Graeme Gill's measurements: some comments
Graeme Gill's measurements: some comments
- Subject: Graeme Gill's measurements: some comments
- From: "email@hidden" <email@hidden>
- Date: Mon, 24 Sep 2007 16:48:35 -0400
Hi to all,
Graeme is "spot" (very bad pun) on all of his comments. As you look
down the "Y" column of his data, you can see the luminance increasing
in the order of the FOV. I can't comment on the Spyder 2 because I
never measured the FOV, but for the other devices, luminance is scaling
quite niecely with FOV (field of view). Now why is that? Imagine that
we are in dark room and someone is pointing a flashlight at us. We see
a bright spot surrounded by black. Now let's measure that with a spot
meter and calibrate a sensor to that spot meter reading. As I adjust
the flashlight brightness, my sensor measurement will follow or track
the spot meter quite well . Now let's keep the spot meter pointed at
the flashlight and now turn on the room lights very low. Now, my wide
field sensor is seeing the room light and the spot meter is only seeing
the flashlight. Now I measure higher than the spot meter. As most
LCD's get dimmer, they not only lower the light level, but they begin to
radiate the light over broader areas. When we calibrate on the LCD in
"flashlight mode" using a spot reference, we are very accurate, but as
the LCD attenuation is increased, the light becomes more diffuse. If
you plot this out, it looks very much like flare in a camera. Laptop
manufacturers try to wring every nit they can, out of a display. This
leads to "enhancements" that make the problem even worse.
Another complication comes from the back light itself. The BLU
manufacturer has a similar problem: They need maximum nits as well, so
there is a natural tendency to design backlights that radiate
non-uniformily. When you combine this with the non-uniformity of the
panel, the light distributions become absolutely untenable. Backlight
units are often made by companies other than the panel vendor. This
means that a given display model, may have different vendor components
that match form, fit, but probably not absolute function. This is why
it is so hard for us to track down problems when we make a low end
calibrator that is used on a low end display. The physical display that
we use in our calibration standard, may be the same model number as
yours, but we have no way of knowing the characteristics of the glass &
backlight in you unit with the identical model number. This is probably
not true with Eizo because they don't make huge volumes. NEC is also
probably more consistent. Samsung makes their own glass so one would
hope that it is consistent. On the whole, the display industry is very
dynamic and unfortunately for us, displays for image critical
applications are far from common....
Regards,
Tom
If I measure the black level on my MacBook (which has a very poor vertical viewing
angle, with quite a lot of light coming from it when you are a few degrees below
the 90 degree mark - a rather disappointing display from Apple), I measure the
following for different instruments:
X Y Z D50 L* a* b*
Spectrolino: 0.366415 0.352546 0.692548 3.184534 1.069715 -7.584697
Eye-One Pro: 0.405160 0.391266 0.731828 3.534291 1.126687 -7.723253
Eye-One Display: 0.909546 0.913935 1.476462 8.252869 1.108462 -10.501008
DTP94: 1.010000 1.010000 1.850000 9.074471 1.321416 -13.169153
Spyder 2: 2.827000 2.660000 4.919000 18.627292 4.922603 -18.434490
(As an aside, the agreement between instrument as to absolute luminance levels
seems abysmal, something Roger Breton note some time ago).
Now whether a very narrow acceptance angle is a realistic representation
of how the viewer sees the screen is an interesting question, as is how
much it actually affects the end result.
Graeme Gill.
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden