Categories
Uncategorized

How Apple Destroys OLED

$2000 and the MacBook Pro doesn’t even have and OLED Display. Never mind the $1100 iPad Pro. Why? Because OLED is… hella problematic. It’s expensive, leading edge supplies are limited, it allows for always-on but it’s prone to burn in, smearing, and off-axis color shifting. it decays unevenly, and while it can be flexible and provides the deepest, truest blacks, it struggles with peak brightness. Like hitting that subscribe button so we can build the best community in tech… brightness.

So, what do you do? I mean, besides hitting that button and cursing in the comments. Well, if you’re Apple, you power word kill regular old LCD, just obliterate it from the premium lineup. Then you divide and conquer. Build up OLED on the smaller devices while hedging with mini-LED on the bigger ones. Then… then you make them fight it out, like Kaiju and Jaegers… or just wait for micro-LED to mature and extinction level event both of them.

And the way that works is… well, I’m not going to go all the way back to when Apple switched from CRT, cathode ray tube, to LCD, liquid crystal display to begin with. Basically switched from the OG TV set iMac to the Pixar Lamp iMac. Not today Satan. Just to the original iPhone and it’s 3.5-inch LCD display. 3.5-inches not because Apple cared about single-handed ease of use or anything like that, not yet, but because 3.5-inches was literally the biggest LCD display Apple could get for the iPhone back then.

Over time, they increased density, first with 2x Retina on the iPhone 4, then height, with the 4-inch iPhone 5, then size in general with the 4.7- and 5.5-inch iPhone 6 and iPhone 6 Plus. But in all that time, 2007 to 2014, just over 7 years, Apple never switched the iPhone from LCD to OLED. Even though Samsung began getting into it around 2009. Even though the Apple Watch, Apple’s own Watch, was introduced with an OLED display in 2014.

Why? Because OLED isn’t really just OLED. Well, it is, of course, but it’s so damn finicky different suppliers have developed different ways to handle it, some based entirely on how much OLED they have to handle. Electronics with OLED screens that are typically at arms length use entirely different processes than TV sets with OLED screens that are typically across the room. LG OLED TVs, for example, emit all white light then use filters to create the colors. That’s not only cheaper when you’re dealing with comparatively huge TV-sized displays, but it means no one color decays faster than any other. Something that’s way more problematic on smaller OLED displays.

But for the Apple Watch, battery life was so limited and the screen sizes so small, Apple felt they had no choice but to go with OLED. Specifically, LG OLED in an RGB stripe. Not the classic with LCD version of RGB stripe, where each pixel is literally divided up into a line of equal-sized red, green, and blue sub-pixels, but for reasons I’ll get in a blue decaying minute, unequal sized and shaped red and green squares and a long, thin blue line.

All this is because OLED can be more power-efficient than LCD, because OLED, technically AMOLED these days, generates its own light. Active Matrix Organic Light Emitting Diode. It’s right there on the label. And LCD… does not. It needs a backlight. Typically an LED, or light emitting diode backlight. The non-organic kind. Insert your species 8472 joke here. The backlight adds mass to the device, sure, but in standard LCD implementations, it’s also all or nothing. All on or all off. The whole thing. Which is why it consumes a fairly consistent amount of power. Also why true blacks are almost impossible — the backlight leaks worse than the iPhone supply chain.

With OLED, since it generates its own light, off pixels are off and on pixels are on. Basically it’s every pixel for itself. And, the less pixels that are on, the less power they draw. So, with a mostly black interface like the Apple Watch, you mostly get a hella lot of power savings.

Smaller displays also inherently have higher yield because, even if there’s a defect or several, you’re not throwing out 13-inches of tablet or laptop display, or 6-inches of phone display. You’re still chopping out a bunch or a ton of totally usable 1-inch displays from that same sheet. And with RGB stripe, even if it causes some color aberration, like aliasing around high-contrast edges, you’re getting a pixel density pretty much equivalent to LCD, which means it’s still sharp, crisp, and bright. Even if it’s also still way more susceptible to burn in and color shifting.

So, yeah, let’s get into that. Basically, because with OLED, the individual pixels light up, those pixels also burn out… individually. Decay, actually. So if you leave something like a bright logo on screen for a ridiculous amount of time, the pixels showing that logo will decay faster than the ones that aren’t. But, Shyalam-style plot-twist, the blue OLED sub-pixels also decay faster than the green and red, making those pixels not only dimmer but… yellower. And because it’s only those pixels, if you switch to something else on the screen, that logo will still persist, you’ll still be haunted by the dimmer, yellower ghost of its decayed pixels.

On a watch screen, where you’re not lighting it up anywhere nearly as much or as often as phone or TV, it was a smart tradeoff. Especially with the kind mitigations I’ll get to in a minute.

For phones, though, because sharpness and brightness weren’t considered as critically important, but blue sub pixel decay was kind of a deal break, they went with a diamond-shaped sub-pixel arrangement. What Samsung calls Pentile. That puts a giant blue sub-pixel in the middle and surrounds it with two much smaller red and two green sub-pixels. Sure, you get a lower effective sub-pixel density and brightness, but that giant blue sub-pixel is going to last you. Plus the lower density made them cheaper to produce than RGB stripe of the same physical size. Which became increasingly important as phones became increasingly bigger.

But burn-in is only one of OLED’s hot messes. There’s also off-axis color shifting. Basically, the thickness of the OLED layer changes when you look at it from an angle, causing it to desaturate and take on a red or blue tint. Because of the response time, or the time it takes to go from off to full on — black to white — moving bright text or graphics across a dark background can cause a smearing effect. And there used to be graininess to lower density OLED as well, especially in the mid-tones. Exasperated by no one else really seeming to care about color calibration or color management back then either. So OLED could look… hella garish. Boosting sat like a first year Instagram clout chaser or something.

And Apple just didn’t think those trade-offs were worth it for the iPhone. Not back then. Because if there are two things that dictate Apple display moves, it’s fussiness and scale. Fussiness, because there was just no way Apple was going to slap OLED on an iPhone until they could get panels that were high enough quality for their display team to start working on. And scale, because there was just no way to get enough quantity of that quality… not for hundreds of millions of iPhones, not for years.

Which has been a recurring issue for Apple. It’s why they’re sometimes lagged behind in screen technology more than a few times over the years. Like when Amazon got higher contrast TFT LCD into their tablets before their was enough supply for the iPad. Or the Apple Watch, even some other phones, got LTPO OLED for adaptive refresh before there was enough supply for the iPhone. It was easy to make those leading edge panels in small batches. It was impossible to make enough for hundreds of millions of iPhones.

See, Apple may contract Samsung’s OLED process for their displays, like they’ll contract TSMC’s 5 nanometer process for their chipsets. But Apple’s doesn’t do anything commodity or off the shelf. Everything is custom designed. And Apple Display is every bit as exacting and annoying about how Samsung fabs their OLED panels as they are how TSMC fabs their chips. Maybe more so even, since they’ll ask for completely different materials than Samsung is using for Galaxy panels on the same process, for example. And then they’ll take those panels and implement everything else, including all their own mitigations, like machine learning processes that study how you’re using your display so it can micromanage exact positioning and brightness levels to greatly reduce burn-in. Implement their own filtering to effectively eliminate off-axis red shift, though not entirely blue. Factory calibrate and completely manage the color system.

And because OLED can be fabbed on flexible plastic substrates, they can be flexible. Apple used that to fold the display back on itself, so they didn’t have to stuff the drivers into the bottom of the phone, creating a big bezel beard of a chin. Hell, they even built a dedicated OLED display engines into their A-series chipsets to control the whole damn chain at the silicon level.

It was at that point that Apple decided OLED was mature enough for the iPhone. But the supply of leading edge OLED was still so constrained, and the price so high, Apple only implemented it in one iPhone. The first $1000 iPhone. The 2017 iPhone X. At least at first. By 2020, Apple there was enough supply for Apple to go OLED across the entire iPhone 12 product line. Though that and 5G still resulted in a price bump.

Same thing with 120Hz adaptive refresh. Earlier implementations forced people to manually select lower resolutions, or abandoned color management, or lost high refresh rates at low brightness levels. And Apple hated all of that, so they waited until the quality they wanted was available in the quantity they needed.

Specifically in for this, the same LTPO OLED they’d used for adaptive refresh on the Apple Watch’s always-on display. It just took until 2021 before they could get enough just for the iPhone 13 Pro to implement up to 120Hz ProMotion Adaptive Refresh. Something they’d done with IGZO LCD on the iPad Pro back in 2017. Right when they were switching the iPhone to OLED to begin with.

And that’s just exactly why we don’t have OLED on the iPad or MacBooks yet either. On top of everything else — on top of mitigation mountain — they also have trouble with consistent brightness. With every pixel lighting up at the exact same level as every other pixel. That’s not a showstopper on phone-sized panels, but it can really stand out on bigger tablet and laptop sized panels. Like the snow on Hoth just looks kinda splotchy. Also, peak brightness, because watching HDR on a phone-sized display and battery is one thing, producing HDR on a tablet or laptop-sized screen is a complete pro workflow another.

And that’s just something not LG, not Samsung, and not BOE… who… liberated a previous generation Samsung process years ago and has been working to catch up and undercut the market ever since, have so far been able to solve. I mean, Samsung uses S-Stripe OLED for their tablets, which is similar to what the Apple Watch uses, but that’s not something that scales in price or yield to… Apple scale. Especially not LTPO OLED, which Apple would need at iPad Pro and MacBook Pro size and quantity to maintain that 120Hz adaptive refresh functionality.

That’s the whole entire reason Apple went another way for the 12.9-inch 2021 iPad Pro and the 14- and 16-inch MacBook Pros — Mini LED.

Now, Mini LED is still like LCD in that it’s true RGB stripe and requires an LED back light, but instead of one giant backlight to rule them all… them all being the pixels all. It has way smaller LEDs, like 10,000 of them, and 2500 local dimming zones, each of which can turn on and off independently. So, light control isn’t granular to the sub-pixel level like OLED, but it’s a jump to Lightspeed beyond traditional all-or-nothing LED. And it can also get super, super bright.

But… yeah, like social media influencers, there’s always a but… because of those local dimming zones, there can be a blooming or haloing effect where bright areas still bleed out over the dark. And, at least in Apple’s implementation on the MacBook Pros, the same type of smearing as OLED, maybe worse, as the black to white response times can’t keep up with the refresh rate, so you can see trailing on white text on a black background, for example.

Now, there have been a bunch of reports about Apple switching mini-LED out for OLED on the iPads and MacBooks Pro… a year or several from now at least. But it’s tough to tell how much of that is real and how much is LG and Samsung using local industry rags to posture in public.

Realistically, it’s just a question of whether Mini LED can get better to the point Apple stops flirting with OLED for those size displays, or OLED can get to a high enough quality, in sufficient quantity, that Apple decides to go all-in and swap it in.

So, for now, Apple seems content to leave them both in the gladiator pits of Sakaar and let them fight it out.

If I were a bustabit betting man, my money would be on OLED coming back late in the 3rd round and just Hulk-smashing mini-LED out of the lineup. Even despite it being so damn mitigation thirsty. But also, that a new challenger will enter the pixel combat ring. Yup, Micro-LED.

You can kinda sorta think of MicroLED being to OLED what mini-LED is to LCD. In other words, a fresher, newer, better looking younger sibling that basically just fixes almost all the problems of the OG.

Specifically, MicroLED offers pixel-level light emission, like OLED, but it’s not organic, like OLED, so it won’t be plagued by burn in or premature blue sub-pixel decay, like OLED. But since it doesn’t even need local dimming zones, like Mini-LED, it also won’t suffer from blooming, like Mini LED.

And, according to reports, Apple’s had hundreds of engineers working on microLED for almost a half-a-decade now, doing what they did with chipsets, and maybe modems next — making way, way more of the process internal. And if past is at all prologue, they’ll do with MicroLED what they did with OLED, and that’s start with the Apple Watch and then slowly expand — literally — to the iPhone, and then on from there as quality and quantity allow.

And what’ll be super exciting about this — I mean, if it works and doesn’t just get AirPower’d along the way — is that Apple will finally own their whole entire display pipeline. Not just from engine to design, but fully from transistor to pixel. Silicon to screen. And that’s one of those highly differentiated experiences involving complete control of key technologies that makes Apple, Apple.