The Case Against OLED

[Kachda]

People constantly talk about OLED burn in, completely ignoring that LCDs TVs also have issues over time. Interesting to look at the rtings results below that include LCD panel issues

Longevity And Burn-In Investigation: 10-Month Results

It has been about ten months since we started our long-term accelerated longevity and burn-in test, and our 100+ TVs and monitors have cumulated over 500,000 hours of runtime.

www.rtings.com

 

[Tim Link]

There is often a power saving setting. Turn it off. Modern flat panels don't use much juice.

Power saving is definitely off. Why it is that setting the local dimming to low lowers the entire scene brightness, I don't know. I guess they reduce the highs and lift the lows at the same time rather than just having the entire LED array at closer to full blast all the time. There may be a power supply or heat issue with that, kind of like how home theater receivers can not often run all 7 channels at once with their individual full power ratings. I watched a video where a guy added a whole bunch of extra LEDs to the backlight of an LCD, and it got up to over 300 watts power draw on I think just a 32" monitor. He was stacking LCDs to get extreme contrast. So when a high intensity full LED array is all on at peak brightness the power draw can be significant.

 

[Curvature]

It doesn't have to. The huge dynamic range of visible light is compensated for by constriction and dilation of the iris. What TV's need to do is reproduce the dynamic range of the retina, which is substantially less.

It is the same with sound, the muscles in the ossicles of the middle ear tighten and loosen via a reflex, which reduces the dynamic range of real life sound being transmitted through the fenestra ovalis into the Organ of Corti.

There needs to be a real, physical range for the ear/eye to compress and map. You can't skip directly to the psychovisual.

The PQ EOTF curve already exists and is the equivalent of an equal loudness contour.

 

[Tim Link]

32,000 nits – luminance of white card illuminated by sunlight+skylight at noon on a clear day
320 nits – luminance of diffuse white rendered by a typical consumer TV
32 nits – luminance of typical diffuse white in cinema
3.2 nits – the light from a single candle at 1 foot distance

No TV is capable of this dynamic range. I'm not even sure we have cameras that can capture it.

A Day with Charles Poynton - Insight Media

I recently had the opportunity to attend a workshop with Charles Poynton, a renowned mathematician and video expert. The informal workshop took place in New York with just a few others, providing a great opportunity to listen and discuss hot button issue around High Dynamic Range (HDR). Poynton...

www.insightmedia.info

And then on the PQ EOTF curve.

Should HDR Displays Follow the PQ Curve? - Insight Media

by Chris Chinnock, Insight Media This topic of discussion started as an email thread in the HDR work group of the International Committee of Display Metrology (ICDM) and then sparked a long discussion in the recent face-to-face meeting in Mt. View, California. Per agreement with ICDM, they...

www.insightmedia.info

Modern TVs use a lot of processing, including curves, including noise models, to deliver the picture.

The issue as I see it is that TVs are mass market devices. The most expensive TVs are usually about furniture and decor and aren't necessarily the best performers. As it is with the most expensive speakers. However, with speakers you have full access to the same tools that engineers use in studios to do their work, and they are relatively inexpensive, easy to setup and play.

With TVs, mastering monitors run tens of thousands of USD and the very best are not available for casual purchase at all. Most have fairly small displays as well, but are heavy and thick. Even if you do buy one, and I was considering doing just that to avoid the panel lottery and reliability issues, these are professional tools, and not consumer-friendly. As far as I know you have to manually select colorspace and other elements when using them, and they have no HDCP compatibility, and usually require special connections.

The more technical consumer will not be able to afford good display measurement gear, while excellent lab-worthy measurement microphones, although not cheap, are affordable.

Say you do measure the TV. Calibration only goes so far and some processing features are not open to be manipulated.

If there was ever an example of gatekeeping, it's with TVs.

Fortunately or unfortunately I've done enough work to understand display flaws, and the only conclusion I've reached is that I would rather be slightly grumpy as I watch a cheap display then buy an expensive TV I would consider artifically crippled by mass market considerations and manufacturing.

Maybe it's why I prefer audio.

Thanks for sharing these articles. Interesting stuff that expands my understanding of the concepts and issues. It was good to hear confirmation of what I had noticed and argued with my dad about years ago - that you need to boost colors a little with landscape photgraphy because of the dimmer indoor viewing conditions. Or maybe you shouldn't. It depends. My dad just kept it all accurate and figured the viewer would make some mental adjustments. I think there's some truth to that because there's always a price to artificailly boosting colors. Better maybe to just view under better lighting.

 

[Tim Link]

people might find this interesting.

Very interesting. Confirms some of what I suspected. Playing with DaVinci resolve to try to color grade my RAW photos as still frame videos I quickly came to appreciate some of the issues. Now you can do it directly in Adobe Lightroom. No matter how good your camera or display is, you're gonna have to tone map a lot of images, let some things fall into the noise floor or blow out as highlights. How do you make it look as nice as possible? It seems to vary with each image.

 

[Aerith Gainsborough]

People constantly talk about OLED burn in, completely ignoring that LCDs TVs also have issues over time. Interesting to look at the rtings results below that include LCD panel issues

Longevity And Burn-In Investigation: 10-Month Results

It has been about ten months since we started our long-term accelerated longevity and burn-in test, and our 100+ TVs and monitors have cumulated over 500,000 hours of runtime.

www.rtings.com

Obviously LCD's backlights age too. Every illuminant does. LEDs especially if they are driven hard / get hot.
It's the main reason why I don't think µLED will be the "God walking on water" everyone thinks they will be.
I've seen tons of blue anorganic LEDs fail way before their green/red counterparts in computer RGB components. Mostly because they are driven too hard.

That being said: my TV (Sony XF90) is 5 years old now and has a crapton of hours under it's belt. Picture is as good as ever.
Though to be fair: outside of HDR content, it always runs near minimum brightness since I am a light shy cave dweller by nature. :'D

 

[Tim Link]

I just had my mind blown doing a little test. My Pixel 4 phone screen is OLED. My 2015 MacBook Pro has an LCD with very noticeable backlight bleed. Sitting here in the dark I compared the same video of Osaka Christmas Lights in HDR, using a small portion of the LCD screen to match size. At first the OLED was killing it with highlight brightness, contrast, and color. But then I thought maybe I should try to adjust the color on the LCD. I switched the color profile from "Color LCD" to "Apple RGB." Now the two screens look nearly identical, with the LCD having a slightly more neutral looking tint. The OLED looks a touch too bluish green to my eye. It's really, really close. What's amazing is that the OLED is accepting the signal as HDR while my LCD does not support that so it's just SDR. It's so close to the same in every way it's a real shocker.
I've generally preferred the color LCD profile for it's more natural look. The Apple RGB is a punchier, more dynamic look. That flatter profile can be mistaken for a lack of ability of the display. Color me surprised!

 

[Aerith Gainsborough]

It's really, really close.

Calibrated, display technologies should look very similar in regards to white balance and colors (assuming both are wide-gamut displays).
The obvious differences come from each technologies particular strengths and weaknesses:

OLED:

• small bright objects on dark backgrounds w/o visible light bleeding into the dark area (not sure it's obvious on a smartphone but it IS obvious on a large panel TV)
• fairly agnostic to the angle it's viewed at (again: irrelevant in case of a tiny smartphone screen, very much relevant if you glance at a TV from the side or even sit close in front of it. I can clearly see colors of my TV losing intensity towards the edges due to viewing angle)
• potential motion clarity, especially older VA panels can have trouble with grey to grey transitions leading to "smearing".

LED:

• still able to appear bright in bright environments (not much of a factor if you watch in dark rooms, as both technologies should get bright enough once your eyes are adjusted to the dark)

So yes, your findings do not surprise me.

 

[Tim Link]

Calibrated, display technologies should look very similar in regards to white balance and colors (assuming both are wide-gamut displays).
The obvious differences come from each technologies particular strengths and weaknesses:

OLED:

• small bright objects on dark backgrounds w/o visible light bleeding into the dark area (not sure it's obvious on a smartphone but it IS obvious on a large panel TV)
• fairly agnostic to the angle it's viewed at (again: irrelevant in case of a tiny smartphone screen, very much relevant if you glance at a TV from the side or even sit close in front of it. I can clearly see colors of my TV losing intensity towards the edges due to viewing angle)
• potential motion clarity, especially older VA panels can have trouble with grey to grey transitions leading to "smearing".

LED:

• still able to appear bright in bright environments (not much of a factor if you watch in dark rooms, as both technologies should get bright enough once your eyes are adjusted to the dark)

So yes, your findings do not surprise me.

The differences matter less to my viewing pleasure than I though, so long as the overall brightness is limited on the LED. No matter how dark the viewing environment is, sunlit scenes need a lot of brightness to look good to me. The eye can adjust for contrast but there's a different feel to everything when your vision is set to very low light. Dim sunlight looks surreal, like you're in a dream, or wearing dark sunglasses. 1000 nits sustained brightness is enough to mitigate that effect to a large degree. I'd rather sit in a medium-low light environment and let the TV be a window into a world that is darker at times and much brighter at times than the environment I'm viewing from. It's just a much more pleasant experience. Similarly, the best clarity you'll hear is in an anechoic room. But it's not the most comfortable and pleasant listening experience for most.
The only time I might go for a really dark viewing environment is when the screen is taking up the majority of my field of view, perhaps a huge wrap around screen. Similarly, I might be more open to a very anechoic listening environment with a full surround sound system. But even then, the dark or anechoic room is just not a particularly comfortable environment, so I'm going to prefer some natural lighting and acoustic ambience in any room that' not an AV laboratory.

 

[Rednaxela]

Replying to @pseudoid on his thoughts posted elsewhere about the Sony X90L series. Follow the first quote for full context.

What type of content are you going to watch on it?

Regular TV, Spotify, and occasionally, say 5-10 times a year, some cinematic content.

There is a reason for implementing 'local (zoned) dimming':

• It is a band-aid fix for the shortcomings of LCD panels... and you get the 'blooming' effect as a "feature, for free.

I know that there will be a price tag to anything you buy that is totally @ your discretion but may I remind you that what ever you purchase today most likely (and hopefully) will be with you for at least 5years?

Definitely. Agree with both points.

How does that specific TV compare compare with any of Sony's OLED offerings?

This is Sony’s 2023 lineup in a nutshell.

Screenshot from Youtube, can’t remember which video/channel, maybe you recognise the logo.

Reluctant about OLED mostly because of Spotify, which I use a lot but is really quite static. Even though it now has a night mode which is very neat, burn in still worries me.

I am guessing that will be about $300(USD) cost increase but could be justified.

Most certainly. I do like a good picture, accurate colours etc. and don’t mind paying a bit more for it. The biggest jump I expect to make compared to my current 2015 Sony 49X8305 LED TV will be in the black levels and user interface speed. I’m aware of the local blooming effect but it may still be better than the global blooming I have now if this makes sense. It may not even bother me at all in practice. There’s an amazing LED offering above it in the X95L series, which is a step up in most every regard including dimming zone count. Its smallest available size is 65” however. I have tried very hard to (virtually) fit it in the living room but so far no success. Would have certainly gone for that otherwise.

Thank you for your comments. All great points and much appreciated.

 

[Keith_W]

I walked in to an A/V showroom while I was waiting for my wife. I saw a Sony A80L and A95L side by side. There was an obvious difference in colour saturation and sharpness between the two panels, and this was before I looked closer and saw the model number and the price. Both are OLED 4K panels, but the A95L was QD-OLED.

Question: are TV's like speakers in that you can implement correction and get it close to correct? Is it possible to adjust saturation and sharpness so that the two panels are closer, or even the same?

 

[EJ3]

Still a good portion earn well, making a 55 or 65 inch oled a good compromise among videophiles ( if there even is such a thing).

Of course there is: you can de-phile anything!

 

[EJ3]

well, yes and no
i admit, last gen 48" was a bit much, but the newer 42" is ok if you're not actually glued to the screen (been using a LG C2 for a few months now, distanced some 60-80cm from it)

regarding plasmas, i still have my old panasonic 42" and there's simply no comparison, at least not after a decade of the plasmas use which got a lot dimmer and washed out in that time, not to mention it was unuseable for use with a pc (besides watching movies)

my main tv is still a 55" '18 or was '19 Sony model with a VA screen, got it for half the price 3 years back and it's enough for my watching habits

My mothers 48" Panasonic PLASMA bought new in 2014, I think, still has one of the greatest video screens that I have seen. Perhaps because it isn't on for more than a couple of hours a day) And the audio is pretty good, too.
I & my wife, have not owned a TV since 2007.
That will likely change by April.

 

[Rednaxela]

Question: are TV's like speakers in that you can implement correction and get it close to correct?

TV calibration certainly is a thing.

Is it possible to adjust saturation and sharpness so that the two panels are closer, or even the same?

Closer probably yes. In some cases maybe so close that they’d be hard to tell apart. But this would heavily depend on the panels compared.

 

[Keith_W]

TV calibration certainly is a thing.


Closer probably yes. In some cases maybe so close that they’d be hard to tell apart. But this would heavily depend on the panels compared.

Yes, I was wondering if I was looking at the out of the box settings which was accounting for all the difference. I would love to see an objective test between calibrated panels to see if measurements are the same, and whether they can be picked apart in a blind test.

 

[Chrise36]

I walked in to an A/V showroom while I was waiting for my wife. I saw a Sony A80L and A95L side by side. There was an obvious difference in colour saturation and sharpness between the two panels, and this was before I looked closer and saw the model number and the price. Both are OLED 4K panels, but the A95L was QD-OLED.

Question: are TV's like speakers in that you can implement correction and get it close to correct? Is it possible to adjust saturation and sharpness so that the two panels are closer, or even the same?

As mentioned calibration can work up to a point but the woled will look a bit different at the end.

 

[Rednaxela]

Screenshot from Youtube, can’t remember which video/channel, maybe you recognise the logo.

Found it.

 

[Rednaxela]

Yes, I was wondering if I was looking at the out of the box settings which was accounting for all the difference. I would love to see an objective test between calibrated panels to see if measurements are the same, and whether they can be picked apart in a blind test.

Something like this?

 

[Aerith Gainsborough]

Question: are TV's like speakers in that you can implement correction and get it close to correct? Is it possible to adjust saturation and sharpness so that the two panels are closer, or even the same?

If you only refer to color / white point calibration: yes. Though external hardware is needed that may be more expensive than the TV itself.

What you absolutely cannot "correct" for are the inherent weaknesses of each tech like blooming around bright objects on full array LEDs or limited brightness on OLEDs. There you need to be aware of your practical needs and pick the lesser evil, so to speak.

Also keep in mind that unlike speakers, a lot of TVs have picture processing going on that may or may not be able to be deactivated and can throw a spanner into the works. :S

 

[holdingpants01]

interesting to see how dim and bland plasma tvs seems now, yet they were SOTA at their time