Curiousity . . . Flicker and Default GrayScale in Standard Mode at 1080p/24 on 5020FD - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #1 of 268 Old 02-26-2009, 07:21 AM - Thread Starter
Member
 
Scraps628's Avatar
 
Join Date: Feb 2008
Posts: 60
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Okay, I'll do it. I'm still curious, and I know the rest of the old-timers are going to have a stroke if we keep talking about this on the other thread.

Here's why I started this thread.

One, there's this video - http://www.youtube.com/watch?v=ijq0mV9GPbY. It's infamous now, given that the person who shot it, video313, is unnecessarily abrasive with everyone who tries to weigh in on the subject. Consequently, he is encouraged to leave this thread alone so as avoid chasing away people who may help us answer these questions.

Two, people other than video313 have seen the same flicker issue in both the Pure Cinema "Standard" and "Advanced" settings, so it does not appear to be a complete fluke:

Quote:
Originally Posted by sheedoe View Post

Please don't kill me for bringing this up, but I have a question regarding the PC modes and 60hz vs 72hz (yes I know... here we go again! )
Last night while I was setting my G9 digital camera up to take some screen shots, I noticed something similar on the LCD screen of my camera to Video313's finding here. The camera was set @ 1/10 shutter speed and f 8.0 aperture.
PS3 was set to output 1080/24P and my Onkyo 885 pre/pro video out set to "Through" to ensure proper 1080/24P feed.

With a 1080/24P feed from the PS3 (playing Wall-E blu-ray), switching between the different PC modes, I've noticed a faster flicker (of the same type) in "Standard" and "Advanced" mode, and a slower flicker (of the same type) in the "Off" and "smooth" modes, same as on video313's youtube video.

With a non 1080/24P feed from my digital cable box, I noticed a slow flicker for all modes as if they were being refreshed at 60hz.

Now I know a camera is not a proper tool to measure refresh rate. But,
1) Why am I seeing these distinct flicker patterns between the pair of PC modes?
2) Is this the refresh rate that is being picked up on the camera? If not, what is it?

If it is indeed the refresh rate, then it would seem that video313 observation is correct in that both the "standard" and "advanced" mode is refreshing at similar 72hz and the other two modes is rereshing at 60hz. I'd really appreciate an answer and thanks in advance.

Three, other people have apparently noticed that when the signal is 1080p/24, the TV will not use the ControlCal calibrated grayscale, so there is at least other corroborating evidence to suggest that SOMETHING different is going on in Standard mode than in Off mode when the TV receives a 1080p/24 signal:

Quote:
Originally Posted by Michael St. Clair View Post

This is true except with 1080p/24 native input. I did ControlCal greyscale adjustment this weekend. In 1080p/60 plus 1080i/720p/480p/480i, it is true that Off, Standard, and Smooth use the modified greyscale. However, with 1080p/24, Standard goes back to using the original greyscale and only Off/Smooth use the adjusted one. I verified this both with my eye, and photographically (taking DSLR pictures of test patterns of the different modes using manual settings). I even set up a greenish grayscale, temporarily, to make it easier to spot the differences at a glance (no, nothing so green that it would wear the green phosphors unevenly, and even then only for a few minutes).

This has been observed by others:

http://www.avsforum.com/avs-vb/showt...7#post15891567

http://www.controlcal.com/forum/show...&postcount=226

So here is the question I hope we might punch around, and it has nothing to do with perceived "smoothness" of the video, so stay away video313 - by your own admission this was not your original question, nor what you are interested in discussing. The question is:

What could explain the apparent flicker when a video camera views 1080p/24 content on a PDP-5020FD set to "Standard" or "Advanced" Mode, and why doesn't it show up in "Off" mode since that one is apparently refreshing at 72 Hz (by virtue of the fact that the TV automatically refreshes at 72 Hz when it detects a 1080p/24 signal if left in "Off" mode)?

One thought as a jumping off point - is it possible that the processing algorithms for performing inverse telecine are still taking place in the Standard and Advanced modes (albeit unnecessarily with 24 fps content), and as a result of the power demands of those processing circuits the panel is changing the frequency of the current being supplied to the panel? Is that crazy? Any thoughts would definitely intrigue some of us (at least me and sheedoe).

Thanks, and - you can't see this - but I have my hands up as if to say "don't kill me! - I'm just asking a question!" Hopefully we can have a fun, insult-free discussion of this without getting anyone too riled up.
Scraps628 is offline  
Sponsored Links
Advertisement
 
post #2 of 268 Old 02-26-2009, 07:42 AM
AVS Special Member
 
Clint S.'s Avatar
 
Join Date: Jun 2008
Location: Baton Rouge
Posts: 2,735
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 0
I'm listening, LOL.

God Bless,
-Clint
Clint S. is offline  
post #3 of 268 Old 02-26-2009, 07:48 AM
AVS Special Member
 
Michael St. Clair's Avatar
 
Join Date: Apr 2000
Location: Cincinnati, OH USA
Posts: 1,179
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Just to clarify for anyone who may misunderstand (though the original post captures the situation perfectly):

1) I'm not making any claims about the 3:3 processing (24fps->72hz). I don't even own a video camera. I haven't noticed any flickering when taking still photos but I also haven't been looking for it.
2) Regarding "Standard" PureCinema mode not using the adjusted greyscale when fed 1080p/24, yeah, I have verified this carefully. I repeated the ControlCal process three times (each time rolling back to the original factory settings) to make sure I was following the instructions to a 'T'. I've tested all four Pure Cinema modes in 1080p24, 1080p60, 1080i, 720p, 480p, and 480i. That is twenty-four different combinations I tested (times three tests), writing down the results each time. On my set, "Standard" only uses the adjusted greyscale when you feed the set something other than 1080p/24. My set was manufactured July 2008 and has not been updated with the new firmware.

Quote:
Originally Posted by Scraps628 View Post

Hopefully we can have a fun, insult-free discussion of this without getting anyone too riled up.

Good luck with that.
Michael St. Clair is offline  
post #4 of 268 Old 02-26-2009, 08:05 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 Scraps628 View Post

What could explain the apparent flicker when a video camera views 1080p/24 content on a PDP-5020FD set to Standard Mode, and why doesn't it show up in "Off" mode if all of the modes are apparently refreshing at 72 Hz (by virtue of the fact that the TV automatically refreshes at 72 Hz when it detects a 1080p/24 signal)?

Sorry, I'm no camera expert so forgive me if this is a dumb question. What is the sample rate of the original camera taking the video? And what is the shutter speed per sample?

Over thinking, over analyzing separates the body from the mind
xrox is offline  
post #5 of 268 Old 02-26-2009, 08:19 AM - Thread Starter
Member
 
Scraps628's Avatar
 
Join Date: Feb 2008
Posts: 60
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Well, I see that sheedoe had listed 1/10 shutter speed, but I think the frame rate on a video camera would be 30 fps or perhaps 60 fps, right? Either way, it is certainly confusing that the shutter would be open 3 to 6 times as long as the frame was being exposed. In the case of 30 fps, that would suggest that frames 1-3 were all from the same exposure. Can that be right?
Scraps628 is offline  
post #6 of 268 Old 02-26-2009, 08:35 AM
AVS Special Member
 
Trackman's Avatar
 
Join Date: Dec 2000
Location: Sacramento,Ca., usa
Posts: 1,207
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Michael St. Clair View Post

Just to clarify for anyone who may misunderstand (though the original post captures the situation perfectly):

1) I'm not making any claims about the 3:3 processing (24fps->72hz). I don't even own a video camera. I haven't noticed any flickering when taking still photos but I also haven't been looking for it.
2) Regarding "Standard" PureCinema mode not using the adjusted greyscale when fed 1080p/24, yeah, I have verified this carefully. I repeated the ControlCal process three times (each time rolling back to the original factory settings) to make sure I was following the instructions to a 'T'. I've tested all four Pure Cinema modes in 1080p24, 1080p60, 1080i, 720p, 480p, and 480i. That is twenty-four different combinations I tested (times three tests), writing down the results each time. On my set, "Standard" only uses the adjusted greyscale when you feed the set something other than 1080p/24. My set was manufactured July 2008 and has not been updated with the new firmware.



Good luck with that.

Mike, what were your results with the other PC modes?

Go Duke !
Trackman is offline  
post #7 of 268 Old 02-26-2009, 08:40 AM
AVS Special Member
 
Michael St. Clair's Avatar
 
Join Date: Apr 2000
Location: Cincinnati, OH USA
Posts: 1,179
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Trackman View Post

Mike, what were your results with the other PC modes?

These are my notes showing which PureCinema modes use the adjusted greyscale:


1080p24

Off: Yes
Standard: No
Smooth: Yes
Advanced: No


1080p60, 1080i, 720p, 480p, 480i

Off: Yes
Standard: Yes
Smooth: Yes
Advanced: No
Michael St. Clair is offline  
post #8 of 268 Old 02-26-2009, 08:50 AM - Thread Starter
Member
 
Scraps628's Avatar
 
Join Date: Feb 2008
Posts: 60
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Scraps628 View Post

Well, I see that sheedoe had listed 1/10 shutter speed, but I think the frame rate on a video camera would be 30 fps or perhaps 60 fps, right? Either way, it is certainly confusing that the shutter would be open 3 to 6 times as long as the frame was being exposed. In the case of 30 fps, that would suggest that frames 1-3 were all from the same exposure. Can that be right?

Ah, I think I've got it. Sheedoe was looking at the LCD of his camera while he (or she?) was preparing to take a still. The shutter speed was at 1/10, but that is irrelevant here because sheedoe was not taking a still. I assume the camera's LCD refreshes at 30 fps or 60 fps, and that was where sheedoe was seeing the the flicker, so the shutter speed of 1/10 is not part of the equation. From what I understand, there is still a shutter (so to speak) on video cameras, in that the light sensor may collect light for the entire 1/30 of a second (in the case of 30 fps) or less, depending on ambient lighting, but at most it won't expose for longer than the duration of the frame.

That does bring up an interesting point though. Assuming that the camera thinks it is very bright, and thus doesn't allow the light sensor to collect light for the entire 1/30 of a second, then the frame of video it displays will represent a shorter slice of time. If that shorter slice is 1/72 of a second, for example, then it could theoretically capture one frame of 72 fps and display it for a duration of 1/60 of a second on the LCD display, thus looking smooth.

That doesn't necessarily help us find the answer as much as suggest that the video evidence we have may be meaningless. Nevertheless, the fact that video313 and sheedoe both found the same results, repeatedly, makes me think that there's still an explanation out there.
Scraps628 is offline  
post #9 of 268 Old 02-26-2009, 09:05 AM
Advanced Member
 
Jeff_DML's Avatar
 
Join Date: Jan 2005
Location: San Diego
Posts: 788
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Michael St. Clair View Post

These are my notes showing which PureCinema modes use the adjusted greyscale:


1080p24

Off: Yes
Standard: No
Smooth: Yes
Advanced: No


1080p60, 1080i, 720p, 480p, 480i

Off: Yes
Standard: Yes
Smooth: Yes
Advanced: No

basically that seems to match with the modes that the TV is in 72hz refresh mode. So it seems modified grayscale is not used during 72hz.
Jeff_DML is offline  
post #10 of 268 Old 02-26-2009, 09:10 AM
Senior Member
 
jesyjames's Avatar
 
Join Date: Mar 2003
Posts: 370
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Michael St. Clair View Post

These are my notes showing which PureCinema modes use the adjusted greyscale:


1080p24

Off: Yes
Standard: No
Smooth: Yes
Advanced: No


1080p60, 1080i, 720p, 480p, 480i

Off: Yes
Standard: Yes
Smooth: Yes
Advanced: No

I measured exactly the same thing. It was at first looking like Standard with 1080p24 was using the modified grayscale, but I have since concluded the disk I was using wasn't encoded in 24p and by switching to a disk I knew was encoded in 24p and measuring confirmed that standard does not use the modified grayscale for 24p content.
jesyjames is offline  
post #11 of 268 Old 02-26-2009, 09:20 AM - Thread Starter
Member
 
Scraps628's Avatar
 
Join Date: Feb 2008
Posts: 60
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Jeff_DML View Post

basically that seems to match with the modes that the TV is in 72hz refresh mode. So it seems modified grayscale is not used during 72hz.

Well, that's sort of what we're trying to get deeper into. D-Nice has assured that - with a 1080p/24 signal - the display goes into 72 Hz mode in "Off" and "Advanced," but that it refreshes at 60 Hz for "Standard" and "Smooth." We're trying to explain why the flicker appears to correspond to "Standard" and "Advanced" and why the grayscale also is different in those two. I don't think we have any ability to conclude that the flicker means it's in 72 Hz mode though, unfortunately.


Quote:
Originally Posted by D-Nice View Post


these are the facts....

IF you send your Kuro a 1080p/24 signal, it WILL display it @ 72Hz (with 3:3 pull down) in Pure Cinema OFF and Advance. The ONLY difference between Pure Cinema OFF and Advance is that Advance will extract 24fps material out of 480i, 480p, 720p, and 1080i signals. AGAIN Pure Cinema OFF and Pure Cinema Advance are IDENTICAL when it comes to a 1080p/24 signal.

IF you use Pure Cinema Standard with a 1080p/24 signal, it WILL display it @ 60Hz with 3:2 pulldown.

IF you use Pure Cinema Smooth with a 1080p/24 signal, it WILL display it @ 60Hz and frame interpolation with ZERO pulldown. That's right ZERO pulldown.

Scraps628 is offline  
post #12 of 268 Old 02-26-2009, 09:30 AM
Advanced Member
 
Jeff_DML's Avatar
 
Join Date: Jan 2005
Location: San Diego
Posts: 788
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Scraps628 View Post

Well, that's sort of what we're trying to get deeper into. D-Nice has assured that - with a 1080p/24 signal - the display goes into 72 Hz mode in "Off" and "Advanced," but that it refreshes at 60 Hz for "Standard" and "Smooth." We're trying to explain why the flicker appears to correspond to "Standard" and "Advanced" and why the grayscale also is different in those two. I don't think we have any ability to conclude that the flicker means it's in 72 Hz mode though, unfortunately.

yeah I saw that

our testing is not that scientific but it is at least consistent and makes somewhat sense. Only issue that conflicts with D-NICE statement is for 24p input and PC off and standard.

my unscientific testing results

Quote:
Originally Posted by Jeff_DML View Post

I have video313 on ignore but you guys keep replying so I noticed his youtube video and thought it was interesting.

Setup:
PS3->Onkyo 606->5020
Verified 24p is coming through the chain via the HDMI PC trick

Used a Canon ZR500 and set the shutter to 1/500


Test 1:
PS3 playing the dark knight bluray
24p set to ON

flicker when set to standard and advance, no flicker on off or smooth. Matches the results in video313 test.
Note, removed the kuro setup menu to verify it was not some weird gui related issue

Test 2:
PS3 playing the dark knight bluray
24p set to OFF, i.e 1080p60

no flicker on any of the modes

Test3:
DISH VIP722->Onkyo 606->5020
1080i wheel of fortune
No flicker in any of the modes

DISH VIP722->Onkyo 606->5020
1080i Cinemax Todd sweeny movie
Flicker in PC advance
No flicker in the rest


side note: did notice flicker going on/off when set to advance and watching 1080i comercials


Based on these results i have to agree with video313, it seems that with 24p input , standard outputs at 72hz and off does not based on this crude test.

http://www.avsforum.com/avs-vb/showp...postcount=9409
Jeff_DML is offline  
post #13 of 268 Old 02-26-2009, 10:23 AM - Thread Starter
Member
 
Scraps628's Avatar
 
Join Date: Feb 2008
Posts: 60
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Jeff_DML View Post

yeah I saw that

our testing is not that scientific but it is at least consistent and makes somewhat sense. Only issue that conflicts with D-NICE statement is for 24p input and PC off and standard.

my unscientific testing results



http://www.avsforum.com/avs-vb/showp...postcount=9409

Check this out: http://www.cybercollege.com/tvp019.htm - specifically the part about fluorescent lights and flicker. Each subpixel in a plasma is basically a tiny fluorescent lightbulb. According to this page, you can expect to see flicker when videotaping fluorescent lightbulbs at higher than 1/60 shutter speeds.

My hypothesis is that the flicker effect of the fluorescent cells in the plasma TV is more noticeable in "Standard" and "Advanced" due to some difference in the way they are being powered up. If so, then repeating Jeff_DML's test above with the shutter speed at 1/60 s should at least change the perceived flicker if not eliminate it. If we can eliminate it by changing the shutter speed, then we know that flicker is not indicative of the refresh rate of the panel, but rather, the fluorescent flicker of the cells themselves (which presumably flicker on and off at 120 Hz like normal fluorescent bulbs?).

Would you be willing do that test Jeff_DML?
Scraps628 is offline  
post #14 of 268 Old 02-26-2009, 10:30 AM
Advanced Member
 
Jeff_DML's Avatar
 
Join Date: Jan 2005
Location: San Diego
Posts: 788
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Scraps628 View Post

Check this out: http://www.cybercollege.com/tvp019.htm - specifically the part about fluorescent lights and flicker. Each subpixel in a plasma is basically a tiny fluorescent lightbulb. According to this page, you can expect to see flicker when videotaping fluorescent lightbulbs at higher than 1/60 shutter speeds.

My hypothesis is that the flicker effect of the fluorescent cells in the plasma TV is more noticeable in "Standard" and "Advanced" due to some difference in the way they are being powered up. If so, then repeating Jeff_DML's test above with the shutter speed at 1/60 s should at least change the perceived flicker if not eliminate it. If we can eliminate it by changing the shutter speed, then we know that flicker is not indicative of the refresh rate of the panel, but rather, the fluorescent flicker of the cells themselves (which presumably flicker on and off at 120 Hz like normal fluorescent bulbs?).

Would you be willing do that test Jeff_DML?

I tried different shutter speeds such as 1/60 and the results seem the same

Edit again, I will retest tonight at 1/60 to double check
Jeff_DML is offline  
post #15 of 268 Old 02-26-2009, 10:37 AM
Senior Member
 
bmwolf's Avatar
 
Join Date: Mar 2007
Posts: 214
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by xrox View Post

Sorry, I'm no camera expert so forgive me if this is a dumb question. What is the sample rate of the original camera taking the video? And what is the shutter speed per sample?

I've done this test using my 24p HDV camcorder and I get similar results to Video313. With a 24p source, If I set the shutter to 1/24 or 1/48 of a second I don't get flickering in standard and advanced. If I set the shutter to 1/60 of a second I get the opposite; flickering in standard and advanced. This combined with watching credits for far to long, has led me to believe that 72hz mode is engaged in standard and advanced, and not during off or smooth. I can't prove it as fact and it goes against what I've read and thought in the past... so take that for what it's worth.
bmwolf is offline  
post #16 of 268 Old 02-26-2009, 10:47 AM
Senior Member
 
bmwolf's Avatar
 
Join Date: Mar 2007
Posts: 214
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Michael St. Clair View Post

These are my notes showing which PureCinema modes use the adjusted greyscale:


1080p24

Off: Yes
Standard: No
Smooth: Yes
Advanced: No


1080p60, 1080i, 720p, 480p, 480i

Off: Yes
Standard: Yes
Smooth: Yes
Advanced: No

I get the same results. I wonder what happens if you send a 24p signal to the TV when making adjustments with ControlCal. I'm gonna try it tonight.
bmwolf is offline  
post #17 of 268 Old 02-26-2009, 11:06 AM - Thread Starter
Member
 
Scraps628's Avatar
 
Join Date: Feb 2008
Posts: 60
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by bmwolf View Post

I get the same results. I wonder what happens if you send a 24p signal to the TV when making adjustments with ControlCal. I'm gonna try it tonight.

I haven't done the calibration yet (still in break-in phase), but I have seen repeated warnings not to try anything outside of the instructions. Maybe check with turbe and DisplayHarmony first - I think they would counsel against this (or perhaps they have tried and can tell you what will happen). Either way, be careful - I wouldn't want anyone to break their screen trying to figure this out.
Scraps628 is offline  
post #18 of 268 Old 02-26-2009, 11:08 AM
Advanced Member
 
Jeff_DML's Avatar
 
Join Date: Jan 2005
Location: San Diego
Posts: 788
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
here's my flicker hypothesis:

At 60hz the video camera is capturing at the same frame rate as the normal 60hz tv. Even if they are not exactly aligned you still only see 1 frame transition per frame captured. If the tv is in 72hz mode and since 72 is not evenly divisible into 60 then the video camera is capturing sometimes 1 frame or at other times 2. Since PDP is not a sample and hold display, i.e. full frame is not displayed 100% of the frame, then this causes a variation the amount of light the camera sensor since it basically getting twice the amount during the 2 frame capture, thus a flicker.
Jeff_DML is offline  
post #19 of 268 Old 02-26-2009, 11:14 AM
Senior Member
 
bmwolf's Avatar
 
Join Date: Mar 2007
Posts: 214
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Back in college whenever we would be shooting video with TVs or CRT monitors in the background, we always tried to make sure and use a shutter speed of 1/60, 1/120 etc. for this very reason.
bmwolf is offline  
post #20 of 268 Old 02-26-2009, 11:18 AM
Senior Member
 
bmwolf's Avatar
 
Join Date: Mar 2007
Posts: 214
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Scraps628 View Post

I haven't done the calibration yet (still in break-in phase), but I have seen repeated warnings not to try anything outside of the instructions. Maybe check with turbe and DisplayHarmony first - I think they would counsel against this (or perhaps they have tried and can tell you what will happen). Either way, be careful - I wouldn't want anyone to break their screen trying to figure this out.

I'm pretty sure others have done it. I think I recall reading over at the ControlCal forums that if you calibrate in advance that your settings only work in that mode. My guess is that this also applies to standard with a 24p input. I'm a cautious person, but in this case I'm confident that nothing bad will happen.
bmwolf is offline  
post #21 of 268 Old 02-26-2009, 11:20 AM - Thread Starter
Member
 
Scraps628's Avatar
 
Join Date: Feb 2008
Posts: 60
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Jeff_DML View Post

I tried different shutter speeds such as 1/60 and the results seem the same

Edit again, I will retest tonight at 1/60 to double check

Hmm... looks like bmwolf had the same results, so probably not necessary.

I'm going to look more into the fluorescent phenomenon to see if that can give us the explanation.
Scraps628 is offline  
post #22 of 268 Old 02-26-2009, 12:11 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 Scraps628 View Post

Check this out: http://www.cybercollege.com/tvp019.htm - specifically the part about fluorescent lights and flicker. Each subpixel in a plasma is basically a tiny fluorescent lightbulb. According to this page, you can expect to see flicker when videotaping fluorescent lightbulbs at higher than 1/60 shutter speeds.

Yes each subpixel is essentially a flourescent bulb (photoluminescence - UV excited phosphor). Interestingly, PDP and LCD produce light the same way . However, Plasma displays compose a frame, over time, using weighted subfields. This is a very important point (see below)

That link suggests using 1/60th or faster for very good reason. At 1/60th, regardless of sync, the camera will capture the exact same light output every single frame since the fluorescent light is operating at 1/60th also (remember, regardless of sync). At higher than 1/60th the camera will capture a different amount of light each frame since the sample rate is higher than the frequency the fluorescent light is operating at. The differing amount of light will then repeat itself at a lower frequency than 1/60th making it easy for the viewer to see on the film as flicker. If the camera sample rate is further increased the apparent frequency of the captured flicker will start to increase as well. If the camera sample rate is high enough the apparent flicker will reach near 1/60th and be hard to see again.

Quote:


My hypothesis is that the flicker effect of the fluorescent cells in the plasma TV is more noticeable in "Standard" and "Advanced" due to some difference in the way they are being powered up. If so, then repeating Jeff_DML's test above with the shutter speed at 1/60 s should at least change the perceived flicker if not eliminate it. If we can eliminate it by changing the shutter speed, then we know that flicker is not indicative of the refresh rate of the panel, but rather, the fluorescent flicker of the cells themselves (which presumably flicker on and off at 120 Hz like normal fluorescent bulbs?).

Plasma displays compose an image (each refresh) using 8-15 pulses depending on the manufacturer. And these pulses are weighted in increasing order. This means the first half of the refresh will be dark and the last half will be bright. This creates duty cycle flicker that some can perceive with the naked eye (especially peripheral vision).

Using a camera to capture plasma:

Assuming no motion (repeated frames) - If the sample rate of the camera is faster/shorter than the plasma refresh rate the camera will capture only a portion of the plasma composing a frame (ie - partially composed frame). Like I said above, the first half of the frame composition process is dark and the second half is bright. Therefore, the luminance of the camera's captured image depends on what portion of the plasma frame composition process is captured. This will obviously create visible flicker. Furthermore, you should see “color flicker” as well since the three colors RGB may have different weights according to the required final color for display. Furthermore, the phosphors activate and decay at different rates.

Therefore, the video with flicker suggests the following(theoretically)

1 – The camera sample rate is faster or slower than the display refresh rate

The video without flicker suggests the following(theoretically)

1 – The camera sample rate is equal to the display refresh rate regardless of sync or shutter time

Over thinking, over analyzing separates the body from the mind
xrox is offline  
post #23 of 268 Old 02-26-2009, 12:11 PM
AVS Special Member
 
Dahlsim's Avatar
 
Join Date: Jan 2004
Posts: 3,986
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 5 Post(s)
Liked: 11
Quote:
Originally Posted by bmwolf View Post

I'm pretty sure others have done it. I think I recall reading over at the ControlCal forums that if you calibrate in advance that your settings only work in that mode. My guess is that this also applies to standard with a 24p input. I'm a cautious person, but in this case I'm confident that nothing bad will happen.

At one time during early testing that was thought to be the case. The thought was that Advanced could be calibrated but since calibrating another PC mode gave you access to all 3 modes that's what D-Nice recommends and that's what everyone did, PC:Off or PC:Standard.

D-Nice tells me though that since then it's been found that while Advanced does use a different memory page we don't know how to save to it yet via ControlCal, if it can be saved to. It may be by altering the step by step a bit, such as by calibrating against a different signal type or such you may just discover a trigger that saves to the other memory page. ( Good luck ) Again I think most of the intense trial and error testing on Non-Elite models stopped a while ago. (Everyone was pretty happy with the results)

If you did find a trigger in the sequence that hit the other memory page then you would be able to save 2 different calibrated video mode grayscales on the XX20. As it stands the HDMI PC mode memories are the only way to calibrate more than grayscale.

Family DVD Collection
Insider thread last question? R.I.P.
Nothing will ever be attempted if all possible objections must first be overcome.
Samuel Johnson 1709-1784, British Author
Dahlsim is offline  
post #24 of 268 Old 02-26-2009, 12:44 PM - Thread Starter
Member
 
Scraps628's Avatar
 
Join Date: Feb 2008
Posts: 60
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by xrox View Post

Yes each subpixel is essentially a flourescent bulb (photoluminescence - UV excited phosphor). Interestingly, PDP and LCD produce light the same way . However, Plasma displays compose a frame, over time, using weighted subfields. This is a very important point (see below)

That link suggests using 1/60th or faster for very good reason. At 1/60th, regardless of sync, the camera will capture the exact same light output every single frame since the fluorescent light is operating at 1/60th also (remember, regardless of sync). At higher than 1/60th the camera will capture a different amount of light each frame since the sample rate is higher than the frequency the fluorescent light is operating at. The differing amount of light will then repeat itself at a lower frequency than 1/60th making it easy for the viewer to see on the film as flicker. If the camera sample rate is further increased the apparent frequency of the captured flicker will start to increase as well. If the camera sample rate is high enough the apparent flicker will reach near 1/60th and be hard to see again.

Plasma displays compose an image (each refresh) using 8-15 pulses depending on the manufacturer. And these pulses are weighted in increasing order. This means the first half of the refresh will be dark and the last half will be bright. This creates duty cycle flicker that some can perceive with the naked eye (especially peripheral vision).

Using a camera to capture plasma:

Assuming no motion (repeated frames) - If the sample rate of the camera is faster/shorter than the plasma refresh rate the camera will capture only a portion of the plasma composing a frame (ie - partially composed frame). Like I said above, the first half of the frame composition process is dark and the second half is bright. Therefore, the luminance of the camera's captured image depends on what portion of the plasma frame composition process is captured. This will obviously create visible flicker. Furthermore, you should see “color flicker” as well since the three colors RGB may have different weights according to the required final color for display. Furthermore, the phosphors activate and decay at different rates.

Therefore, the video with flicker suggest two possibilities (theoretically)

1 – The camera sample rate is faster than the display refresh rate

2 – The camera sample rate is longer than the display refresh rate and has a shutter time shorter than the sample rate.

The video without flicker suggest two possibilities (theoretically)

1 – The camera sample rate is equal than the display refresh rate regardless of sync or shutter time

2 – The camera sample rate is longer than the display refresh rate regardless of sync but has a shutter time equal to the sample rate.

I like where you're going with this, but what does "sample rate is longer" mean to you? I agree with point 1 in both circumstances (as a hypothesis), but I'm not following on 2 and I suspect it's because I haven't figured out what you mean by "sample rate is longer."

But anyway, I think your point about weighted subfields is very interesting. Is it possible that a different number of pulses is being used for "Off" and "Smooth" than are being used for "Standard" and "Advanced"? Or, is it possible that the TV is weighting them differently during Standard/Advanced vs. Off/Smooth? On the latter, for example, what if in Standard/Advanced the TV is weighting the pulses not for each 1/60 of a second frame, but rather for every two 1/60 of a second frames - or something like that? Is that possible? If so, something of that nature would essentially make the display appear to be flickering at 1/30th of a second, and then explain how a 1/60th camera could pick up flicker.
Scraps628 is offline  
post #25 of 268 Old 02-26-2009, 12:52 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 Scraps628 View Post

I like where you're going with this, but what does "sample rate is longer" mean to you? I agree with point 1 in both circumstances (as a hypothesis), but I'm not following on 2 and I suspect it's because I haven't figured out what you mean by "sample rate is longer."

Camera sample rate (ie - 60Hz or 30Hz). If the camera sample rate is 60Hz (16.7ms) and the Plasma is refreshing at 72Hz (13.9ms) you will see flicker on the camera "only if the camera shutter speed is shorter than 16.7ms". Of course this is only my thoughts. I could be wrong

Quote:
Originally Posted by Scraps628 View Post

But anyway, I think your point about weighted subfields is very interesting. Is it possible that a different number of pulses is being used for "Off" and "Smooth" than there for "Standard" and "Advanced"?

Yes, but it would make no difference regarding camera captured flicker.

Quote:
Originally Posted by Scraps628 View Post

Or, is it possible that the TV is weighting them differently during those two modes? On the latter, what if the TV is weighting the pulses not for each 1/60 of a second frame, but rather for every two 1/60 of a second frames - or something like that? Is that possible? If so, something of that nature would essentially make the refresh 1/30th, and then explain how a 1/60th camera could pick up flicker.

The only way the weight would affect camera capture flicker is if they were weighted differently on successive frames (even on a still image). Seems very unlikely.

Over thinking, over analyzing separates the body from the mind
xrox is offline  
post #26 of 268 Old 02-26-2009, 01:20 PM - Thread Starter
Member
 
Scraps628's Avatar
 
Join Date: Feb 2008
Posts: 60
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by xrox View Post

Camera sample rate (ie - 60Hz or 30Hz). If the camera sample rate is 60Hz (16.7ms) and the Plasma is refreshing at 72Hz (13.9ms) you will see flicker on the camera "only if the camera shutter speed is shorter than 16.7ms". Of course this is only my thoughts. I could be wrong

Yes, but it would make no difference regarding camera captured flicker.

The only way the weight would affect camera capture flicker is if they were weighted differently on successive frames (even on a still image). Seems very unlikely.

On the last point, I think it would induce flicker, no? Basically, if the weighting was spaced out over two 1/60's of a second (i.e., 1/30 of a second), and the first half (read first 1/60 of a second) was darker than the second half (read second 1/60 of a second), then a video camera would get successively a darker frame and then a lighter frame, and so on and so forth. Additionally, since the camera basically gets the average of the first 1/60th and the average of the second 1/60th, it would turn some 16-30 weighted pulses (likely not visible as flicker) into 2 distinctly different pulses. This could explain it, theoretically, right? - though I don't know that I see any reason for them to change the pulses like this in the first place.

Someone feel free to point out why this makes no sense. It doesn't feel likely, but it technically fits, right?
Scraps628 is offline  
post #27 of 268 Old 02-26-2009, 01:34 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 Scraps628 View Post

I like where you're going with this, but what does "sample rate is longer" mean to you? I agree with point 1 in both circumstances (as a hypothesis), but I'm not following on 2 and I suspect it's because I haven't figured out what you mean by "sample rate is longer."

Actually, I drew it out on paper and you are right. There will be flicker on the camera film if the camera is set to 60Hz and display is at 72Hz or vice versa regardless of shutter speed. I'll draw it out on power point to show you later I'll also edit my post.

Cheers

Over thinking, over analyzing separates the body from the mind
xrox is offline  
post #28 of 268 Old 02-26-2009, 01:38 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 Scraps628 View Post

On the last point, I think it would induce flicker, no? Basically, if the weighting was spaced out over two 1/60's of a second (i.e., 1/30 of a second), and the first half (read first 1/60 of a second) was darker than the second half (read second 1/60 of a second), then a video camera would get successively a darker frame and then a lighter frame, and so on and so forth.

This would produce 30Hz duty cycle flicker that the viewer could easily see with the naked eye right from the display itself.

Over thinking, over analyzing separates the body from the mind
xrox is offline  
post #29 of 268 Old 02-26-2009, 01:52 PM - Thread Starter
Member
 
Scraps628's Avatar
 
Join Date: Feb 2008
Posts: 60
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by xrox View Post

This would produce 30Hz duty cycle flicker that the viewer could easily see with the naked eye right from the display itself.

Yeah, that's what I was suspecting too. I don't understand exactly how the weighted pulses are interpreted by the viewer's brain, so I was thinking (in the abstract) that perhaps there was a way to weight the pulses over 1/30th of a second, thus keeping the eye from seeing flicker. If that is theoretically possible, then you wouldn't see the flicker when viewing the screen itself, but a camera would see it because - as I said before - frame 1 would be darker than frame 2, and there wouldn't be any gradual build, just a pop from dark for the first 1/60th of a second to light for the second 1/60th of a second. Still though, I am inclined to say this theory sucks because I can't think of any reason for them to do this in the first place, since it seems like it would certainly add complications and seemingly offer little benefits.

I was just taking a shot in the dark.
Scraps628 is offline  
post #30 of 268 Old 02-26-2009, 01:56 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
If the flicker on the video is ~12Hz then it solidly suggests that the conflict between 72Hz display and 60Hz camera is to blame (or vice versa).

Over thinking, over analyzing separates the body from the mind
xrox 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