I would like to provide some more data to this discussion. I compared the response of a brand new out-of-the-box i1Pro 2 with a brand new out-of-the-box stock D3 OEM. I used 5 different display types.
Then I compared WRGB measurements of the i1Pro 2 and the uncorrected D3 with my JETI 1201. Here's what I got.
The i1Pro 2 does better on the plasma and slightly better with the CRT, but with the LCD, LED, and projector the errors are essentially the same. Overall, the error rates between the i1Pro 2 and stock D3 are almost exactly the same. Given that the sensitivity of the D3 is vastly superior to the i1Pro 2, I am a little puzzled by this repeated insistence that modern colorimeters simply cannot be trusted and you are generally better off with a budget spectro. With a properly calibrated D3, even the color accuracy would have been better than the i1Pro 2.
I know of only two other published examples looking at this. One is Michael's, which has been discussed here. However, this is not really an evaluation of the D3's accuracy. It is an evaluation of one of the correction tables on his C6, which is clearly defective. The other comparison I know of was published here
. As with my test, it also failed to show any real advantage when using budget spectros.
I have an open mind about this, but I have personally measured hundreds of D3s on multiple displays prior to calibration. The errors I see are very small. I have also tested the i1Pro repeatedly throughout the years. It, too, shows small errors. However, unlike the D3, its relatively poor signal-to-noise ratio affects it performance with low light sources.