OLED TVs: Technology Advancements Thread - Page 369 - AVS Forum
Forum Jump: 
 234Likes
Reply
 
Thread Tools
post #11041 of 11055 Old Yesterday, 02:15 PM
AVS Addicted Member
 
rogo's Avatar
 
Join Date: Dec 1999
Location: Stop making curved screens
Posts: 30,286
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 236 Post(s)
Liked: 562
Darin (good to hear from you) and TGM, good points from both of you. I basically agree across the board.

Fundamentally, my sense is that the white serves a pretty limited function. And I think TGM, you agree here, too. Getting back some lost output more than trying to "save the pixels from themselves".

I doubt very much the long-run plan is to keep using the white for this reason. The design gets a lot simpler over time without it and given that you can lose the corresponding transistors, you probably don't need much improvement in overall light output to justify dropping it (if you're overall goal is a power budget).

Of course, to the extent that driving the material harder has some effect on lifespan, that's also a concern. But it seems probable that, too, will be mitigated over time.

One other random thought. LG uses a color-filter design that is not a million miles removed from their LCD technology because they're exceptionally good at patterning and producing that layer. The OLED makes light really differently from LCDs, however. It might make sense over time to rethink the color-filter layer and/or add a special film in between the light and the CF. Some sort of purpose-built OLED BEF (brightness enhancer) that takes advantage of the way OLED works. It looks like a lot of potential light is lost to the rigid lines of the color filter and/or the electrodes. That seems like an area where long-term improvement is more than a hypothetical.

There is no difference in HDMI cables. If you can see the picture without visible dropouts or sparklies, the cable is working at 100%. No other cable will display a better version of that picture. You're simply wrong if you think there is a better digital cable than one that is already working. (Oh, and plasma didn't die because of logistics problems, nor does OLED ship in big boxes because it comes from Korea.)
rogo is offline  
Sponsored Links
Advertisement
 
post #11042 of 11055 Old Yesterday, 03:10 PM
Advanced Member
 
sooke's Avatar
 
Join Date: Aug 2003
Location: Raleigh, NC
Posts: 893
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 27 Post(s)
Liked: 42
Quote:
Originally Posted by rogo View Post
Darin (good to hear from you) and TGM, good points from both of you. I basically agree across the board.

Fundamentally, my sense is that the white serves a pretty limited function. And I think TGM, you agree here, too. Getting back some lost output more than trying to "save the pixels from themselves".
Hhmm... I guess I'm going to be the odd man out. Increasing light output for colors with a common grey component, but not for colors without a common grey component, does not seem like you could get much benefit without hurting PQ. I would think you would want to normalize all your colors with how bright you could make your primaries. I don't have the experience with displays and color you folks do, I'm just going by inuition, but adding the white pixel to make some colors brighter does not seem likely. I agree if that is why it is there, it probably won't be around long.

On the other hand, if you look at the diagrams TGM made in is last post, there is much less total wear when the grey component is illuminated with the white pixel compared to doing it with the primaries thanks to the clear filter. I have no idea how hard the white would need to be driven to equal the light output of the combined 3 primaries, but using TGM's conceptual diagrams as an example, you can see there is 1/3 the total wear compared to using the primaries. This seems like a good reason to add the white pixel.

As for the terms "common mode" and "grey component", sounds like TGM has much working experience with colors and displays, so if he is used to "grey component" that is what I will use. I liked fafrd's use of common mode because it is a EE term and so I understood immediately what he was saying.

Panasonic TC-P60ZT60 plasma panel
Sony BDP-S5100 Blu-ray player
Samsung SMT-H3272 DVR
Yamaha RX-V650 AVR
XBOX 360
URC MX-780 remote control
sooke is offline  
post #11043 of 11055 Old Yesterday, 03:47 PM
AVS Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 3,748
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 620
Quote:
Originally Posted by rogo View Post
Darin (good to hear from you) and TGM, good points from both of you. I basically agree across the board.

Fundamentally, my sense is that the white serves a pretty limited function. And I think TGM, you agree here, too. Getting back some lost output more than trying to "save the pixels from themselves".

I doubt very much the long-run plan is to keep using the white for this reason. The design gets a lot simpler over time without it and given that you can lose the corresponding transistors, you probably don't need much improvement in overall light output to justify dropping it (if you're overall goal is a power budget).

Of course, to the extent that driving the material harder has some effect on lifespan, that's also a concern. But it seems probable that, too, will be mitigated over time.

One other random thought. LG uses a color-filter design that is not a million miles removed from their LCD technology because they're exceptionally good at patterning and producing that layer. The OLED makes light really differently from LCDs, however. It might make sense over time to rethink the color-filter layer and/or add a special film in between the light and the CF. Some sort of purpose-built OLED BEF (brightness enhancer) that takes advantage of the way OLED works. It looks like a lot of potential light is lost to the rigid lines of the color filter and/or the electrodes. That seems like an area where long-term improvement is more than a hypothetical.

I'm just not understanding why you are saying this (in bold above) - do you have some other specific information or are you just commenting on the same examples that have been posted.

Extracting the 'grey component' (or 'common mode' if you are EE like sooke and I) is a pretty important function and leads to significantly less use of the primaries (especially blue).

And since the examples shown never show all three primaries lit up, it's pretty strong evidence that this is exactly how the white sub is used (as pointed out by TGM).

When there is no grey component, the white sub must be off by definition (including the case of any two of the primaries being used all the way up to 100%).

But let's consider the color composed of 100% R + 100% G + 100% B (ie: a white).

Let's assume the color filters are 50% efficient and further that the light output fro the WOLED layer is evenly divided 33% into each primary (as far as the color filters are concerned).

So each of the colored subs is only going to be ~17% efficient and driving the R & G & B primaries at 100% is only going to result in light output which is ~17% of that produced or ~50% of the light output of a single sub.

That means if the white sub pixel is going to be used to drive that same white light output, it only needs to be driven to the 50% level (because none of the primaries are filtered out and there is no color filter to further degrade efficiency.

Both for reduced power consumption as well as increased lifetime, I would think LG would want to make use of the white sub whenever possible and as strongly as possible (meaning to the limit offered by the grey component).

We can debate this ad nauseum, but the easiest thing would be to convince an owner to run color ramps or a color palette and check with a magnifying glass for any examples where all 4 subs are used. My guess is that there are none (with the possible exception of the brightest possible white which would be ~150% of what the white sub could put out all by itself driven to max).

By way of comparison, if LG sacrifices the white sub and increased the size of the other primary subs (to 133% of the current size), max white output would increase from ~50% of the current white sub to ~67%.

To me, a white subpixel makes a heck of a lot more sense than a yellow subpixel (as in Sharp) and I won't be at all surprised to see it hanging around for a long while.

All of this analysis glosses over the impact of the inter-subpixel spacing which would make the 'eliminate the white sub' scenario a bit more favorable (but not enough to change the overall message).
fafrd is online now  
post #11044 of 11055 Old Yesterday, 05:56 PM
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 6,421
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 430 Post(s)
Liked: 797
Well one thing's for sure, I really wish I knew what the LG engineers know. I bet there's quite a bit of "learning the hard way" type information about all of this that would be cool to know.

Beware the statistical correlations that sound like they're indicative of something. Drowning deaths are tightly correlated to ice cream consumption. In fact, be wary of any statistic that is stated as if it comes with a self-evident conclusion: there is no such thing.
tgm1024 is offline  
post #11045 of 11055 Old Yesterday, 06:12 PM
Member
 
ChaosCloud's Avatar
 
Join Date: Dec 2011
Posts: 29
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 12
Quote:
Originally Posted by Wizziwig View Post
You will see there is never a case where they only use the white sub-pixel, even on gray ramps. Only pure red and blue seem to use a single sub-pixel.
The white subpixel alone is used for dull greens/browns. In my brief time with a 55EA9700 I did some testing. See pics below.
https://dl.dropboxusercontent.com/u/...8/P1010045.JPG
Attached Thumbnails
Click image for larger version

Name:	xp9.jpg
Views:	23
Size:	188.6 KB
ID:	286050  
ChaosCloud is online now  
post #11046 of 11055 Old Yesterday, 07:35 PM
AVS Addicted Member
 
rogo's Avatar
 
Join Date: Dec 1999
Location: Stop making curved screens
Posts: 30,286
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 236 Post(s)
Liked: 562
"Both for reduced power consumption as well as increased lifetime, I would think LG would want to make use of the white sub whenever possible and as strongly as possible (meaning to the limit offered by the grey component)."

This makes no sense to me. If you overdrive and utilize the white subpixel, you just torch that one instead. It's the same exact chemistry as any other subpixel -- they all have the exact same OLED "stuff" in there. If you keep using it "as strongly as possible" you will just kill it off.

You have to always remember the "blue subpixel" isn't blue. It has nothing to do with blue lifetime. It's white. Ditto the red. Ditto the green. You have to balance your use of the white one, or else it just lives fast and dies hard, and again you have an uneven wear problem and you're no longer going to get the expected colors out of the display when you use it.

There is no difference in HDMI cables. If you can see the picture without visible dropouts or sparklies, the cable is working at 100%. No other cable will display a better version of that picture. You're simply wrong if you think there is a better digital cable than one that is already working. (Oh, and plasma didn't die because of logistics problems, nor does OLED ship in big boxes because it comes from Korea.)
rogo is offline  
post #11047 of 11055 Old Yesterday, 08:58 PM
AVS Special Member
 
Wizziwig's Avatar
 
Join Date: Jul 2001
Location: SoCal, USA
Posts: 1,269
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 192 Post(s)
Liked: 145
LG slashes 55-inch OLED TV price (In Korea)
http://www.koreatimes.co.kr/www/news...33_165343.html


That's roughly $3784. Still much cheaper here in the U.S.
Wizziwig is offline  
post #11048 of 11055 Old Yesterday, 09:34 PM
AVS Special Member
 
Wizziwig's Avatar
 
Join Date: Jul 2001
Location: SoCal, USA
Posts: 1,269
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 192 Post(s)
Liked: 145
Quote:
Originally Posted by fafrd View Post
We can debate this ad nauseum, but the easiest thing would be to convince an owner to run color ramps or a color palette and check with a magnifying glass for any examples where all 4 subs are used. My guess is that there are none (with the possible exception of the brightest possible white which would be ~150% of what the white sub could put out all by itself driven to max).

If there really are no cases where all 4 sub-pixels are lit, doesn't that mean that 1/4 of the panel is off on average?


Looks like we finally have some proof that LG is starting to change the pixel structure a little.

http://www.lesnumeriques.com/tv-tele...re-n36031.html


Quote:
This year, LG reviews the architecture of its Oled tiles. While the 55EA970V and 55EA980V proposed the sub-pixels of equivalent sizes, the 55EC930V has white and blue two times larger than the Green and red.

Also of note, green, blue, and Red sub-pixels now consist of two under cells, but according to what we have seen, the top and bottom are not controlled separately.
First 2 pics are from 2013 (55EA970V and 55EA980V). Third pic is from 2014 (55EC930V).


In the oldest 2013 pic, you actually do see several pixels with all 4 subs enabled.
Attached Thumbnails
Click image for larger version

Name:	LG_55EA970V_woled-sous-pixels-g.jpg
Views:	35
Size:	214.4 KB
ID:	286906   Click image for larger version

Name:	LG-Oled-incurve-sous-pix.jpg
Views:	29
Size:	98.6 KB
ID:	286922   Click image for larger version

Name:	LG_Oled_TV_55EC930V_sous-pixels_g.jpg
Views:	36
Size:	329.0 KB
ID:	286930  
ChaosCloud likes this.
Wizziwig is offline  
post #11049 of 11055 Old Yesterday, 10:04 PM
AVS Addicted Member
 
darinp2's Avatar
 
Join Date: Sep 2003
Location: Seattle, WA
Posts: 21,182
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 57
Quote:
Originally Posted by Wizziwig View Post
If there really are no cases where all 4 sub-pixels are lit, doesn't that mean that 1/4 of the panel is off on average?
1/4 of the sub-pixels would be off at minimum, so the average would be even more off than that.

However, the first picture you attached has an example of all 4 sub-pixels lit a little above the center of the picture.

--Darin

This is the AV Science Forum. Please don't be gullible and please do remember the saying, "Fool me once, shame on you. Fool me twice, shame on me."
darinp2 is online now  
post #11050 of 11055 Old Yesterday, 10:34 PM
AVS Addicted Member
 
rogo's Avatar
 
Join Date: Dec 1999
Location: Stop making curved screens
Posts: 30,286
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 236 Post(s)
Liked: 562
It's very clear there are times when all sub-pixels are very much alive. But those test images show they are rare.

What they don't necessarily show is a lot of very representative content.

There is no difference in HDMI cables. If you can see the picture without visible dropouts or sparklies, the cable is working at 100%. No other cable will display a better version of that picture. You're simply wrong if you think there is a better digital cable than one that is already working. (Oh, and plasma didn't die because of logistics problems, nor does OLED ship in big boxes because it comes from Korea.)
rogo is offline  
post #11051 of 11055 Old Today, 05:22 AM
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 6,421
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 430 Post(s)
Liked: 797
Quote:
Originally Posted by Wizziwig View Post
In the oldest 2013 pic, you actually do see several pixels with all 4 subs enabled.
Quote:
Originally Posted by darinp2 View Post
However, the first picture you attached has an example of all 4 sub-pixels lit a little above the center of the picture.

--Darin
Quote:
Originally Posted by rogo View Post
It's very clear there are times when all sub-pixels are very much alive. But those test images show they are rare.

I'm not sure they're there at all. They actually look like two pixels neighboring each other that look like there are 4 subs on because of their proximity. This is what allows subpixel rendering to work in RGB stripe layouts; it looks like they're borrowing from each other.

Somewhere LG did post a quote indicating that the scalar was present (I don't think they used the word scalar) which would allow for that if the scalar was less than 1.0, but are you sure you can see a place where that happens? Be careful to identify what is likely the start and end sub of the group of 4.

Beware the statistical correlations that sound like they're indicative of something. Drowning deaths are tightly correlated to ice cream consumption. In fact, be wary of any statistic that is stated as if it comes with a self-evident conclusion: there is no such thing.
tgm1024 is offline  
post #11052 of 11055 Old Today, 05:29 AM
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 6,421
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 430 Post(s)
Liked: 797
Quote:
Originally Posted by rogo View Post
This makes no sense to me. If you overdrive and utilize the white subpixel, you just torch that one instead. It's the same exact chemistry as any other subpixel -- they all have the exact same OLED "stuff" in there. If you keep using it "as strongly as possible" you will just kill it off.

You have to always remember the "blue subpixel" isn't blue. It has nothing to do with blue lifetime. It's white. Ditto the red. Ditto the green. You have to balance your use of the white one, or else it just lives fast and dies hard, and again you have an uneven wear problem and you're no longer going to get the expected colors out of the display when you use it.
I'm fairly certain fafrd knows that. We need to keep in mind that the calculation for this is outside our current information set----we have no data on the pass-through specs of the filters, and the white pixel (and this is likely very important) is unencumbered by any loss due to a filter. Or if there is a filter on it, it's likely very close to clear.

It might well be that you're right, and that scalar I (and LG) speak of is "a work in progress". It's also at least conceivable that the firmware keeps track of the white usage over time and throttles that scalar down appropriately in an effort to achieve "wear balancing".

The interesting thing about that scalar is that if it's done properly, no matter what it is it won't change the color of the overall pixel (as a whole); or very minimally so. So it's free to change dynamically as the firmware sees fit.

Beware the statistical correlations that sound like they're indicative of something. Drowning deaths are tightly correlated to ice cream consumption. In fact, be wary of any statistic that is stated as if it comes with a self-evident conclusion: there is no such thing.

Last edited by tgm1024; Today at 06:26 AM.
tgm1024 is offline  
post #11053 of 11055 Old Today, 11:28 AM
AVS Addicted Member
 
darinp2's Avatar
 
Join Date: Sep 2003
Location: Seattle, WA
Posts: 21,182
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 57
Quote:
Originally Posted by tgm1024 View Post
I'm not sure they're there at all. They actually look like two pixels neighboring each other that look like there are 4 subs on because of their proximity. This is what allows subpixel rendering to work in RGB stripe layouts; it looks like they're borrowing from each other.
I'm not following how one of the pixels on the bottom of this crop doesn't have 4 subs on. Seems to me that on the bottom row there is a single green sub on and it either belongs to the left pixel or the right pixel. Whether they borrow or not I see 7 subpixels in a row on the bottom right that are turned on, which to me means that at least one of the pixels has 4 subs on.

I think there are 9 pixels in this cropped image. Are you seeing a different number?

--Darin
Attached Thumbnails
Click image for larger version

Name:	lgcrop.jpg
Views:	3
Size:	10.3 KB
ID:	287769  
tgm1024 likes this.

This is the AV Science Forum. Please don't be gullible and please do remember the saying, "Fool me once, shame on you. Fool me twice, shame on me."
darinp2 is online now  
post #11054 of 11055 Old Today, 11:34 AM
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 6,421
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 430 Post(s)
Liked: 797
Quote:
Originally Posted by darinp2 View Post
I'm not following how one of the pixels on the bottom of this crop doesn't have 4 subs on. Seems to me that on the bottom row there is a single green sub on and it either belongs to the left pixel or the right pixel. Whether they borrow or not I see 7 subpixels in a row on the bottom right that are turned on, which to me means that at least one of the pixels has 4 subs on.

I think there are 9 pixels in this cropped image. Are you seeing a different number?

--Darin
Nope, lol, I'm seeing it now. Not sure how I missed it---I didn't see it when I looked through the images before. Well, it's consistent with their scalar approach anyway, so I'm not "surprised"....I just wish I knew the precise algorithm used, and especially whether or not there is any kind of dynamic wear balancing.

Beware the statistical correlations that sound like they're indicative of something. Drowning deaths are tightly correlated to ice cream consumption. In fact, be wary of any statistic that is stated as if it comes with a self-evident conclusion: there is no such thing.
tgm1024 is offline  
post #11055 of 11055 Old Today, 02:00 PM
AVS Addicted Member
 
rogo's Avatar
 
Join Date: Dec 1999
Location: Stop making curved screens
Posts: 30,286
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 236 Post(s)
Liked: 562
Quote:
Originally Posted by tgm1024 View Post
I'm fairly certain fafrd knows that.
Again, sometimes the explanation isn't directed at the person whose post is being commented on. It's there for anyone who might read this.
Quote:
We need to keep in mind that the calculation for this is outside our current information set----we have no data on the pass-through specs of the filters, and the white pixel (and this is likely very important) is unencumbered by any loss due to a filter. Or if there is a filter on it, it's likely very close to clear.
There is still some polarizing going on with the LG design, for whatever reason that's needed. Exactly what it's doing to output, I'm less sure.
Quote:
It might well be that you're right, and that scalar I (and LG) speak of is "a work in progress". It's also at least conceivable that the firmware keeps track of the white usage over time and throttles that scalar down appropriately in an effort to achieve "wear balancing".
Definitely possible. It seems less likely that something is monitoring usage across 2 million pixels, heading for 8 million, and storing that data then algorithmically adjusting how the display works based on it... But possible.
Quote:
The interesting thing about that scalar is that if it's done properly, no matter what it is it won't change the color of the overall pixel (as a whole); or very minimally so. So it's free to change dynamically as the firmware sees fit.
This may be true. I'm less sure than you are.

Quote:
Originally Posted by darinp2 View Post
I'm not following how one of the pixels on the bottom of this crop doesn't have 4 subs on. Seems to me that on the bottom row there is a single green sub on and it either belongs to the left pixel or the right pixel. Whether they borrow or not I see 7 subpixels in a row on the bottom right that are turned on, which to me means that at least one of the pixels has 4 subs on.
This of course was correct. Glad we came to consensus.

There is no difference in HDMI cables. If you can see the picture without visible dropouts or sparklies, the cable is working at 100%. No other cable will display a better version of that picture. You're simply wrong if you think there is a better digital cable than one that is already working. (Oh, and plasma didn't die because of logistics problems, nor does OLED ship in big boxes because it comes from Korea.)
rogo is offline  
Reply OLED Technology and Flat Panels General

Tags
Led Hdtv , Lcd Hdtv , Plasma Hdtv , Oled Tv , Lg , Samsung

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