rMBP display very yellow

[kfscoll]

Post-calibration results - rMBP vs. cMBP

As promised earlier, here are the post-calibration uniformity results of my rMBP and my early-2011 cMBP (with hi-res anti-glare display). I calibrated both displays to 6500°K and 100 cd/m2.

First up, here are the rMBP's results:

Second, here are the cMBP's results:

Notice the maximum luminance delta of 9 cd/m2 for the rMBP and 18 cd/m2 for the cMBP. The maximum white point delta for the rMBP is 153K and for the cMBP it's 187K.

So, what this means is that my "shoddy" LG rMBP display has better brightness and color temperature uniformity than my older MacBook Pro's 15" anti-glare display. I think maybe I thought the cMBP's uniformity was subjectively better because its color uniformity is symmetric; in other words it's warmer in both lower corners of the display whereas the rMBP is cooler in one corner and dead-nuts on in the other corner (the corner, incidentally, that I thought was the most yellow).

Of course, the calibration tool can't take measurements in the extreme corners of the display so the delta values could be worse (or better) there. Still, this is pretty solid objective evidence that, in least in my case, despite what my eyes tell me, the uniformity of my rMBP's display is pretty significantly superior to my cMBP's display.

Interesting note -- I guarantee that most people would consider the post calibration results on all of my monitors (iMac included) to be "yellow". What's also interesting is that the post-calibration color temperature of my rMBP didn't change much from the default display profile. In other words, based on anecdotal evidence, the LG displays aren't too yellow; the Samsung displays are too blue.

Of course, all of this matters not a whit if you have a display with a color gradient. Calibration won't help you at all.

 

[terraphantm]

As promised earlier, here are the post-calibration uniformity results of my rMBP and my early-2011 cMBP (with hi-res anti-glare display). I calibrated both displays to 6500°K and 100 cd/m2.

First up, here are the rMBP's results:
Image

Image

Second, here are the cMBP's results:
Image

Image

Notice the maximum luminance delta of 9 cd/m2 for the rMBP and 18 cd/m2 for the cMBP. The maximum white point delta for the rMBP is 153K and for the cMBP it's 187K.

So, what this means is that my "shoddy" LG rMBP display has better brightness and color temperature uniformity than my older MacBook Pro's 15" anti-glare display. I think maybe I thought the cMBP's uniformity was subjectively better because its color uniformity is symmetric; in other words it's warmer in both lower corners of the display whereas the rMBP is cooler in one corner and dead-nuts on in the other corner (the corner, incidentally, that I thought was the most yellow).

Of course, the calibration tool can't take measurements in the extreme corners of the display so the delta values could be worse (or better) there. Still, this is pretty solid objective evidence that, in least in my case, despite what my eyes tell me, the uniformity of my rMBP's display is pretty significantly superior to my cMBP's display.

Interesting note -- I guarantee that most people would consider the post calibration results on all of my monitors (iMac included) to be "yellow". What's also interesting is that the post-calibration color temperature of my rMBP didn't change much from the default display profile. In other words, based on anecdotal evidence, the LG displays aren't too yellow; the Samsung displays are too blue.

Of course, all of this matters not a whit if you have a display with a color gradient. Calibration won't help you at all.

Hmm, interesting data. Makes me feel a bit better about my MBP. Subjectively, I'd say my color and luminosity variances are similar.

I agree about the color tempeature - this rMBP is much closer to my calibrated NEC monitor than pretty much any other laptop I've used. So while that's technically yellower, 6500K is still pretty darn white.

I'd be interested in seeing a similar calibration done on one of the samsung displays.

 

[ilgreatluigi]

Yeah, I actually didn't notice that until you mentioned it, but now I can see it, especially the yellow in the bottom-right corner. The colors in the photos themselves look a little blown out, so it's maybe a little harder to see details compared to the pictures taken by "cplusd".


Hmmm...


I could be jumping to conclusions at this point, but as I pointed out in a post to the other thread I linked to earlier, I also visited an Apple store, and saw 4 display laptops: 3 LG, 1 Samsung, and the Samsung was noticeably the best display on the table. This is causing me to lean towards thinking that perhaps all of the color gradient displays are LG. The only anomaly so far is that the OP of the other thread was complaining about a dark spot in the "lower-left" corner of his display, which later turned out to be Samsung. But I went back and read it again, and he specifically mentioned that the corner was just slightly darker and definitely *not* "yellower" than the rest of the display.


The "it's just the glue still curing, guys" people kill me more than the calibration people do. That whole myth started back when the iPhone 4 with its new Retina display was first introduced. Some people were seeing distinct and quite small yellow spots in various places on their phone displays that disappeared after just a few days; example pics here. The so-called "glue" is clearly not what is causing either the entire display's white-point to be overly warm or the white-point gradient issue. That is a totally separate issue, but because it included the words "yellow" and "display", people latched onto it and figured it was a reasonable explanation for every possible issue with the display that had anything whatsoever to do with yellow-ness.

Sigh.

-- Nathan

Ah yes, it must be the make... Not. I'm sure if there was a huge difference between the screens, apple QA would have spotted it. What makes you an expert on the issue? I have credible newsworthy sources and you're presenting an amateur opinion.

 

[kfscoll]

Ah yes, it must be the make... Not. I'm sure if there was a huge difference between the screens, apple QA would have spotted it. What makes you an expert on the issue? I have credible newsworthy sources and you're presenting an amateur opinion.

Credible sources? I'll believe that once you name them. Besides, this yellow "issue" looks nothing like the iPhone's yellow glue issue.

It isn't the glue. Differences between LG's and Samsung's manufacturing processes causing the differences between the two manufacturers' displays is a much more plausible root cause than "the glue's not dry." But hey, if I'm wrong, I'll eat my rMBP's LG screen.

 

[terraphantm]

I think this display might have an A-TW polarizer, which would also explain the slight green and redish hues seen in the corners.

The NEC in the above picture has an A-TW polarizer, while the HP doesn't. Now look at the picture of my rMBP vs another IPS LCD below:

similar difference IMO.

As I mentioned, the one draw back to the A-TW polarizers was that they caused the LCDs to exhibit a slight greenish or redish hue depending on the angle. So when looking at them dead on, the corners would end up exhibiting such a hue.

 

[kfscoll]

I think this display might have an A-TW polarizer, which would also explain the slight green and redish hues seen in the corners.

[image]

The NEC in the above picture has an A-TW polarizer, while the HP doesn't. Now look at the picture of my rMBP vs another IPS LCD below:

[image]

similar difference IMO.

As I mentioned, the one draw back to the A-TW polarizers was that they caused the LCDs to exhibit a slight greenish or redish hue depending on the angle. So when looking at them dead on, the corners would end up exhibiting such a hue.

It wouldn't surprise me. If you turn the display off and shine a light into it, the resulting "flare" is green on one side and red on the other.

 

[skydesigner]

It wouldn't surprise me. If you turn the display off and shine a light into it, the resulting "flare" is green on one side and red on the other.

I just wanted to update my "yellow-tint" problem on my Samsung panel. I think it´s not yellow... it´s just a darker area (lower left) and on white backgrounds it gets a mix of yellow/grey. Viewing angle does affect it a little.

By the way: did someone here in the forum already post a color-profile for the samsung panel? I know those profiles have to be made especially for every panel itself, but I want to see how it affects the whites/ colors overall.

 

[iaymnu]

After 4 exchanges I finally got a Samsung panel. What a difference when viewing anything with a white background. No more "dirty" yellow gradients/spots.

 

[skydesigner]

After 4 exchanges I finally got a Samsung panel. What a difference when viewing anything with a white background. No more "dirty" yellow gradients/spots.

You also have a totally uniform "white"? Lilke I worte, mine has some darker areas in the lower left. Could you check it for me? (switch to fullscreen with the browser)

http://jasonfarrell.com/misc/deadpixeltest.php?p=2

thank you!

 

[mzbe]

Writing this on my next rMBP (another 2.6/512/16), which I received today and am going to return tonight at the Apple store.

Another LG display, LP154WT1-SJA1: Native white point 5200K, slight bleeding, perceptible screen refresh flicker.

 

[AZREOSpecialist]

I'm not looking for any faults but I've just received my RMBP and i'm finding my display is very yellow so i compared it to my old Sony Vaio Z Series and the result is that my Sony Vaio display is significantly whiter and brighter

Does it look yellow if you aren't comparing it with another computer with a less yellow screen?

 

[IGregory]

Received my retina MBP today, and it has a terrible yellow tint compared to my previous MBP.

I called apple and after a long discussion they would agree to replace my machine, however, they said I would have to send mine in first and THEN they would send out a new one.

Unfair, considering its not my fault this machine is faulty.

Anyone dealt with this before?

Standard operating procedure. If Apple sends you a replacement before you return the unit you have, what's to say you won't return the unit in need of repair. I am not saying you would do such a thing, but there are plenty people out there who would. Its just good business practice.

Edit: Of course, if you are near an Apple store you could take it there for an even exchange.

 

[nickbarbs]

lets play a game

It's called guess the manufacturer of the panel in this picture-

is it

A. Samsung

or

B. LG


picture taken with my DSLR, curious to see what you guys think.

Display is all default settings, max brightness.

 

[NathanA]

So, I got my replacement for my unit with an LG (that had the lower-left yellow corner).

It's a Samsung.

Overall white point is markedly improved. HOWEVER, even this display has a darker left corner. I would say that on this panel, it doesn't look so much "yellow" as it does "dingy" or slightly dirty. Of course, the surface of the display is perfectly clean.

Ugh. I'm tried of playing this game, and I don't understand why it's so hard for panel manufacturers to produce panels that are relatively free of these kinds of defects.

Ah yes, it must be the make... Not. I'm sure if there was a huge difference between the screens, apple QA would have spotted it. What makes you an expert on the issue? I have credible newsworthy sources and you're presenting an amateur opinion.

Wow, what crawled up your butt and died? Of course it's an amateur opinion...I never presented it as anything but. It is an opinion, however, supported by experience and the anecdotal evidence of others. This isn't my first time in the Apple display rodeo...

I just wanted to update my "yellow-tint" problem on my Samsung panel. I think it´s not yellow... it´s just a darker area (lower left) and on white backgrounds it gets a mix of yellow/grey. Viewing angle does affect it a little.

My issue matches yours exactly, as well as that of poster Davieis from this thread.

On this new Samsung panel, some colors are more obviously affected by the different brightness/white point of the left side of the screen. White is one, but I think it is more obvious on, say, a Facebook profile Timeline page...that blue-grey color looks greyish-yellowish on the left side and much more blue on the right.

I'm honestly not sure what I plan to do at this point. I could return it and just call the whole thing off. I could keep replacing it until I get an acceptable screen, but I'm losing confidence that this will ever happen. Or I could hold onto it, hope that the issue is sorted in a few months, and then send my computer in to have the screen swapped.

-- Nathan

 

[mzbe]

To put things into perspective: Every rMBP I bought had a significantly worse display (more bleeding, uneven brightness, white point, color shifts) than my late 2008 MBP.

Putting both the old and the new Mac side by side, I could not stand the LG panels in the rMBP. No way.

BTW the quality of displays has subjectively declined (!) every time I got a new replacement in. It will be a small wonder if the next one I get will have any brightness or show any other color besides yellow.

I am getting a i1display pro colorimeter tomorrow, so for the next rounds of replacements (I hope not!) I would be able to provide some measurements.

 

[kfscoll]

I just measured the native white point of my rMBP and it's 6437K. Pretty good.

 

[thekev]

Notice the maximum luminance delta of 9 cd/m2 for the rMBP and 18 cd/m2 for the cMBP. The maximum white point delta for the rMBP is 153K and for the cMBP it's 187K.

So, what this means is that my "shoddy" LG rMBP display has better brightness and color temperature uniformity than my older MacBook Pro's 15" anti-glare display. I think maybe I thought the cMBP's uniformity was subjectively better because its color uniformity is symmetric; in other words it's warmer in both lower corners of the display whereas the rMBP is cooler in one corner and dead-nuts on in the other corner (the corner, incidentally, that I thought was the most yellow).

Of course, the calibration tool can't take measurements in the extreme corners of the display so the delta values could be worse (or better) there. Still, this is pretty solid objective evidence that, in least in my case, despite what my eyes tell me, the uniformity of my rMBP's display is pretty significantly superior to my cMBP's display.

Interesting note -- I guarantee that most people would consider the post calibration results on all of my monitors (iMac included) to be "yellow". What's also interesting is that the post-calibration color temperature of my rMBP didn't change much from the default display profile. In other words, based on anecdotal evidence, the LG displays aren't too yellow; the Samsung displays are too blue.

Of course, all of this matters not a whit if you have a display with a color gradient. Calibration won't help you at all.

I wanted to note that the display hardware isn't really addressable directly. It's being adjusted via commands fed to the framebuffer to give a desired output aside from the backlight which I think is still a hardware control (some displays use software to lock the black level, but it would be difficult with sub 10 bit panel hardware). Anyway the tracking of your results is probably much closer on the rMBP than the cMBP because the native hardware response is much more similar to the desired target. The cMBP has a really cold white where the rMBP seems to be much closer to sRGB, which is very nice. This is why I hate when they identify gamut in percentage. If you look at any standards validation, it's always measured in Delta E tolerance from test patches rather than by absolute volume. A patch can be outside the gamut of the display, yet it's determined valid if it remains within the set tolerance level. With a volumetric standard it can be achieved in some weird ways.

Writing this on my next rMBP (another 2.6/512/16), which I received today and am going to return tonight at the Apple store.

Another LG display, LP154WT1-SJA1: Native white point 5200K, slight bleeding, perceptible screen refresh flicker.

That's quite warm. Some soft proofing systems use a D50 white point which is similar to a white value of 5000k (note similar for anyone who thinks I don't understand physics I know the math behind black body temperature values). sRGB is closer to a value of 6500k at its brightest point, so yeah that is pretty warm (more yellow in appearance). It could be an inconsistency in the way the hardware levels were set, and those can't really be addressed outside the factory. Even in displays with LUT based systems (or matrix based in the case of Quato) factory measurements still matter.

I just measured the native white point of my rMBP and it's 6437K. Pretty good.

That is really good. That's pretty much perfect. Colorimeters themselves vary. Remember that they're basically consumer grade devices unless you spend a fortune on one, so trying to match it much finer than that quickly becomes impractical. I like to stay close to hardware native whenever possible. Even then a profile can generate an accurate description of the hardware, and if you're using software with built in tracking, you can see how well the display holds up over time. Some sensors are affected by temperature, humidity, and may drift over time. Even then if you have a recent model colorimeter, it should be reasonably stable for several years. I'd be interesting to know how much the displays shift over time. My typical experience is that with full time use, it accelerates greatly after the 2 year mark.

 

[kfscoll]

@thekev -- good info -- keep it coming! My colorimeter is a new X-Rite i1Display Pro FWIW. I'm using standard default settings in i1Profiler to calibrate the display (aside from increasing the default patch set size to "medium"). I'm not using any software to track the display's color shift over time. As you alluded to, I figured the native white point I measured was well within the margin of error of the colorimeter so I felt pretty good knowing that it's essentially dead-on 6500K. I do have a minor bit of color non-uniformity but I'm probably not going to be able to do much better than what I have.

 

[mzbe]

Macbook Pro with Glaucoma display

Before returning the lemon, I showed the rMBP to my 75 year old mother in law, who is a passionate amateur photographer.

Her comment: This looks like when I had glaucoma on one eye, before the surgery.

As Apple is so fond of using the retina designator, it seems more than appropriate to me to henceforth refer to the bad displays as "Glaucoma edition".

 

[Dralt]

Hello,

Is this the right thread for therapy? I see yellow in every computer Apple sends to me. When I take my meds, I even see gradients and rainbows in various corners of the screens. Besides, I am not sure, but I really think I saw an image retention of Elvis for a brief moment in the last one I received, unless it was Jesus, I am not quite sure, honestly, since I'm having trouble sleeping due to all the stress all these exchanges and returns are causing me.

I hope the next one is OK or I am going back to paper!

 

[tooony]

my 2.6/16/256 rmbp today with a lg screen... with the screen on full brightness compared to my iphone 4s screen, it looks yellow... should i just get a replacement?

 

[answer348]

Purchased the rMBP (2.3/8/256) and waited several weeks for it to arrive at a reseller that operates through my university. Finally brought it home two days ago and placed it next to my cMBP (late 2008). I quickly noticed how the whites were much whiter and brighter on my old laptop. After a discussion with Apple over the phone, they considered it a hardware defect and had me return it to the reseller.

Since the reseller could not exchange it, I've placed a new order with Apple and I'm hoping for the best. I just can't imagine that such a beautiful display is supposed to look like that.

 

[sofianito]

my 2.6/16/256 rmbp today with a lg screen... with the screen on full brightness compared to my iphone 4s screen, it looks yellow... should i just get a replacement?

Put the Coldplay's Yellow song or the Beatle's Yellow Submarine and enjoy it...

 

[kfscoll]

At the risk of sounding like a broken record, chances are that your other displays (iPad, cMBP, iPhone, whatever) are actually too blue and your rMBP is fine. The native white point of my rMBP looked much more yellow than the native white points of my iMac, cMBP, iPhone, and iPad, but measuring my rMBP's native white point yielded a reading of 6437K -- an almost perfect match to the reference D65 white point. After calibrating my other displays, they wound up looking more like my rMBP color-wise (read: more yellow)...the white point of my rMBP was so close to reference out of the box that calibration didn't noticeably change it.

 

[cherishzm]

Not to hijack the thread but among several iPhone 4/4's I've had, I noticed some screens are more yellower than others at full brightness.