You might have taken more time understanding how the test was performed before commenting on it.
The measurement of the plasma by the corrected D3 did not match the reference measurement by +/-0.001 because the reference measurement was taken on a different plasma, different make and model. In fact, that was the whole point of the test--to show that a correction to a D3 using one plasma as a reference was transferable to other plasmas.
If you truly believe that the test was useless, then of course you could easily provide your own evidence in support of your skepticism. Instead of that, we get yet more hackneyed testimonials about your wonderfulness and blanket statements about abstract theory for which you provide no data whatever.
The pointlessness of this is illustrated by the fact that a question was asked that I attempted to answer with specific data, data that is broadly consistent with what I have seen many, many times before. You ask readers to ignore the data without providing any of your own or making any attempt to answer the original question. This is not helpful to anyone. A little more useful information and a little less self-promotion would be welcome.
So, I'll do what you didn't. I'll test this.
I'll measure white on the same display with the i1Pro in contact mode in a completely dark room at night. Then, I'll immediately measure again offsetting the i1Pro from the screen on a tripod by 5" and then 8" (the D3 PRO was at 8"). Next, I'll change the read angle by a few degrees, which also changes the read point by an inch or two. Finally, I'll repeat the readings with all of my equipment at the back of the room turned on.
- i1Pro contact measurement: x0.315, y0.329--exactly the same as I had measured a few weeks ago.
- i1Pro at a 5" offset: x0.315, y0.330-0.331 (it would flip back and forth).
- i1Pro at an 8" offset: x0.315, y0.329-0.330 (it would flip back and forth).
- i1Pro at 8" angled down slightly: x0.315-0.316, y0.330 (it would flip back and forth).
- Repeated all measurements with LED equipment light on in the room. This made no difference at all. The readings were exactly the same.
The conclusions to be drawn from this data?
- A non-contact reading increases the amount of green the i1Pro reads over a contact reading somewhere between 0.0015 and 0.0005. In other words by a negligible amount in any way you want to measure it--CIE94 0.2, RGB 0.4%, y 0.0015.
- Small changes in read angle also make a difference, but a negligible one, somewhere between 0.0005 and 0.001.
- Having LED lights from equipment on in the same room made no difference at all.
Of the four variables you mentioned,
- contact vs. non-contact reading
- small changes in read angle
- small changes in read point
- LED light from an equipment rack in the same room
one made no difference at all and the other three made a difference so small as to have no bearing on the original test. In fact, to the extent they had any bearing at all, they show the i1Pro to be a little LESS accurate because the y reading was pushed even further away from the reference reading.
LED lights from equipment racks can make a difference with some readings--in particular, when attempting to read black level on a high contrast display, but they make no difference when you are reading a very bright test pattern.
This is a textbook case of the difference between broad theoretical assertions and measurable reality. Is the statement
So with an i1Pro in contact mode on a Plasma it is also seeing the green pixel structure and can give you false results.
true? Perhaps. But the point is that it doesn't matter, because whatever effect this has on real-world measurements is so small that it is just not relevant. If we were at CERN measuring the speed of neutrinos then perhaps these sorts of variations would matter. Since we are measuring commercial TVs with inexpensive mass-market equipment it doesn't.
Is this statement
Lighting control has to be done in an absolute black room not even the light from equipment leds can be allowed.
true? Well, it depends what you are measuring. If you are measuring a very dark test pattern then LEDs at the back of the room may very well contaminate the result, but the suggestion that the tiny amount of light emitted by LEDs 15 feet away would affect a reading of a 118 cd/m2 test pattern is absurd on its face.
Originally Posted by derekjsmith
Also in the above example the i1Display3 that was calibrated to the Plasma using a Jeti 1211 was shown to get closer results than the i1Pro on a Plasma why? Because it was calibrated to it so it should if done correctly have results within +/- 0.001 xy but it does not. Again it's all about having the proper lab, knowledge and procedures to do this.