Q: MacBook Pro Retina display burn-in?

I got my 3rd replacement (4th rMBP) today and another LG, but I checked it all possible manner but seems No IR. But whites has a hint of yellowish look in to it, but we have to carefully look at it. I had enough of this replacements now, So did u have any problems in your display if so what is it? Just a hint of yellowish is look is enough for a ask a display swap to a samsung one. 

I still think there is some confusion, you are only describing the polarization of the sub-pixels/pixels. They are twisted and remain in this polarized form for an unknown reason letting the light pass. That is obviously required for the light to pass and an image be generated. But you are assuming that PWM and the shades gradations all the depends on the twisted design os the pixels, for each voltage a certain twist and a certain amount of light would pass. And in this case the LEDs would have just two configurations, ON and OFF. Do I read you right?

The problem with this design is that the twists are too slow, I believe they usually take at least 2ms to return to the normal state of total blocking of light and the phosphor will take no more than 1ms to be lit and go off. It is also more efficient to modulate the intensity of the phosphor luminosity for temporal dithering then the twists of the pixels. And this brings me to an apparent contradiction in the twists-for-shades theory, since the pixels will have to change their configuration a few times in just one second to generate a proper shading how come a "state memory" would develop leaving the pixel persistently stuck in the same position?

Now the way I am considering the LEDs are the the ones submitted to modulation and changes in luminosity while the twists are either on or off only. So the retention would be cause by the phosphors and not by the twist configuration, which have to be many and always return to a balanced state among the WHOLE GROUP of pixels affected by the image and in a darker position to leave just a shadow a faint persistent image.

In any of these two cases, the array described in johns1's post is completely irrelevant.

Well that's what he said at first but after I told him he could just check with terminal with an installed display, he went to speak to his supervisor and told me they would get me a Samsung display, so I'm guessing they will just keep replacing the displays until they get a Samsung one.

bjiibj wrote:

 

shayster98 wrote:

 

Thanks Apples_8212! I guess even if it has no ghosting there's still a potential for it to develop, and the fact that the native white-point of the LG screen is also wrong deters me. (Although the LG floor sample in the Apple Store's color seemed fine to me, but there was no Samsung to compare it with. Furthermore, it even had IR.) Do you guys know how to get an advanced replacement and what really is it?

 

 

No offense intended, but I think this comment very succinctly represents the kind of illogical thinking that has been pervading some of the discussion here.  To say that the LG looked fine to you but you can't conclude it really was good unless you had a Samsung to compare with means that you are assuming that the Samsung has a more correct color representation and that whatever difference you would have perceived between the LG and Samsung would have been chalked up to the LG being wrong when the Samsung was right.  But what if it's the Samsung that is wrong?  Without objective measurement by equipment made for the purpose of testing color accuracy, we can't really say which is better or worse.  Maybe the Samsung has better color accuracy, or maybe not; I don't think anyone has presented any really compelling evidence either way.  But people really do seem to be quickly jumping to conclusions about other ways in which the Samsung is better than the LG, even without any objective evidence.  It seems that people are just looking for ways to dislike the LG just for further justification of the panel being inferior so that they have better reason to get Apple to replace them.  Which I think is not really necessary because the image retention problem is all that we really need to feel confident that the displays have a problem and need to be replaced.  And if you are an LG owner that doesn't have image retention or another real problem like dead pixels or white spots, I don't think you should feel like the display needs to be replaced just because it isn't a Samsung.

Bjiibj, I do get what you're saying. I did think it was a waste too to keep returning shiny new laptops. However, some who received rMBPs with LG displays did not have any image persistence issues at first, but developed the problem, interestingly, after the 14-day return policy expired. I really don't need that kind of issue as I'm starting school soon; I have little time to return and re-order a new one which doesn't seem to have the problem, but then starts to develop it three weeks into school. I have to take and edit many photographs and design layouts, and accuracy is pretty important. And, while it is entirely possible we're all completely wrong and the LG is the one that has a more precise native color calibration, the Samsung screen seems to be the preference of most. The whites look brighter and more neutral, and less warm. I'm not "looking for a way to dislike the LG just for further justification to get Apple to replace them", as you said. I just need what I paid for, and have no time to keep transferring data and reinstalling software from computer to computer. By the way (just to everyone in general), I've ordered a BTO rMBP; Apple says it should come on the 28th even with express shipping. I'll keep you guys posted on what manufacturer it is and if it has any noticeable issues.

High-Death

In any of these two cases, the array described in johns1's post is completely irrelevant.

Still trying to get back at someone else with a different opinion or logic then your own with a little jab at the end.  Such persistence, but lacking relevency.  

Just to add my experience.  After reading all of these posts, I became concerned since my rMBP is now about 1 week old and I have an LG screen.  I do notice some mild image retention.  I called my local store and asked them about the issue with the image retention.  They had not heard of it and suggested coming to the store to see if the floor models had this problem since they had been there and used for over a month.  We tested 4 of the rMBP that were on display and all of them were LG screens.  We tested them by leaving up the google screen and other websites.  The sales people and I were shocked that you were able to see the imprint of google and the other websites on the background and on other programs when opened.  You could clearly see the retained images on any background including the space background.  We came back to the computer almost 20-30 minutes later and they still had the retained image although it had faded.  They searched through their stock to see if they could find some kind of code to find a Samsung screen, but they were concerned that all of their computers had the LG screens.  They said I could bring the computer back for a full refund since I am within the 14 days or they could send it back to Apple to be repaired.  Personally after paying this much for a computer, I am going to bring it back and wait for the issue to be fixed before even considering buying another rMBP.  My biggest concern is the problem is only going to get worse.

There is nothing to get back at, it is a simple and direct conclusion.

Worriedperson

We tested 4 of the rMBP that were on display and all of them were LG screens.  We tested them by leaving up the google screen and other websites.  The sales people and I were shocked that you were able to see the imprint of google and the other websites on the background and on other programs when opened.  You could clearly see the retained images on any background including the space background.  We came back to the computer almost 20-30 minutes later and they still had the retained image although it had faded.

All the apple stores I visit have a active screensaver that acts as a demo that runs after so many minutes of non-user activity.   If you were to start using the MBPr you would never notice IR, which explains why none of the staff would ever notice ISP panel image retention. 

When you tested the four laptops with their staff were you changing the background to the solid dark grey background as mentioned earlier?  Or was this using the checkerboard pattern?

High-Death wrote:

 

I still think there is some confusion, you are only describing the polarization of the sub-pixels/pixels. They are twisted and remain in this polarized form for an unknown reason letting the light pass. That is obviously required for the light to pass and an image be generated. But you are assuming that PWM and the shades gradations all the depends on the twisted design os the pixels, for each voltage a certain twist and a certain amount of light would pass. And in this case the LEDs would have just two configurations, ON and OFF. Do I read you right?

 

The problem with this design is that the twists are too slow, I believe they usually take at least 2ms to return to the normal state of total blocking of light and the phosphor will take no more than 1ms to be lit and go off. It is also more efficient to modulate the intensity of the phosphor luminosity for temporal dithering then the twists of the pixels. And this brings me to an apparent contradiction in the twists-for-shades theory, since the pixels will have to change their configuration a few times in just one second to generate a proper shading how come a "state memory" would develop leaving the pixel persistently stuck in the same position?

 

In the first paragraph you correctly summarized: the amount of twist is a function of voltage, so you can achieve different degrees of brightness of a pixel using different voltages applied to the crystals.

In the second paragraph you deviate from this explanation and assume that you have to change the crystal configurations "a few times in just one second to generate a proper shading".  I don't know where this came from.  All that the display has to do is to supply the appropriate voltage for the desired twist of the crystals to produce the required transmissivity of light, and then the crystal's orientation remains fixed and doesn't change until a new color is to be displayed.

One thing I have glossed over here is that what actually happens is that a certain percentage of the crystals will twist fully at a certain voltage, resulting in an overall transmissivity that is dependent on the voltage.  An individual crystal either twists fully or not at all, and the opacity is a function of what percentage of crystals twist at a given voltage (even this is probably a simplification as I can't believe that every single crystal always twists 100% one way or the other, and there is never any stuck in an intermediate state, but in any case the effect is the same: the opacity is a function of the voltage, which produces a steady state orientation of the crystals).

Here is an excerpt from Wikipedia's article on "Twisted Nematic Field Effect":

"The amount of opacity can be controlled by varying the voltage. At voltages near the threshold, only some of the crystals will re-align, and the display will be partially transparent. As the voltage is increased, more of the crystals will re-align until it becomes completely "switched". A voltage of about 1 V is required to make the crystal align itself with the field, and no current passes through the crystal itself. Thus the electrical power required for that action is very low."

High-Death wrote:

 

Now the way I am considering the LEDs are the the ones submitted to modulation and changes in luminosity while the twists are either on or off only.

To reiterate, the twists for any given crystal are either on or off only; but of course each cell (subpixel) is made up of millions (billions?) of crystals and the overall opacity is a function of how many have twisted one way vs the other.  There is no need for temporal modulation to produce an effective opacity.  The static orientation of the crystals is enough.  And it's a divergence of the static orientation from what it is supposed to be, due to lingering voltages or some other effect that is causing the image retention.

Hey Guys,

As this post has now become so sprawling it is hard to see what is going on. The guys at mac rumours have set up a Poll

http://forums.macrumors.com/showthread.php?t=1422669

Asking for the people with rMBPs to do the checkerboard test and see what IR problems they notice. What this poll shows so far is that there ARE people with the Samsung screens that notice IR as well as there being people with the LG screens that DON'T  see any IR.

It would be cool to have as many people as possible take part so a picture of the true extent can be clearer.

Personally I have an LG screen and can't see any IR...

There is a problem with this theory about each twist degree representing a shade, that would mean there is 256 different voltages that will be simultaneously applied to the subpixels/pixels. So you are saying there are 256 different degrees of a twist. When you agree with me saying there is indeed only two mode, on and off, twisted subpixel or resting subpixel, you are incurring in an error, in this case there is nothing to control the shades but the LEDs. "Percentage of the crystals rearranged" in each cell each and every time an specific shade is requested? That would only be possible if you have 256 different potentials.

And there are two other problems with your theory, first these displays are not Twisted Nematic, they are IPS (In Plane Switching) and there are only 2 electrodes per PIXEL (not subpixel) therefore different arrangements for shading are all dependable on different differential potentials applied. Meaning, 256 different combinations of twists from all 3 subpixels (8bit) or temporal dithering, which is a combination of fewer different shades throughout time to replicate a final desired shade. There is also spatial dithering that you sort of suggested, but that would decrease real resolution and achieve a worse effect. So, this is where the "change in the crystals configuration" (quoting you) comes from.

And there is something else, this display is a 6bit display, therefore it has to recur to temporal or spatial dithering for achieving the 256 shades of a true 8bit monitor. So, either the twist's arrangement changes quite a few times in a second, or the shades are related to the LED's luminosity.

1. Yes, there are 256 different voltages that can be applied to each subpixel, each voltage causing more crystals to rotate, each voltage thus resulting in a different opacity for that cell.

2. Even IPS displays use crystals which rotate when voltage is applied.  The topography is different but the basic mechanism is the same.

3. Each cell of an IPS display is a subpixel, and each subpixel has its own electrodes; IPS just puts them in a different layer of the substrate than TN does.

This will be my last post of this discussion.  I really can't be bothered to continue to try to describe to you how LCD panels work.  You may find this page elucidating:

http://www.tftcentral.co.uk/articles/panel_technologies.htm

You will note that the backlight is never even mentioned in technical descriptions of how individual colors and brightnesses for each pixel are achieved.

Hi

http://youtu.be/QPxdhZHm3qc

I got the same problem.

I've exchanged the Macbook Pro for three times and unfortunately it's so difficult that I just want to have a perfect Macbook Pro Retina after spending US 2400?!

Also, Apple didn't offer good customer service, their attitudes just make me feel like I'm the one finding troubles with them and they don't really care if I buy the product or not.

I think I don't have so many times to waste on returning the product, after all, I need the computer to work.

As a customer, I think the basic thing I want is to buy a flawless product, and enjoy using it. Maybe Apple should think if there is any problem with its quality control or at least show the right attitude to solve the problems.

LG display, built in: Week 6 (February) or week 33 (August), no IR till now (after 1 day of testing)

bjiibj wrote:

 

1. Yes, there are 256 different voltages that can be applied to each subpixel, each voltage causing more crystals to rotate, each voltage thus resulting in a different opacity for that cell.

 

2. Even IPS displays use crystals which rotate when voltage is applied.  The topography is different but the basic mechanism is the same.

 

3. Each cell of an IPS display is a subpixel, and each subpixel has its own electrodes; IPS just puts them in a different layer of the substrate than TN does.

 

This will be my last post of this discussion.  I really can't be bothered to continue to try to describe to you how LCD panels work.  You may find this page elucidating:

 

http://www.tftcentral.co.uk/articles/panel_technologies.htm

 

You will note that the backlight is never even mentioned in technical descriptions of how individual colors and brightnesses for each pixel are achieved.

1. WRONG: from your own document:

When voltage is applied to a cell, the crystals of that cell all make a 90-degrees turn. By the way, an IPS panel lets the backlight pass through in its active state and shutters it in its passive state (when no voltage is applied), so if a thin-film transistor crashes, the corresponding pixel will always remain black, unlike with TN matrices.

IPS cells are either ON or OFF ONLY Like I said it, there are NO diferent levels of "twists".

2. The mechanism is not the same when compared to what you have said before, when you implied and yet again on the third item that more than two electrodes were present per pixel, again this is wrong.

3. WRONG, there are only 2 electrodes PER PIXEL, not subpixel. and IPS does not put electrodes in a different substrate, and you can even find this explicitly explained in the document you posted, and you can even find a few graphics there where it is more than crystal clear that the IPS crystal doesn't even twist, it simply turns - It is not a TWISTED crystal from like TWISTED Nematic - and the there is also a few graphics where you can see the subpixels between the 2 electrodes.

IPS matrices differ from TN Film panels NOT ONLY in the structure of the crystals, but also in the placement of the electrodes – both electrodes are on ONE WAFER and take more space than electrodes of TN matrices. This leads to a lower contrast and brightness of the matrix.

BTW, lower contrast = less shades.

The shades (therefore different colors) are achieved based on the orientation of the IPS cells (and combination of the orientation of the subpixels in the same cell (on or off)) and the intensity and frequency of the lit LEDs.

The other method is Frame-Rate-Control (FRC), also referred to sometimes as temporal dithering. This works by combining four colour frames as a sequence in time, resulting in perceived mixture. In basic terms, it involves FLASHING between two colour tones rapidly to give the impression of a third tone, not normally available in the palette. This allows a total of 16.2 reproducible million colors.

From wikipedia:

LED backlights are often dimmed by applying pulse-width modulation to the supply current, switching the backlight off and on again like a fast strobe light.

The interesting part of all this, is that by dimming a White LED (the case here) you can make it turn blue or yellow by increasing its intensity (the pink and yellow shift a few are experiencing and like I said before - many posts before actually - could related to an energy supply configuration.)

QUOTING YOU:

I really can't be bothered to continue to try to describe to you how LCD panels work.  You may find this page elucidating:

Yeah right, the page is very educating it is a shame you haven't read it... About the patronization (ONCE AGAIN), so you are trying to look wise by just looking (actually overlooking) for information in a on-demand basis... you don't even understand how IPS works and a few posts back didn't even know what kind of display was being used here...