Hello, Bill. I can't recall if I've ever communicated with you directly, but I have a Pentax K1ii and was part of the agonizing thread resulting from DPReview's reliance upon negative test scores and the consequent poor rating they gave the K1ii. I have subsequently used the K1ii with a wide variety of lenses and my own experience is that it is a superb camera for Landscape (hiking) use. I haven't left Pentax but I did get a little bored over Ricoh's time lag in producing new gear and have begun adding Nikon gear. Of the three Nikon cameras I've purchased (The D500, D800e and the D700), I liked the D800e more than the D500, but after getting the D700 am inclined to like it best -- for the sort of photos I take.
I am not totally at odds with Tony Beach. I've had cause to take all such test scores with a grain of salt. They may all be true, but they haven't really tested for the sort of shooting I do; so when I find that a camera, e.g., the K1ii, performs better than the tests would lead me to believe, I conclude they must have been testing for something I don't need.
Also, I was in charge of a lab testing department for verifying that the C-17 had met all the requirements the Air Force had contracted with us for. A variety of test parameters were established and agreed upon by both McDonnell Douglas and the Air Force. My little group took each required test, coordinated with the lab test engineer, design engineer, and anyone else as necessary to get the evidence that would satisfy the Air Force that we had met their requirements. Frankly, the DXO information that I've seen thus far would require more selling than I would feel comfortable doing. Imaging Resource, at least, provides the illusion of clarity. Of course, all it would have take to bring the selling process to a halt would be an Air Force engineer asking a question like Tony Beach's to bring the selling of the Imaging Resource results to a screeching halt.
Imaging Resource does their own testing as far as I know:
https://www.imaging-resource.com/TIPS/TESTS/TESTS.HTM They test, or at least claim to test for more parameters than DXO seems to.
Every site I'm aware of has issues with their testing. Broadly speaking you can get a good idea what's going on, but within the margins of error they are a waste of time.
FWIW, the PhotonsToPhotos measurements have a pretty small margin of error.
See
Measurement and Sample Variation for some specifics.
However, the D7500 and D800 are not within the margin of error here because the D800 has a much larger sensor than the D7500.
Yes, sensor area is the key factor (unless you get into equivalence arguments which I never entertain).
Why, for example, does DXO say that they test ISO for sports?
Yeah, the DxOMark Sports Score is a confusing combination of multiple factors.
If anyone wants to see the breakdown they can visit
DxOMark Sport Score Audit .
I'm not sure what I'm looking at, but I checked some of the cameras I'm familiar with & see, for example, that the D600 handily beats the D850 in terms of "published score." I'm pretty sure the D850 is a better camera than the D600; so I'm not sure how valuable it is to know that the D600 has a higher "published score."
Ignore that, and their lens comparisons too. However, their measurements are largely consistent with Bill Claff's, both of which I provided you links for above.
Yes, I used to use DxOMark as a "sanity check" but found that if anything my measurements were more accurate.
Having been retired from Boeing Engineering since December 1998, my "sanity checks" have become much more modest: If most of a day's shots look pretty good to me in Lightroom 6 and I have 20 or so I'm willing to upload to my Smugmug account (a modest thing of interest to a few friends and Rhodesian Ridgeback fanciers) then I am happy for the rest of the evening.
Lawrence