Thunderbolt Display out the box or calibrate

I sometimes feel like I haunt these threads, but there is too much misinformation on the topic on here at times. First the only control any of this software will grant you on an imac or thunderbolt display is a rewritten profile. In system preferences---> displays, it will say something like color lcd by default. Specifically referring to the calibration of Apple branded displays, there is no way to correct backlight bleed or poor uniformity. People do sometimes compare these things to other displays or their phones, but it doesn't work. They all shift as they age, and Apple has used displays with different characteristics in different lines. If half the screen is a different color, it won't fix that.

Can you explain what you meant by whites were not white? I suspect a lot of these things are are just that users aren't happy with the pass or fail tolerance levels used. The "calibration tools" typically referred to as colorimeters can be helpful if you're picky or want to track the behavior of the display. If you use one, block out as much ambient lighting as possible when you run the software, and give the display 30 minutes to warm up prior to calibration. This helps with consistency of measurements. Letting the profiling software run with a colorimeter typically takes 5-15 minutes depending on software and settings used. Some have options for how many patches are tested. That affects how long it takes. Some software waits for the hardware to stabilize for a couple seconds between patch measurements.

Keep in mind that this is a software adjustment only. None of it carries over if you also boot into bootcamp unless you run it separately there. I would say in all cases, leave as many settings as possible to native or sRGB unless that doesn't accomplish the desired results. With consumer grade tools and ICC profiling, it's generally a good idea to to make things as easy as possible for the software.

 

I just bought a colorimeter which should be delivered from Amazon today. Datacolor Spyder4 Elite. I chose this model because it supports color matching multiple monitors attached to the same computer.

My goal is two-fold:

1) Acceptable color accuracy for my printed photos
2) Consistency across my two screens (27" iMac & 27" ATD)

Different color temperatures became evident when I started using dual displays attached to the same computer. To date... printing has only been snapshots for friends and/or small projects. I am about to start printing/framing some of my best work for our house.

/Jim

 

I think it is typical for various displays to look different... especially when viewed side by side.

Hopefully my colorimeter will be delivered today and I'll be able to report back any progress in making my dual displays look similar.

/Jim

 

The Spyder4Elite will also check for conformity across your panel. Nothing you can do about it of course but at least you can check the results.
As you mention you do printing, Can I ask why you didn't go for the Spyder Studio? Then you could have calibrated your printer as well. Another alternative is the ColorMunki Photo which will do both your screen and printer.

 

Sorry...

I do not "do" printing... although I have owned a ton of photo printers. I've given up on them and just have a color laser now. I just never printed enough photos to make it worthwhile.

I am planning to send out my work to a local photo pro shop. I thought I should calibrate my display first. But truthfully... it was the iMac and TBD sitting side by side with mis-matched color profiles that pushed me over the edge to get the colorimeter.

My Spyder arrived today and maybe I'll get around to run a calibration attempt tonight... depending if I get all my work done first.

/Jim

 

Should take no more than 20 minutes per monitor. Let us know how you get on.

 

No problem man. I try to mention some of the limits of the technology relating to how it's presented, as it can be confusing (calibration sounds like a low level hardware adjustment doesn't it?). Apple has changed characteristics in terms of color temperature settings and other things many times. Even among their current lineup, there's a pretty huge variation. I suspect they want everything to be roughly sRGB, but it hasn't happened. My cmbp notebook screen is actually very cold. Measured natively i1 profiler says roughly 8000K. sRGB is closer to 6500K. Both color temperatures represent a range of colors. I'm talking specifically about the white points, as most displays shift somewhat throughout their ranges.

I will add I've never recommended the thunderbolt display as a display. A lot of people buy it for its functionality as a dock, but there are displays that I like better. I suspect they're going to update it to match the newer imacs soon. I figured June would be an appropriate target initially. My old posted estimate was no earlier than March or April, more likely mid-year. The reason was they wouldn't push a new/expensive process onto peripheral hardware until things stabilized and yields improved. I think you would probably be happier with one once they are updated. Otherwise if you don't mind matte displays and don't require the docking functionality, there are other nice displays on the market.

I'm just going to mention a couple things here. The colorimeter helps, but remember that print shops vary, and if you want it to match your display, the color temperature and to a lesser degree the brightness of the light illuminating that print has to match. The image will still look good under other light later, as normally things are viewed in context. Fluorescent lighting is kind of ugly, but if you're at the grocery store, everything you see is lit by it. If it was just one object, the effect would be much more obvious.

Make sure you get a proof either way if you're going to have a large job run. Printing houses still have kind of a draconian way of doing things. I really think the LUT systems used in grading video are much nicer than print workflows.

There are various programs to help compensate for that, but it's important not to over-correct. Typically matching two displays is kind of a lowest common denominator scenario, and they aren't likely to match on every value. If you're running any kind of profile check that goes through a patch test, you will see many odd colors. They correspond to LAB or XYZ values, which your display matches as closely as possible based on the written profile. I wouldn't expect all of them to match between two, especially those of different models or ages. Typically I would designate a primary display and get the other as close as possible.

I would not try to get a new imac to match a display that has been in service for several years. That would be artificially limiting the imac. I would probably start off with the imac as close to native behavior as possible, so the profile basically just provides your system a better description of the underlying hardware. This is useful for color managed software that needs to make conversions so things will display properly. You can actually demand any set of of target values that the software supports, but that doesn't make it a good idea. After they're both done on native, if the older display is too warm or too magenta or something like that, you can adjust the target settings somewhat. It uses a transformation matrix on pivotal points channel by channel, so overdoing your requests isn't a good thing.

 

I'm glad I found this topic.

I purchased a new 27" iMac and a thunderbolt display. After setting them up side by side, the thunderbolt looked amazing compared to the iMac. The whites are truly white and it seemed more vivid. I decided I wasn't going to cope with two different displays, so I returned the iMac and got a second thunderbolt to tether to my MBP retina. Much to my dismay, the second thunderbolt is also more yellow or washed out than the first thunderbolt. I then held up my macbook pro retina to the first thunderbolt and it too looks like it has more of a yellow hue to it than the bright white thunderbolt #1.

The only way I can get the displays to match is to calibrate #1 and drag the native point down to match the crappy whites of the second display.

Is this normal?? I understand some variation should be expected... but it's pretty noticeable side by side.

Kind of wishing I'd gone for the Dell monitors.

Will the spyder device help with this?