Color calibration problems on mac pro

Question

I have a bunch of quicktimes I captured via firewire from a 1200a deck. These quicktimes I converted to Quicktime PNG via compressor so they could be opened on a PC as well. When I open the exact same quicktime file on PC it appears much brighter than when I open it on my Mac. And I dont just mean it appears brighter to the eye, but when I hover my mouse cursor over pixels the values are much brighter (I tested this in aftereffects on both machines).
The PNG and DVCPROHD quicktimes give the exact same values on my mac (unsurprisingly), and when I view them in aftereffects or FCP they look quite dark (much darker than what I saw at the shoot). But when I view them in quicktime player they look brighter - just like how they look on PC (normally exposed). BUT when I do a screen grab using CMDCTRL+SHIFT4, and open the grab in photoshop, it looks much darker again! with the darker values!
Both PC and Mac monitors are calibrated with the eyeone Display 2 - but I dont think that is really the problem, because the actual pixel values I'm reading with aftereffects or photoshop are much darker on the mac! I have a feeling that the values the mac is giving are wrong, because everything looks really underexposed, the blacks are completely crushed etc... but I'm sure on the shoot there was much more detail in the blacks and everything was exposed properly.
Any ideas what is going on? it seems like some wierd color profile thing... but would that actually change the pixel info in the quicktime file?

Answer 1

Hi memosan-
I completely understand. My experience is with medical imaging, where each monitor is absolutely required to match the other. Please step back and look at the bigger picture.
You are making the leap of faith that the way each system interprets pixel values is based upon the identical math, which it cannot possibly be. Each system will do what I thinks it needs to do to display a certain pixel value based on your calibration to a desired result.
When you think of displaying the image, you must think about what happens after the data is read off of the storage device and assigned to the graphics system and prepared for display. Up until that very instant the data is identical. What happens next will determine where those white values start differentiating.
When you calibrate a monitor you essentially build a table within your system that the graphics card uses to calculate a specific value to display based on the capabilities of that particular system. Even with identical systems those numbers will be different system to system. Maximum white will equal one number, maximum black another and so on.
When you read a pixel value it is based upon the math that the system is doing in order to display the image that you have requested.
If any part of the system is different, the math that it must do to obtain a particular pixel value will be radically different than the next system. If the temperature gradients are not the same, or if the actual pixel depth of one monitor is 10 bits and another 12, or if they cannot display the same black value the results will all be the same in that pixel values will be different, sometimes very radically.
The only way to actually demonstrate this is for you to use the same NTSC or PAL monitor, matching bit depth cards in each system, calibrate each system to that monitor without adjusting the monitor, then display the same image from the same file and see what the values read. My experience would tell me that only then will you be able to get numbers that are close.
This has everything to do with the monitor displays and your calibration and your different systems.
Luck-
-DaddyPaycheck