

Incidently the more I think about it, the more I'm lead to believe that any rainbow effect is the direct result of PWM and not phosphour decay. If the on-off periods are not in sync between sub-pixels, the result when darting eyes and picking up just a fraction of the image would be the rainbow because the image could be caught with 1 sub-pixel on and the other off.

Incidently this artifact would appear on a 3-chip DLP for exactly the same reason. Didn't someone mention earlier that he could see it on a 3-chip DLP? To me this would verify that the effect is solely a result of PWM and not phosphour decay at all, as DLP doesn't experience decay and has on-off cylce times of microseconds.


In the 8bit system the 128 gray level is acheived as I have shown above, with one sustain pulse at the end of the frame period. However it is more complicated than that.

The sustain pulse is not a single long pulse as you say. It is made up of bipolar pulses of equal length that match the AC frequency driving the display. Since Plasmas are AC addressed the 128 gray level sustain pulse will have 64 positive pulses and 64 negative pulses that alternate and the total sume is preceeded with one priming pulse and ended with one erase pulse. This however does not require the phosphor to turn on and off 128 times. It is just a consequence of the driving frequency.

In short, 8 subfields require phosphors fast enough to turn on and off 8 times per frame with pretty short rise and fall times.

Note: As I said in my previous post, out of 16.6 ms per frame only 4-6 ms is actually used to emit light due to addressing (write and erase)

Over thinking, over analyzing separates the body from the mind
xrox is offline  
post #300 of 883 Old 05-17-2006, 02:49 PM
Member
 
cirob's Avatar
 
Join Date: Oct 2005
Posts: 175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Hmm I have been reading through and I don't doubt all the people that see the rainbow, I don't see it on my nec, my buddy tom doesn't see it on his elite.
Fellas maybe you should watch the program instead of the set, I mean nothing man made is perfect.
cirob is offline  
Reply Plasma Flat Panel Displays

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off