That's pretty cool - I hope they bring these out soon!
But, since I hate incorrect terminology, this is wrong:
4K = 3840×2160
Some TV manufacturers incorrectly use the term 4K like this, but the way it's been used for the last decade in the film industry (and is specified in standards like the DCI spec) is that 4K refers to images with a horizontal resolution of 4096 pixels [1] (the number of pixels vertically changes to accommodate different aspect ratios - the pixels are always square).
The TV industry is using 3840 because it has four times the resolution on 1080p. I don't think there are any specifications for this resolution, so there's no official term. I've heard it called QuadHD or 4xHD though.
1. There is one exception, which is 3996x2160 (1.89:1 aspect) but all the others are 4096.
All this Retina talk makes me sick. Apple doesn't have this technology under control. They screwed up.
My first Retina screen had horrible image persistence problems. The replacement from Apple now has the same problems. Coincidence? In the Munich Apple Store I had the opportunity to check four machines (of the seven they had standing around), all four had the exact same problem.
Apple can't build them now, they really shouldn't try to build more of them.
I have one and it's completely fine. I saw quite a few, between WWDC and (too many) trips to my local Apple Store, and I find it hard to believe that the flagship model of a large multinational computer company is fundamentally broken.
You might have a Samsung screen. All the screens I tested were LG screens.
The test is quite easy but you wouldn’t notice the defect just like that. Change the background to the dark grey, open up Safari, leave it in one place for two minutes, hide the window. The outline of the address bar was very clearly recognisable on all four machines I tested in the Munich Apple Store.
It’s a bit harder to see on colourful background and you need more than two minutes, but it’s definitely also visible there.
With my replacement the image retention defect also only showed up after one week of daily use.
I think it’s very likely that a large number (likely quite a few more than one percent) are affected by this problem. I think it’s exceedingly unlikely that this is not a problem with shoddy quality control.
So first you say these display got "horrible image persistence problems" and then you say "you wouldn't notice it just like that". So can't be that big of a problem? I have one with a LG panel, and the only time I noticed some ghosting is when someone on the internet told me to turn my screen completely gray - otherwise I wouldn't even know about it.
Yep. I'm on my 2nd display with that problem. The thing is, when I replaced the first one in store, I (along with my father and the Apple employee) scrutinized the screen to see if it had image persistence. It did not. It does now, which indicates that the problem developed over time. Both screens have been LG -- I think all LG screens will eventually develop the problem to some degree. It's not noticeable except under certain circumstances (light to dark backgrounds).
Just tried it on my clearly LG panel. Did not have the problem you mentioned. I've had the machine for a week, from CDW running 10.8.1 so perhaps they've figured it out?
All of a sudden, @jzw's iPad port of xscreensaver turns from "geek-joke" into "prescient foresight"… (assuming the iPad/iPhone retina displays exhibit similar defects…)
Why is that exceedingly unlikely? All the display units were probably from the same manufacturing batch, and may have even been among the earliest manufactured (which could have an effect on defect rates).
As I see it this seems to be a massive issue Apple does not have under control. I mean, if it were a problem with a bad batch, why was my replacement screen also affected?
I can only work with what I have access to. And what I have access to makes the situation look very dire indeed.
It's obviously a problem. The question is whether you're overestimating the chance of a single person receiving a defective unit, a defective replacement, and seeing defective units at an Apple store. Considering how many people go to Apple stores and buy the computers, even an extremely low defect rate could explain your experience.
Oh, man. That old bullshit again. Beyond a certain point that just doesn’t work. If you think about it. Also seeing how my experience is shared with many other people who also received defect units and saw several defect units in different Apple Stores.
The employees in the Apple Store know about the issue even before I arrived, the Apple Care guy I talked to knew about the issue before I brought it up (there is just no word yet from higher up) … this is very likely a big fucking deal. Antennagate shit, only this time with an actual defect.
I've got a perfectly good retina screen, and I looked at all the screens in my local apple store, and they are all good.
Thus all Retina screens are fine, and any defects must be a minor problem.
Alternatively I could realise that my local store make up a tiny % of the global market and I can't make generalisations without a more substantial sample.
I get you aren't happy you got a bad screen. But the sky isn't falling.
Again, if there are any defects, the problem is real. My concern, which I think you have confirmed, is that your misunderstanding statistics led you to wrongly conclude from your personal experiences that the defect rate is higher than it may actually be.
I agree with arrrg. I don't believe it's a bad batch issue at all. I bet I could demonstrate the problem to just about anyone who has had a MBPR for at least a month and who has an LG screen (Samsung screens do not seem to have the problem.)
Look, it’s unlikely that this has anything to do with a bad batch. People are still getting defect machines all the time, the week in which they were built don’t seem to play a role.
From 2000 through 2005 I purchased four LCD screens. Not a single one came without either (A) dark pixels, or (B) pixels that were "Always on" - in one case I had three dead pixels. Dell and Apple (vendors of these screens) indicated that the failure rate/location of the bad pixels was low enough, that they wouldn't take a return. I lived with it.
I find it interesting that since 2006, the three Screens I've purchased have had 0 bad pixels. And, in general, people are reporting significantly fewer LCDs with bad pixels (either always on or always off).
With regards to your Mac - was it within the first 30 days? Is Apple now allowing returns for a single bad pixel?
One nice thing about the Retina Display MBPro - a single dead pixel is probably irrelevant - you'll never notice.
I'd suspect two causes behind this. 1) as the manufacturing experience increases, the defect rate will (probably) drop. And 2) there is now an established market for minorly defected LCD panels in cut-price no-brand-name monitors, so panels sold to brand names can afford to have QA tightened up more by selling marginal panels into the secondary supply chain instead of just shipping or junking them, allowing the "brand name ship" cutoff point to be somewhat higher.
I think it is valid to point out the flaws of a product which is regarded as market-leading. For example: why aren't other companies offering similar products? Could in part be because of problems such as this.
If we can't be a little honest about a company that is (I believe) bigger than Wal-mart then I don't know what company WOULD be strong enough to withstand mild criticism.
I’m sorry, but selling a product with a defect like that is just unacceptable. There are no ifs and buts about that. That’s just how it is.
It would be great if they could fix this issue. I would be overjoyed. However, they seem to be unable to, now even going so far as talking the defect down.
>>I’m sorry, but selling a product with a defect like that is just unacceptable. There are no ifs and buts about that. That’s just how it is.
Unacceptable to who? The products in question are selling by the millions. Clearly, the market finds them very acceptable. In the grand scheme of things, what you think is largely irrelevant.
arrrg obviously has an ax to grind about Apple in general or just has some irrational, intense hatred for retina screens. From the replies I've seen, reasonable discussion is out of the question. "Illegal"? Please.
Ignoring the defect screen, I love everything about the machine. I think it’s pretty plainly the best laptop on the planet right now (at least for me personally). I would much rather Apple fix this than having to give back my MBPr. But I can't accept a defect screen. It's quite simple.
CRT burn-in never got classified as a "contract of sale voiding defect".
Is the retina display "defect" a significantly different thing? (I can't tell from any of the commentary I've heard whether its a "permanent, irreversible screen change", or a temporary "ghosting" of previous display only persistent over a few second/minute/hours timescale.)
I'm wondering the same thing. From what I see described it's the same kind of thing that some very early LCD screens had 10-15 years ago, and might even be tolerable if you use something like LCDScrub[1] to fight it regularly. Xscreensaver has a work-alike in it's set also that i've used before on some very old screens that had problems.
What the hell? How could that be even remotely acceptable? It plainly is not.
I mean, Apple seems to agree, at least for now. Every single Apple employee I talked to immediately recognised it as an obvious and unacceptable defect. (The one employee I showed it to first exclaimed “Holy shit!”, I didn’t even have to say anything beyond “Could you please hide the Safari window?”) And that’s what it is.
Nobody is saying you have to live with it. Of course Apple will exchange it with a new one for free. What you seemed to be saying in your original comment is that there should not be defects, period. Which is pretty much a pipe dream when it comes to any product ever made.
I’m not crying. I accept that something like this is a possibility. Fixing it, however, is not my responsibility and I will just declare the sales contract null and void if Apple continues to be unable to fix this issue. It’s quite simple and I have the right.
And it’s a damn shame, actually. In every other respect the MBPr is an awesome machine. (Also, I never had any problems with first gen Apple products. All through the last ten years Apple has been excellent in releasing pretty great first gen models.)
I'm curious as to why OP keeps insisting that current Thunderbolt ports can't support a QFHD (3840 x 2160) display.
Thunderbolt layers a few lanes of PCIe with a DisplayPort 1.2 signal. DisplayPort 1.2 supports QFHD at 30 bits/pixel and 60Hz.
Of course you'd be limited to one display per port, and I'm not sure whether that much bandwidth on the DisplayPort lanes would crowd out the PCIe traffic.
Alternatively (as anyone who bought Macs in the 90's can attest), Apple has never been particularly reluctant to introduce a new interconnect when it was convenient for them.
Obviously if they simply pixel-doubled their current 27" panel it would be far beyond Thunderbolt's capabilities, which may be where the "Thunderbolt can't support Retina resolutions" idea originated.
In theory the display signal would need 12.8 Gbps leaving 7.2 Gbps for PCIe traffic, so it's not obviously impossible. It's not clear whether Thunderbolt controllers can mux and demux over 10 Gbps worth of DP though.
1. They only name important features.
2. They give the spotlight to only one new feature at a time and they make it the cornerstone of their marketing efforts.
3. They are consistent across their entire product line.
That why almost everybody knows the word unibody. Now everyone knows the word retina.
It's genuinely impressive how they successfully put together all the things that you mention and at least a handful more while others juggle incredibly cumbersome, abstract names.
The power of these things seem obvious to me, but the fact that only Apple pulls it off suggests that I'm nuts.
I don't have a problem with Apple inventing a term for "high enough PPI". In fact, we probably needed such a term to raise awareness about the importance of PPI. I think the problem is, Apple has the monopoly for using the term retina (they applied for a trademark which will probably be granted). As a result, Sony Xperia S has a higher PPI value than iPhone 4S but since it can't/doesn't market its display as retina, consumers may be misguided into thinking iPhone 4S has a sharper display. If we simply called them both "high-PPI displays", the consumer is more likely to become aware of the fact that retina is not a technology and it is trivial to determine which of these displays is sharper by simply comparing their PPI values.
obvious solution is obvious: Sony just needs to market their phone as having a mega-ultra-hyper-clarity display, and then make a side by side comparison chart where they list the iPhone as only having a mega-clarity display.
Your choice of Sony is ironic, because Sony names every damn feature on their products. I could replace "retina" in your comment with "MotionFlow" and complain that Sony is tricking consumers into buying their TVs.
But at least us, as people who understand retina simply means "high enough PPI", should not argue about whether a particular display qualifies as retina or not. Because it is subjective to decide how much PPI is enough for each type of product and there is no actual numerical barrier that we can agree upon. It is meaningless to discuss whether Galaxy Nexus display (315 ppi) qualifies as retina. It is simply a 315-ppi display.
I'd argue that while the word Retina happens to be a marketing term, in this context it is just as relavant. What we're really discussing is "do the pixels disappear?" - the reason they used the term retina display is that they are indistinguishable to your retina.
"315ppi" or "super high ppi" is meaningless to me even as a techie because for me the gold standard for text clarity is 600dpi on laser printed paper, so even though it's adequate for the screen, it wouldn't impress me unless I saw it.
"Looks like glossy magazine print" makes sense to any consumer. Again, yes in this case Retina happens to coincide with the trademark, but it's also a functional property, one that's relavant to the discussion.
I don't know. If Lenovo, Motorola and Microsoft had teamed up and said "HiDPI is the future, we're putting it on our flagship devices and making it a primary feature of Windows". If they had done that, and implemented it well (eg, most people say it's gorgeous), then they could have named it "Windows ThinkDensity MotoDisplay Technology 8" and that is more or less what we'd be calling it.
Point is, it may be something that is conceptually simple and technologically practical but if company X bets their flagship product on the tech and ends up being the driving force in the mainstream adoption of said technology, then I think they get to brand it.
I believe a big part of "Retina" working is the simple, flowing, suggestive name.
Sony achieved this with "Walkman".
Motorola got it with "RAZR".
Samsung is right there with "Note", and "Galaxy" would be a great qualifier but has become a bit diluted by being stamped on everything. (In my opinion, Galaxy should have been transitioned to the ultimate qualifier to let people know that it's a premier, high-end product.)
Stuff like "Super AMOLED Plus" and your "Windows ThinkDensity MotoDisplay Technology 8" there is madness.
Microsoft actually did put a lot of high-DPI work into Vista, but PC makers and app developers completely ignored it. An advantage of Apple's vertically-integrated model, I guess.
Interesting math, 3840 × 2160 at 32 bpp (ARGB32) is just about 32MB per frame (3840 * 2160 * 4) / (1024 * 1024). At 60 fps that is 1.92GB/second of pixel churn. Perhaps I'm just getting too old but to me that is an impressive amount of data to throw around.
I just did those same calculations and got the same somewhat boggling numbers (actually, I only assume 24bit pixels at the panel and got ~24MB frames).
I then wondered how the OS & Window Manager & Applications & video driver & memory bus & pci bus & video card deal with the respective parts of their chunks of a data stream that big… (Then managed to realize I've got work to do before heading down the google-research-of-interesting-thing rabbit hole…)
But, since I hate incorrect terminology, this is wrong:
Some TV manufacturers incorrectly use the term 4K like this, but the way it's been used for the last decade in the film industry (and is specified in standards like the DCI spec) is that 4K refers to images with a horizontal resolution of 4096 pixels [1] (the number of pixels vertically changes to accommodate different aspect ratios - the pixels are always square).The TV industry is using 3840 because it has four times the resolution on 1080p. I don't think there are any specifications for this resolution, so there's no official term. I've heard it called QuadHD or 4xHD though.
1. There is one exception, which is 3996x2160 (1.89:1 aspect) but all the others are 4096.