Ad
Ad
Ad
Pages: [1]   Bottom of Page
Print
Author Topic: Best screen calibration for LR on Mac?  (Read 7008 times)
Mosccol
Jr. Member
**
Offline Offline

Posts: 74


« on: March 14, 2007, 02:13:25 PM »
ReplyReply

One of the questions you would like to ask Michael as he talks to you through the video downloads...  

I have calibrated my screen using a Spyder 2 on my Mac Book Pro 15 as well as my Samsung 19 LCD. In both cases, the calibration has done a good job but the standard way to look at photos (especially those coming from an sRGB environment) of a Gamma of 2.2 and a temperature of 6500 K looks horrible (serious pink hue). I have found that the most natural calibration was 1.8 Native, with Gamma 2.2/Native being warmer but also looking good.

At present I am not getting much accuracy from my printer (for other reasons outside this topic) so I am stuck with subjective observation of my monitor.

Any views from Mac LR users?

Thanks!
Logged
61Dynamic
Sr. Member
****
Offline Offline

Posts: 1442


WWW
« Reply #1 on: March 14, 2007, 02:24:56 PM »
ReplyReply

Monitor calibration is not tied into the software being used. One calibration and profile work for all applications.

If you are seeing a pink hue either you are not calibrating things properly or you are way over caffeinated. More likely the former since Gamma should have no effect on how warm or cool the screen appears. You cannot depend on your subjective view of the screen to judge a calibration. The whole reason for the spectrometer (your spyder) was to get away from using your eyes for calibration and obtaining accurate and consistent results.

Outline the full process you go through in calibration and we can find out what's going wrong.
Logged
Mosccol
Jr. Member
**
Offline Offline

Posts: 74


« Reply #2 on: March 14, 2007, 02:50:01 PM »
ReplyReply

Quote
Outline the full process you go through in calibration and we can find out what's going wrong.
[a href=\"index.php?act=findpost&pid=106631\"][{POST_SNAPBACK}][/a]

Hello 61Dynamic & thanks for the offer...

I have followed the Spyder software instructions including warmup and lighting arrangements. I have replicated the experiment and obtained substantially the same results using ambient halogen lighting, daylight and total darkness.

I have also calibrated using the Apple built-in software to cross check results and the pink hue only happens with the Spyder. The results are as follows (using one of the GretagMacBeth test images on screen). Interestingly my Intel MacBook Pro and my G4 MacMini (with DVI connection to a 19" Samsung LCD) give fairly similar results.

Default Apple LCD setup (1.8/Native): a bit blue/cold

Apple software calibration to 1.8/Native: pretty close to 'natural', perhaps still a bit cold

Spyder 1.8/Native: very natural looking but dark skins a touch pale

Spyder 2.2/Native: as above but warmer/darker, it's a toss which I prefer between this and the one above; very easy to get used to between the two

Apple 2.2/D65: even warmer and darker but actually closest to paper prints (compared to the other settings)

Spyder 2.2/D65: pretty horrible; pink hue, not natural looking at all. Can get used to it, but not completely

So there you have it... my full experience!

I guess that the question boils down to: do you need to use D65 when dealing with sRGB pictures? I would be quite happy to work with 1.8 Native or 2.2 Native. How are other Mac users set up?
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9092



WWW
« Reply #3 on: March 14, 2007, 03:37:37 PM »
ReplyReply

In theory, a 2.2 TRC gamma and D65 should be your best bet. Even better is Native white point and native gamma. I don't recall if the Spyder software provides either setting (it should at the very least, allow you a Native White Point).

Be sure you evaluate the color in an ICC aware application like Photoshop. Some app's haven't a clue the display profile exists.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
61Dynamic
Sr. Member
****
Offline Offline

Posts: 1442


WWW
« Reply #4 on: March 15, 2007, 09:55:41 AM »
ReplyReply

You don't need the monitor to be any specific color temp for working with any files. The calibration and profile of the monitor is independent of the color space used. That being said, 6500K is the best choice because that is the temp the human eye responds to best. And it's what most displays operate at anyway. The ACD is an exception. It operates at 7000K but I profile mine to 6500K with no ill effect.

Your gamma depends on the video card. Macs classically used a gamma of 1.8 however recent Macs seem to be 2.2--at least the systems I've had access to are, including mine. If you calibrate to Native settings, 2.2 is probably where it will be.
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9092



WWW
« Reply #5 on: March 15, 2007, 11:02:55 AM »
ReplyReply

Quote
Your gamma depends on the video card. Macs classically used a gamma of 1.8 however recent Macs seem to be 2.2--at least the systems I've had access to are, including mine. If you calibrate to Native settings, 2.2 is probably where it will be.
[a href=\"index.php?act=findpost&pid=106798\"][{POST_SNAPBACK}][/a]

I don't think the TRC gamma is a video card byproduct but rather the physical attributes of the display. There may well be something going on here as when you calibrate a display, you do need to take the card into effect (that's why you can't move the display to a different system along with the profile). Displays are usually in the neighbourhood of 2.0-2.2. Mac's assumed an 1.8 gamma behavior which is part of the OS and I'm afraid this didn't change, as it should have when Apple moved into OS X. They kept this assumption for (silly) legacy issues.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
NikosR
Sr. Member
****
Offline Offline

Posts: 622


WWW
« Reply #6 on: March 15, 2007, 01:33:33 PM »
ReplyReply

Just to add my 2 eurocents worth, although I'm far from an expert.

I have noticed that with many average TFT monitors (that is, not proof quality) calibrating to a colour temperature other than 'native' produced sub optimal results.
Logged

Nikos
dcp
Newbie
*
Offline Offline

Posts: 13


WWW
« Reply #7 on: March 15, 2007, 01:57:22 PM »
ReplyReply

I had the same problem when I was using the Spyder2pro. When i used the blue eye pro that came with my lacie.. the problem went away. I used the same calibration on my other screen and it went away too.
Logged

2x Canon EOS[span style='color:red']1[/span]dmk2; Canon EF16-35mm f2.8[span style='color:red']L[/span], 24-70mm f2.8[span style='color:red']L[/span], 35mm f1.4[span style='color:red']L[/span], 85mm f1.2[span style='color:red']L[/span], 70-200mm f2.8[span style='color:red']L[/span]IS, 135mm f2.8SF, 180mm f3.5[span style='color:red']L[/span] Sigma 10-20mm f4-5.6 - Full time wedding photographer
Mosccol
Jr. Member
**
Offline Offline

Posts: 74


« Reply #8 on: March 15, 2007, 02:22:09 PM »
ReplyReply

Quote
I had the same problem when I was using the Spyder2pro. When i used the blue eye pro that came with my lacie.. the problem went away. I used the same calibration on my other screen and it went away too.
[a href=\"index.php?act=findpost&pid=106829\"][{POST_SNAPBACK}][/a]

Now that's interesting... Of course it is interesting to check the calibrator itslef... As it is new out of the box you have no way of knowing if it is not faulty... What did you do?
Logged
61Dynamic
Sr. Member
****
Offline Offline

Posts: 1442


WWW
« Reply #9 on: March 15, 2007, 11:48:49 PM »
ReplyReply

Andrew,

In the ancient times when mammoth CRTs ruled the earth I learned the gamma was determined by the video card alone. It makes sense LCDs could effect gamma but I haven't come across anything that shows it definitively.

Do you have any sources you can share on LCDs effecting gamma? I'd like to read more into that.
Logged
seanmcfoto
Full Member
***
Offline Offline

Posts: 176


« Reply #10 on: March 16, 2007, 12:28:07 AM »
ReplyReply

Just my 2 ˘ (albeit in €)
I have also found that on many different machines, both Mac and PC, that the Spyder2Pro gives a pink tint using the recommended settings of 2.2/D65. I've been using 2.2/Native instead myself.
The local camera club bought one and I (unfortunately for my free time) ended up running it for a number of the laptop users there.
Logged

Mosccol
Jr. Member
**
Offline Offline

Posts: 74


« Reply #11 on: March 16, 2007, 04:28:08 AM »
ReplyReply

Quote
Just my 2 ˘ (albeit in €)
I have also found that on many different machines, both Mac and PC, that the Spyder2Pro gives a pink tint using the recommended settings of 2.2/D65. I've been using 2.2/Native instead myself.
The local camera club bought one and I (unfortunately for my free time) ended up running it for a number of the laptop users there.
[a href=\"index.php?act=findpost&pid=106921\"][{POST_SNAPBACK}][/a]

Aha, so I'm definitely not alone... I have left a support ticket at Colorvision but no response so far...
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9092



WWW
« Reply #12 on: March 16, 2007, 09:05:21 AM »
ReplyReply

Quote
Andrew,

In the ancient times when mammoth CRTs ruled the earth I learned the gamma was determined by the video card alone. It makes sense LCDs could effect gamma but I haven't come across anything that shows it definitively.

Do you have any sources you can share on LCDs effecting gamma? I'd like to read more into that.
[a href=\"index.php?act=findpost&pid=106917\"][{POST_SNAPBACK}][/a]

I'm pretty darn sure TRC gamma is a physical attribute of the display. CRT's had a true gamma (if you define this using the gamma formula). LCD's don't at all but are made to mimic this response for obvious legacy reasons. The following text was blessed by Karl Lang who's posts on displays has been quoted here and elsewhere for some time. Karl is the father of the PressView and Artisan display.

Quote
Displays have a native Tone Response Curve (TRC) specified by a gamma value, which is a result of their natural behavior. Gamma is a specific formula (output = input gamma ) that describes a very simple curve. This curve, the result of this gamma formula, in the case of a display, describes input amplitude (voltage) and the corresponding light output (brightness). Various values for gamma produce different curves. So we can use a single number to describe this type of curve. However, many devices do not follow this formula, their curves are far more complex and are not gamma curves nor should we use gamma to describe them (tone response curve is the accurate term).

There is no difference between a display a user will connect to a Macintosh versus a Windows system. The operating system does make some assumptions about gamma. The TRC gamma assumed by the Macintosh is 1.8 while the TRC gamma assumed by the Windows operating system is 2.2. Over the years, most Macintosh users were under the impression that the correct TRC gamma to use as a target value was 1.8. This is due to the early Macintosh operating system attempting to produce a screen to print match on devices that also used a TRC gamma of 1.8; the LaserWriter printer. Since color management didn’t exist back in those days, producing a native TRC gamma of 1.8 assured that the Grayscale output and the preview of images on the Macintosh were closer. However, the display’s native TRC gamma was not close to this gamma assumption. The native TRC gamma of most displays is in the neighborhood of 2.0 to 2.2. Today the use of 1.8 gamma within the Macintosh OS is simply a legacy and not useful. If a user calibrates the display to something other than a TRC of gamma 1.8, the appearance of color outside ICC-aware applications will appear darker or lighter. There are compelling advantages to calibrating a display as close to the native, physical TRC gamma as possible. The farther the display calibration is from the native gamma of the display, the more adjustments have to be produced at the graphic card. Macintosh users should set their TRC gamma target value to 2.2 instead of 1.8. The slight darkening of images outside ICC aware applications is not severe enough to present a problem and the results of keeping the display calibration closer to it’s native condition is less aliasing of images.

Since TRC of the display set for calibration is recorded in the ICC profile and this information is provided to Photoshop and other ICC-aware applications images outside an ICC aware application like Photoshop may appear a bit dark. However the soft proofing previews in Photoshop are correct. This is how a Macintosh user and a Windows user can view the same image with the same color appearance even if both systems are assuming a different gamma.

The natural TRC of an LCD is a severe S curve and doesn’t even remotely follow the gamma formula. LCD manufacturers want to have their displays act like CRTs. End users who don’t use color management expect the colors to be at least “in the ball park” of what they are used to seeing on CRTs. To achieve this, the LCD has a built-in 8-bit LUT (Look-Up Table), which makes the LCD follow the gamma formula, usually gamma 2.2. By converting 8-bit input data into 8-bit output data, the result is banding (aliasing) in images.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Mosccol
Jr. Member
**
Offline Offline

Posts: 74


« Reply #13 on: March 16, 2007, 09:24:29 AM »
ReplyReply

Quote
The following text was blessed by Karl Lang who's posts on displays has been quoted here and elsewhere for some time. Karl is the father of the PressView and Artisan display.
[a href=\"index.php?act=findpost&pid=106983\"][{POST_SNAPBACK}][/a]

That's useful Andrew. So I guess that 2.2/Native should be my choice. Still waiting to hear from Colorvision on that pink cast for D65...
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9092



WWW
« Reply #14 on: March 16, 2007, 09:28:07 AM »
ReplyReply

Quote
That's useful Andrew. So I guess that 2.2/Native should be my choice. Still waiting to hear from Colorvision on that pink cast for D65...
[a href=\"index.php?act=findpost&pid=106993\"][{POST_SNAPBACK}][/a]

Actually Native Gamma should be ideal. Yet another famous post by Dr. Lang:

Quote
What is optimal front panel controls or video card LUT for adjusting 
white?

The simple answer is you can't really know. Manufacturers don't give 
you enough information in most cases.

The internal data paths of most LCDs are not published. Modern mid to 
high end LCDs do not have simple 1D LUTs. Most chip sets today have 
3D lookups which are altered by all of the front panel adjustments 
and a factory-set final grey balance 1D curve after that. Display 
features like NECs "colorcomp" are also using up data to make spacial 
corrections (this is never mentioned in the data sheets.)

Using an analog input signal never fixes anything! That input signal 
is not a smooth curve it is a stair-step signal created by the video 
card DAC. The only thing this will do is introduce noise and 
additional aliasing in the system as it is redigitised. Even if a 
front panel control did adjust the analog input side (i have never 
seen this) this would only create worse aliasing due to the 
misalignment of the stair-step signal and the ADC thresholds.

These things said, here is what you can do.

It is always preferable to keep the white point as close to native as 
possible. If you only have one display and native is 6800K use it 
don't correct it, your eyes will adapt. If you need to have two 
displays side by side then you need to correct white point.

If you are using a modern high-end LCD. Use the front-panel to adjust 
the white point, load a neutral LUT in VC. Look at a grayscale, is it 
neutral all the way to black? or did it look much better when set to 
native? This will give you some idea if your internal path has the 2 
stage 3D-1D system I described above. If it does use the front panel 
to correct the white, If not use the cal software and the LUT.

Do not adjust "contrast", "brightness" or "gamma" on the front panel. 
Do adjust "backlight", on some models backlight is incorrectly 
labeled "brightness."

When you calibrate do not change the TRC of the system ("use native 
gamma") changing the TRC will only cause more aliasing.

Do not use a Matrix/TRC based profile. Use a full lookup. This will 
provide the best soft proof. LCDs are not CRTs they do not use 
phosphors that mix in a linear fashion. There is a lot of crosstalk 
between channels especially in the darker tones. This can only be 
properly modeled with the full lookup ICC profile.

Hope this helps,

Karl Lang
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
61Dynamic
Sr. Member
****
Offline Offline

Posts: 1442


WWW
« Reply #15 on: March 16, 2007, 11:29:25 AM »
ReplyReply

Thanks for posting those quotes Andrew. Very informative.

A question: Where is Dr. Lang posting this info?
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9092



WWW
« Reply #16 on: March 16, 2007, 11:39:33 AM »
ReplyReply

One of his more 'famous' quotes was on the Betterlight forums and pasted here among other sites.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Mosccol
Jr. Member
**
Offline Offline

Posts: 74


« Reply #17 on: March 19, 2007, 07:44:18 PM »
ReplyReply

I finally got an answer from Colorvision which pretty much confirms the 2.2/Native choice:

"there is nothing forcing you to use 2.2/6500K for your calibration. If you select the native white point (which can be better in some cases) than the software will determine the white point of your screen and write into the profile all measurements using the values of you screen. This is also a rigth calibration.
The software will calculate everything in order to have a right calibration profile."

OK, the style is Swiss German but I think it says Spyder software will behave properly with Native. Doesn't explain why I get the ugly pink using D65 but it is no longer important, right? ...

Logged
lbalbinot
Jr. Member
**
Offline Offline

Posts: 70


WWW
« Reply #18 on: October 30, 2008, 07:02:52 AM »
ReplyReply

This is old, but I'd like to add some thoughts that I've recently learned. I had the same problem with the pink hue on the new LED-backlit MBP and one thing that helps is to set the backlight to 50% or less before calibrating. It seems there's a color cast when the backlight is too bright and that will throw my colorimeter off (Spyder 3).

Luis
Logged

Luis F Balbinot
luisbalbinot.com
ddk
Sr. Member
****
Offline Offline

Posts: 274


WWW
« Reply #19 on: October 30, 2008, 07:47:13 AM »
ReplyReply

Quote from: Mosccol
OK, the style is Swiss German but I think it says Spyder software will behave properly with Native. Doesn't explain why I get the ugly pink using D65 but it is no longer important, right? ...


Fortunately there are great companies like Integrated Color to turn to when you run into Swiss German!

Somewhere you made a comment that 2.2/D65 setting is closest to the actual prints, this makes me wonder if the ugly pink hue is in your images, its hard to make any determination without seeing one of your files.

david
Logged

david
-----------------------
www.pbase.com/ddk
Pages: [1]   Top of Page
Print
Jump to:  

Ad
Ad
Ad