LG OLED vs Samsung Super OLED - Page 5

I think the problem here is that this design is so unique, many want to talk about the underlying technology - and this design is essentially a two-step fractal. You have an underlying red element, green element, and blue element. These combine into white. You then take a group of those with red, green, and blue filtering (plus one unfiltered) and you have the top level structure.

What I've seen in many threads though is that when specifically discussing the tech behind the white OLED's ... many (including me) have called it a pixel and the elements making it from the red, green, and blue layers as being sub-pixels. While that's a bit of a broad usage, it certainly makes sense why people would call it that. So the question becomes ... just what should we name everything in order to aid in discussion clarity? This two-step fractal makes discussions difficult unless someone is particularly verbose. And since this design is so new, I expect people will continue discussing both levels of the physical tech.

Any recommendations on some good naming conventions to proceed with? I would submit we should use the strict definition of pixel (picture element) and sub-pixel. In other words use it to describe what we can actually see on the screen. So with that in mind, a sub-pixel would be a white OLED and its color filter. A pixel would be a grouping of 4 of those (one with a red filter, one with green, one with blue, and one with none).

Now however, what is a good naming convention so people can discuss the tech used for the white OLED? I think for the sub components, maybe layer element instead of sub-pixel? So it would be red layer element, blue layer element, and red layer element? I can't really think of a cute name for the actual white OLED that's an aggregate of them though. I guess maybe just calling it a WOLED is easiest?


While it may be a second order effect, one has to look at this in terms of actual deltas over the expected usage life time of an average TV - it's a question of magnitudes.

Obviously in a perfect world R, G, B would age at the same rate (or better yet, not at all) ... but that's not realistic at this time. There will be some drift, but we have to think about it in the aggregate. It's not just about comparing the actual half-life age values themselves ... but looking at how far those values push out serious error accumulation.

In modern PHOLEDs, both red and green have pretty long lives. Not the exact same, but it's far enough out there that the error accumulation over the average viewing life of a TV yields a relatively small drift. Regardless of how long the blue FOLED life is, as long as it's long enough ... the actual error accumulation won't be significant. And that's where the LG solution comes into play. Unless Samsung's partners have had some crazy new breakthrough for blue (which to my knowledge they haven't - and I follow Universal Display Corp quite regularly as I'm a stock owner) ... then the amount of drift their display will have will definitely be higher. While the half-life has been pushed out some ... it's still like an order of magnitude or more lower than red and green ... and the bigger problem, it's short enough that I'd expect some sizable error accumulation over the average viewing life of a TV. At least if it's used as your main set.


Wasn't implying you necessarily should be - just discussing the tech differences. Though I'm sure some will use this for spec wars

That said, the answers to this may impact the above. All things being equal, the lower drive voltage being used ... the slower the aging.

So they aren't going to have a wall mountable version without a backpack?

Crap. I was hoping they'd offer one sans speakers ... and with the electronics in a separate box connected via wire. Sounds like the base will not be removable in the backpack-less version.






Until we see numbers, I'm not sure that's inherently the case.

Why is this exactly? What sort of issues does it create?



Why does vertical stripes make it useless with a computer hooked to it?

Computers use subpixel rendering which only works with horizontal RGB stripes. (in some instances, BGR stripes also work, but this is not universally supported) Anything else does not render things correctly. (primarily text)

Regardless of the color source, the TV converts the data to RGB. On the panel, luma can be outsourced to thw white pixel ???

Ya it's called Delta

They improved it every year. Not sure if they still need to use Delta on their new 8 Domain panel.

Good point.

http://en.wikipedia.org/wiki/Subpixel_rendering



So the question becomes, will SW in the future always make the assumption you're using a traditional LCD? I hope at some point OS's or at least browsers will support more than just one layout for font rendering. Really wouldn't be too hard to implement.



To my initial question though, are their other IQ issues to using 4 subs (disregarding potential issues with sub-pixel rendering).

I can only think of softer image since the 4th pixel is just luma. We will have to see if there are any actual problems using a colorimeter.

Hopefully RGB luminance will remain uniform across 0-255

Yeah I was more concerned whether he was implying some sort of visual artifact - because I can't think of any.

As far as it being softer ... I'd actually be happy as long as it can still resolve a 1080p line test. I'm currently using a 60A3000 for my main set, and the thought of losing the awesome fillrate hurts. I love the smooth, 'film-like' look to the image. It doesn't exaggerate aliasing, noise, etc like lesser file rates do.

That's actually my main concern with Sony's Crystal LED. It seems promising, but the fill rate is pretty damn low.

RGBW is an LCD technology (for now). LG OLED's use all white subpixels with color filters on top.

Fill rate?

According to sources, it also features white sub-pixel in addition to the RGB

Pixel fill rate is the percentage of a pixel that's actually illuminated (versus gaps, borders, etc).

An obvious (extreme) example a low fill rate were older LCD projectors and monitors where you could visibly see a 'screen door effect' (SDE). Even if you can't actually resolve SDE though, it can still impact your viewing.

Don't get me wrong, some people actually prefer the image since it appears sharper even though the resolution is the same. And that can be nice an really high quality video, but not everything we watch is perfect. The problem is it can also exaggerate things like noise, aliasing, etc. For me personally, I'd rather have the most natural/accurate image to start off with ... and if I feel like it needs some artificial sharpening, I'll use my video processor.

Ah that

I've never heard anyone use the term Fill Rate in that manner. I usually associate it with Frame Buffer

It would be ideal to have a display with no gaps in between the pixel but we have to wait until Stacked OLED becomes norm. But yes, I too would love to know how bad it is on the LG and Samsung OLEDs.

Sorry to interrupt the learned discussions posted here, but how do you bolt a flat panel mount on a screen only 4mm. thick? That's asking a lot from a millimeter or two of screw.
Is there a bolt welded on to the back?

It seems like the backpack is not attached to the screen (so that could possibly be put somewhere other than behind the screen depending upon the cabling).

It pains me too ... I'm a software developer

But it is the term I've actually seen regularly used for it.

http://www.cine4home.com/know-how/4-...sharpness.html




So right here is CLED (or as I call it, XLED)



What's interesting is that there doesn't seem to be any obvious delineation between sub-pixels which is great (maybe they're stacked?) ... but the actual separation between pixels if pretty huge.

Hopefully that's something that can be addressed.

You sure about that? From the description I read, it unfortunately sounded like the backpack is attached? It's got downward firing speakers on it.

WOW!

Pixels do appear to be stacked and the gap is massive for a 50/60".

So the only correct definitions for pixel and sub pixel are those, in my opinion.

I think those are good terms in general. As a practical matter, once they are constructed, they will have WOLEDs (no matter that they are achieved from the multiple layers).

Basically, if LG can make the color temperature drift minimal through about 20-30K hours, they will have a great result here. I really can't say they will, but for the early adopters I hope so. I also imagine this number will improve over time (across product generations).

This is one of several problems Samsung will have that LG -- in theory -- will either not have or has reduced.

The following was from a previous post which I think was from a review.
Sounds from this that it is not permanently attached.

The LG 55-Inch 55EM9600 has two configurations. One is a table stand version with the speakers and electronics built into the base. The other, a wall mount version with the electronics and downward firing speakers housed within a backpack that sits between the wall and the screen.

Interestingly ... I read that very same description and came away with the exact opposite interpretation.

Ahhhh the beauty of the English language. Ambiguous to a fault

http://www.techotopia.com/index.php/...nt_details.jpg

Besides, iDevices don't use sub-pixel font rendering and they sell like hotcakes. (And doesn't high ppi make it moot anyway?)

I love the idea of "stacked" color elements as I find most direct-view HD displays distractingly noise-ridden from wide-angle viewing distances merely because of the visible pixel structure from side-by-side color elements at the 1080 level.

My LCOS projector, however, allows me to get almost 1 screen width away because it has "stacked" color and small inter-pixel spacing.

If we can bring that pixel character to direct-view displays it will be a large step forward in allowing wide-angle viewing from direct-view panels. That will be a big advantage when projected images are replaced by ultra-large direct-view displays over the next decade or so (my personal time estimate for feasable direct-view 80" + displays).

What you said...100% agree.

This is because they are low power devices (less taxing to render without it) and have very high density screensat least the iPod/iPhone do, at 330 PPI.

A 55" 1080p OLED screen? 40 PPI. Computers (Windows or Mac) do use Subpixel font rendering, and it's essential for anything below about 150 PPI. (and still beneficial at higher densities)


And LCoS isn't even the best example of it, because they're three chip devices that have imperfect alignment. Single chip DLP is the ultimate in sharpness because it essentially has full colour pixels without RGB subpixels, using a single chip, so there are no alignment issues.

D-ILA (JVC LCoS)


DLP:


Source

Full colour pixels (however it's achieved) is definitely the way forward, but it looks like 4K displays are likely to be the best alternative in the near future for direct-view displays, rather than projection-based displays. (doesn't fix the problem, but helps mask it with the increased pixel density)

Oh cool. Do you know off the top of your head if Windows offers something similar? That would make this moot.


Not that I think something like this is why or why not iDevices sell like hot-cakes ... you're right for iPhone 4/4S ... the ppi makes this moot.

A large-sized 1080p display doesn't have that luxury though. A 55" screen at 1080p is like 40ppi

Obviously your viewing distance is significantly farther however. Whether sub-pixel rendering is needed or not depends on where you're sitting. Anything closer than say 10' and you're getting worse perceived performance than an iPhone retina display (assumed to be held at 12").

Agree completely. My current set is a LCoS ... and combination of stacked sub-pixels and near-perfect fill rate is fantastic.

Sony's Crystal LED looks to get it half right

While I generally agree, the proof is in the pudding. There are 3-chip LCoS displays that pass 1080p resolution tests, so getting the alignment right is obviously possible. I assume the issue is that alignment changes over throw distance? So you'd need alignment calibration depending on where it's located?

Certainly you are correct that single-chip solutions don't have to worry about that though. However with DLP, the side-effect is you need to worry about temporal issues due to PWM for color production.


Probably. Thankfully OLED and LED-based displays at least have the potential to eventually offer full color pixels. Whether it ends up being tenable or not remains to be seen though (either cost or performance issues).

Yes doesn't sound like pentile. Am I mistaken but doesn't pentile makes chorma subsampling a permanent "feature" :P

I think the confusion here is also that Sammy's RGB is the pixel per se, but LG's RGBW is NOT the pixel... the pixel we see is the color filter.


Key word is currently Frankly I think as a first generation device it is pretty impressive. Problem is how fast can they make RGB more perceivably better than LG's RGBW (I'm no longer going to use the misleading WOLED terminology, not lest because others were using it). Otherwise it will be just as "successful" as their active 3D


My post in the OLED thread. I am an ardent supporter of OLED but I am not blind I just don't think it is an inherent OLED issue.



Agree that this is puzzling to me too. But literature so far has said the contrary ie the aging is similar. So I will reserve my skepticism.


Another reason why I think IGZO is for large panels, since LTPS solution is already cost parity to LCD. LG cannot compete with Sammy on small/medium sizes, maybe not even monitors.

As I said there, it's clear to me that they were. It's also distinctly possible that what they were hiding is that they had to drive the panels to some insane level to make them pop on the show floor -- and that for whatever reason that's true it will be fixed by shipping time.

That said, here's Harry McCracken at Cnet... Pretty much expressing my feelings in spades:

"Giant OLED TVs aren't inherently amazing.
It's a long-standing CES tradition for TV makers to try and out-do each other with demos of high-end big screens that won't wind up in many living rooms anytime soon. This year, the makers in question were LG and Samsung, which had 55-inch OLED TVs at their booths. I checked out both sets--and didn't come away lusting after either one. The LG, at least when I saw it, was displaying gimmicky 3D videos, which made it hard to judge how good it would be for anything else. And the video clips I saw on the Samsung were way over-saturated, giving everything an artificial, candy-colored effect that reminded me of some OLED smartphones."