Calibrating display with Spyder 4 Pro

[reRESERVEDMD]

Hi,

I'm new to display calibration. I've got a retina and a Mac Pro with an LED display. The LED is an eyesore and I've decided to improve it.

Question I got to you is:

What should I set for the white point on the retina? With current lighting condition, it wanted to set it to 5000K, but that was too yellow. When I turned ambient lightning off and used 6500K setting, it appears too blue. Should I go with 5800-6000K? I'm looking for something versatile (low lighting at home, bright at work and Starbucks).

I assume the 2.2 Gamma, Native Brightness, and no ambient light setting is alright?

Thanks,

Paul

 

[derbothaus]

2.2 Gamma, D6500 White point (Closest to Apple's Native). Calibrate in a black room to minimize the reflections from the glass. Don't trust ambient if it wanted to set at 5000K. It is making corrections for how warm or cool the current room is. I am assuming you will be moving around and lighting will change depending on day, no? LED is usually a greenish cast addition. A little more Red is probably OK. Blue not OK. What SW are you using? The included Spyder 4 stuff I don't have any experience with on quality. I have Spyder 3 and i1 D2. I might suggest getting a demo of ColorEyes Display Pro and running that in Demo for a bit and see what results you get. It is a much much better profiling solution. It costs though. But you can keep using the profiles it creates in demo when demo expires.

 

[thekev]

I would initially leave the white point native. Brightness is entirely up to you. If it's the thunderbolt display, I think the thing is just a bit higher than D65. The actual white point (as in the brightest white) of the rMBP is supposedly right around D65. I recall the TB display being closer to 7000k. If you're trying to match a specific output, it may take a small amount of adjustment either way. As Derbothaus stated, do it in a dark room to minimize the effects of extraneous light. Leave the spyder plugged in a few minutes prior to calibration. Give the display itself half an hour to warm up prior to starting.

Also in what way is it an eyesore? If something is truly wrong, generating a new profile from measured values (what you're doing) won't fix that. If you're still having trouble, it may be time to pay Apple a visit.

 

[reRESERVEDMD]

2.2 Gamma, D6500 White point (Closest to Apple's Native). Calibrate in a black room to minimize the reflections from the glass. Don't trust ambient if it wanted to set at 5000K. It is making corrections for how warm or cool the current room is. I am assuming you will be moving around and lighting will change depending on day, no? LED is usually a greenish cast addition. A little more Red is probably OK. Blue not OK. What SW are you using? The included Spyder 4 stuff I don't have any experience with on quality. I have Spyder 3 and i1 D2. I might suggest getting a demo of ColorEyes Display Pro and running that in Demo for a bit and see what results you get. It is a much much better profiling solution. It costs though. But you can keep using the profiles it creates in demo when demo expires.

I would initially leave the white point native. Brightness is entirely up to you. If it's the thunderbolt display, I think the thing is just a bit higher than D65. The actual white point (as in the brightest white) of the rMBP is supposedly right around D65. I recall the TB display being closer to 7000k. If you're trying to match a specific output, it may take a small amount of adjustment either way. As Derbothaus stated, do it in a dark room to minimize the effects of extraneous light. Leave the spyder plugged in a few minutes prior to calibration. Give the display itself half an hour to warm up prior to starting.

Also in what way is it an eyesore? If something is truly wrong, generating a new profile from measured values (what you're doing) won't fix that. If you're still having trouble, it may be time to pay Apple a visit.

Thanks to both of you for the help! I calibrated both of those using 6500K, 2.2 Gamma, and around 120cd for brightness. Both displays look spectacular, especially the Thunderbolt.

This thing was an eyesore because I could tell that the colors were way off. Now it matches the rMBP (apart from a bit of purple tinge on the retina due to the high resolution display). Overall, I'm very satisfied with it.

I wish there was a way to hook a calibrating tool like that to an iPhone and an iPad.

 

[derbothaus]

Great. Very standard settings.

 

[iLukeJoseph]

You shouldn't have a purple "tinge" and if you do it is not caused by the high resolution display.

I am not even sure I would suggest using 2.2 gamma either (some like to argue, but there is no "standard" gamma setting). When I do a TV, I do gamma depending on room conditions, my person TV's are all calibrated pretty closely to 2.35.

If your software has an option for it, I would try L* or even sRGB. I consider sRGB a gamma setting, as it uses the standard Rec. 709 color gamut (HDTV color gamut), but it a different gamma, off the top of my head pretty sure it is 2.2 but it is a different curve than what you would calibrated a HDTV to.

 

[thekev]

I'm glad that worked out well. Those are pretty standard settings, although the cMBPs can't really hit D65 without problems in my experience. The rMBP was a huge step up for Apple in notebook displays.

You shouldn't have a purple "tinge" and if you do it is not caused by the high resolution display.

I am not even sure I would suggest using 2.2 gamma either (some like to argue, but there is no "standard" gamma setting). When I do a TV, I do gamma depending on room conditions, my person TV's are all calibrated pretty closely to 2.35.

If your software has an option for it, I would try L* or even sRGB. I consider sRGB a gamma setting, as it uses the standard Rec. 709 color gamut (HDTV color gamut), but it a different gamma, off the top of my head pretty sure it is 2.2 but it is a different curve than what you would calibrated a HDTV to.

L* is frequently debated, yet I'm not sure why people like it. A few years ago I remember claims (mostly from sales-people) that a CG211 + basiccolor calibrated and profiled to L* produced shadow detail comparable to whatever reference CRT. 2.2 is pretty standard though. You need to remember that you're applying this as a fairly high level adjustment. It's a fairly simple matrix profile used to tweak the instructions fed to the gpu that drives the display. I try to pick targets as close to native hardware behavior as possible so as not to create new problems. This is especially true for the OP who wanted to match multiple displays, at which point a sort of common denominator target that is close enough to the hardware gamut of both displays must be chosen. You do not want to end up with weird casts in certain value ranges or banding, both of which can occur when trying to apply overly aggressive adjustments to something that really wasn't designed for them.

You might already know this, but sRGB greatly predates HDTV. It was intended to be an ubiquitous target as it could be approximately matched by a new crt of reasonable quality in the mid 90s. It's also not exactly 2.2, but it is close.

 

[iLukeJoseph]

Even though sRGB predates HDTV, that is irrelevant to my post. Rec. 709, which is what HDTV's should be calibrated to. sRGB uses the same exact color space. The only difference between Rec. 709 and sRGB is the gamma curve.

So in your theory of trying to keep things as native as possible, what is native for the R? If native is sRGB wouldn't you want to calibrate with that?