Retina MBP calibrated ICC color profile?

[TibookAktive]

Are there any graphic designers out there with the means to produce a calibrated .ICC color profile using a Spyder unit (or similar) for the new MBP retina display and willing to post the resultant .ICC file on the forum/other file hosting site?

I've always found the macbook pros screen a bit warm and always managed to download a calibrated .ICC from somewhere (like this forum) to get a much better white balance and colour representation - I really think the retina will sing if we can do the same for it.

I haven't found any profiles anywhere on the web yet, so I thought I'd put out the call to fellow macrumorites and see if we can get the ball rolling!

Thanks guys/gals.

Richard.

 

[Fortimir]

Are there any graphic designers out there with the means to produce a calibrated .ICC color profile using a Spyder unit (or similar) for the new MBP retina display and willing to post the resultant .ICC file on the forum/other file hosting site?

If one doesn't surface by next week, I'll post one. I get mine on Tuesday and I'll run a calibration soon after.

 

[Beta Particle]

The point of display calibration is to adjust your individual panel to as close to specification as possible. (D65 white point, 2.2 gamma etc.)

This is because production tolerances mean that no two displays are the same.

Using someone else's ICC profile on your display is likely no more accurate than the factory profile, and in many cases is likely to result in less accuracy.

You need to buy/rent calibration hardware and calibrate your individual display.

 

[TibookAktive]

If one doesn't surface by next week, I'll post one. I get mine on Tuesday and I'll run a calibration soon after.

That would be really helpful! Thank you!

----------

The point of display calibration is to adjust your individual panel to as close to specification as possible. (D65 white point, 2.2 gamma etc.)

This is because production tolerances mean that no two displays are the same.

Using someone else's ICC profile on your display is likely no more accurate than the factory profile, and in many cases is likely to result in less accuracy.

You need to buy/rent calibration hardware and calibrate your individual display.

I fully take your point and if I was in graphic design I would definitely calibrate my individual screen to ensure color accuracy. However, I'm not in graphic design and I just have had good experiences in the past using calibrated ICC profiles (submitted by other users on forums, for example) and personally think that they have given better color and white balance on my previous MBP screens compared to the stock profile (which I feel tends to be a bit washed out and warm).

So just wanted to try a few different calibrated files from those lucky enough to already have all the equipment to see if I got the same positive results on my new rMBP.

 

[Fortimir]

This is because production tolerances mean that no two displays are the same.

Using someone else's ICC profile on your display is likely no more accurate than the factory profile, and in many cases is likely to result in less accuracy.

I agree 100%, but it's better than nothing. No skin off my back to provide one if someone asks.

But yes, ICC profiles are also based off other criteria that are environmental, like screen brightness and ambient lighting. If you are someone who need calibration, you really need to buy a unit and re calibrate every 2-4 weeks. Spyder3 and Spyder4s have a real-time ambient light sensor, as well.

 

[AZREOSpecialist]

Does anyone have a calibrated ICC profile for the Retina? OP never came back.

 

[stevelam]

calibrating your screen with someone elses profile is like getting the answers to a quiz, except you're taking a completely different quiz.

on a side note, totally not surprised that the person above ^^ is asking for it

 

[7even]

Dunno if it will help anyone but here's mine - i1Display Pro, 2.2 gamma, native whitepoint (IIRC).

EDIT: This is the Samsung / A00E panel

 

[AZREOSpecialist]

I found it curious that while running Apple Display Calibrator, it indicates that the display's native gamma is 1.8. I thought Apple moved to a 2.2 gamma starting with Snow Leopard. After going through all of the steps of Display Calibrator, with everything looking perfect at each step, the resulting profile makes everything look exceedingly dark - much more so than my calibrated 2.2 gamma NEC screen at the office.

Has Apple moved back to a 1.8 gamma for all systems, or just the Retina?

 

[derbothaus]

calibrating your screen with someone elses profile is like getting the answers to a quiz, except you're taking a completely different quiz.

I will ape that analogy thank you very much. All true as well. Sometimes someone else's profile can look better but it will be less accurate if you want actual accurate color.

 

[thekev]

I found it curious that while running Apple Display Calibrator, it indicates that the display's native gamma is 1.8. I thought Apple moved to a 2.2 gamma starting with Snow Leopard. After going through all of the steps of Display Calibrator, with everything looking perfect at each step, the resulting profile makes everything look exceedingly dark - much more so than my calibrated 2.2 gamma NEC screen at the office.

Has Apple moved back to a 1.8 gamma for all systems, or just the Retina?

Stop right there. The retina display gamma is not even close to 1.8. Do not feed bad information to everyone. This should tell you why not to use that terrible built in calibration function. Everyone wants to use it, which is why I mention not to overwrite anything. When they realize its results are worse, they'll pick something more logical. It's more like 2.2 to 2.3 given its close match to sRGB. Whatever you read there, it's wrong. That it defaults to such behavior doesn't make it any less silly.

I will ape that analogy thank you very much. All true as well. Sometimes someone else's profile can look better but it will be less accurate if you want actual accurate color.

The generic profiles provided by Apple range from bad to dismal, so it's possible that new unit to new unit the display profile could be an improvement. I set everything to display native on most of the Apple displays and use that as a starting point. I mean you can set up complex targets, but most people are likely to become frustrated by the results. My macbook pro display measures around 120 cd/m2 @ 8000K for its brightest white. I have the brightness around 5 bars down from the max, and I set i1 profiler to take the native maximum brightness and white point using the typical bradford matrix based system and the 504 patch test. Bradford is a pretty typical algorithm, and the lack of documentation doesn't provide enough information on how this is adapted to an LUT based profile (which changes a lot more than the outlying points), so I've avoided that option. Obviously that isn't an acceptable target for everyone, but it interprets everything at native gamut and white point, giving me a reasonably good result without causing any real problems. Just for (extremely nerdy) fun I tried to hit a D65 target. Regardless of measurement options used, it provided some weird results in the quarter tones. There's only so much you can expect. This is a 2011 MBP. The rMBP should be much closer. I suspect they're using a panel that's capable of wider gamuts and tuning it to be a bit warmer than the typical LED results. That works. Trying to emulate such a thing on a TN panel via software is likely to be much more annoying. Even prior to validation measurements it was obvious that certain grey tones had picked up too much chroma.

 

[derbothaus]

The generic profiles provided by Apple range from bad to dismal, so it's possible that new unit to new unit the display profile could be an improvement. I set everything to display native on most of the Apple displays and use that as a starting point. I mean you can set up complex targets, but most people are likely to become frustrated by the results. My macbook pro display measures around 120 cd/m2 @ 8000K for its brightest white. I have the brightness around 5 bars down from the max, and I set i1 profiler to take the native maximum brightness and white point using the typical bradford matrix based system and the 504 patch test. Bradford is a pretty typical algorithm, and the lack of documentation doesn't provide enough information on how this is adapted to an LUT based profile (which changes a lot more than the outlying points), so I've avoided that option. Obviously that isn't an acceptable target for everyone, but it interprets everything at native gamut and white point, giving me a reasonably good result without causing any real problems. Just for (extremely nerdy) fun I tried to hit a D65 target. Regardless of measurement options used, it provided some weird results in the quarter tones. There's only so much you can expect. This is a 2011 MBP. The rMBP should be much closer. I suspect they're using a panel that's capable of wider gamuts and tuning it to be a bit warmer than the typical LED results. That works. Trying to emulate such a thing on a TN panel via software is likely to be much more annoying. Even prior to validation measurements it was obvious that certain grey tones had picked up too much chroma.

Honestly I only use the default on Macbook's. If I owned a retina I would certainly try to profile. Currently using Spyder 3/ i1 D2 and ColorEyes (probably will upgrade when or if I ever get an LED backlit external). But results were always terrible on Macbook TN's. Apple actually kind of mimics proper color correctly if you get a good screen (usually the ones people complain are too yellow or brown). Good enough for me as I never proof on a laptop anyway.

 

[thekev]

But results were always terrible on Macbook TN's.

I'm still working on getting something really pleasing. I don't care so much about accurate on this display. The native colorlcd profile is positively awful. When I looked at the profile, it grabs the native gamma of the red channel and holds that relatively consistent, then mostly adjusts blue and green to match. i1 profiler said showed the display as being quite defunct in cyan values, which makes some amount of sense with how much they're tweaking the blue and green to get those results. My measured version evened them out quite a bit more. What annoys me at times is that people on here get a little too caught up in the aesthetics of hardware design. Personally I care about how things function.

As for Apple though, the rMBP is a nice step up in display quality from those that I've seen. I have yet to dissect the profiles on any of them. When I look at this stuff, sometimes I wish they'd hire fewer designers to obsess over the lines of the chassis in favor of a few more engineers, like that will ever happen.

Also I have an Eizo for anything important. If I had to work on the road, I'd buy a flight case and check it as luggage before I'd use a notebook display .

 

[derbothaus]

Also I have an Eizo for anything important. If I had to work on the road, I'd buy a flight case and check it as luggage before I'd use a notebook display .

I'm not that hard core. Kudos!

 

[thekev]

I'm not that hard core. Kudos!

Blah, I was hoping the response would be on my nerdy analysis of Apple's frame buffer gymnastics. There's a certain range I can work with. It's just hard to get that level of consistency + shadow detail from the Apple displays, and the viewing angles on TN are too frustrating for serious use.

 

[AZREOSpecialist]

Stop right there. The retina display gamma is not even close to 1.8. Do not feed bad information to everyone. This should tell you why not to use that terrible built in calibration function. Everyone wants to use it, which is why I mention not to overwrite anything. When they realize its results are worse, they'll pick something more logical. It's more like 2.2 to 2.3 given its close match to sRGB. Whatever you read there, it's wrong. That it defaults to such behavior doesn't make it any less silly.

Thank you for your help. Something doesn't work with the Apple Display Calibrator, so I have thrown those custom profiles away. I have reverted back to the stock LCD profile and everything looks nice again. I just wanted to give that program a try to see if it improved anything, and it most certainly did not.

 

[thekev]

Thank you for your help. Something doesn't work with the Apple Display Calibrator, so I have thrown those custom profiles away. I have reverted back to the stock LCD profile and everything looks nice again. I just wanted to give that program a try to see if it improved anything, and it most certainly did not.

Yeahhh.... you know my responses probably sound a bit harsh at times, but mostly I'm trying to avoid seeing bad ideas passed around the forum. It's important that the display is actually good and set up correctly. Re-profiling or "calibration" can only modify the instructions fed to the display hardware. It still depends on a properly manufactured display. If you do decide to go that route, the cheapest colorimeters I think are worthwhile are the colormunki display and spyder4. They're each a little under $200. The colormunki display won't work with third party LUT based packages. You can't use it with basiccolor, coloreyes, spectraview, color navigator, etc. X-rite makes you buy the i1 display pro for that. It also comes with slightly more robust software options. It's the kind of thing where that's your next potential step up, and you may or may not find it worthwhile. If you were a photographer, graphic designer, etc. I'd say buy one.