“ Really sad that they don't give us any info on how they measure sensors, so that someone could replicate the results...
That said, what to tell about this results, that show something a bit different. Taken as is from a post from Roger Clark, from clarkvision.com:
"My sensor analysis of the 7D2 is showing the improvements and reasons for some of the posted results by others.
I'll have a full report written up in a few days. A few advanced findings, though should be regarded as preliminary until I do a full writeup.
Full well capacity is slightly higher, but actual photon sensitivity is only 14% higher (0.2 stop in actual sensitivity) than the 7d1.
The noise floor is pushed down and pattern noise is a non issue at pretty much all ISOs and is lower than any other canon camera I have tested. Apparent read noise is an impressive low of 1.6 electrons at ISO 12800, raising to 2.4 electrons at ISO 1600. Image S/N in high signal regions will look similar to the 7D, but as brightness drops, the 7D2 images will look much better, and with the low apparent read noise and very low pattern noise, shadow areas can be lifted quite nicely. All this evidence comes from the sensor data with no raw conversion. The lack of pattern noise and clean low level noise means it should clean up nicely with filtering (unlike many earlier canon cameras).
Overall, I am very impressed. A note on DXO numbers. It appears that DXO is not correcting Nikons truncating of the raw data, which artificially improves dynamic range by about a stop. Also Nikon filters the raw data, improving noise and dynamic range further. I believe, based on some experiments, that if the canon data were treated similarly, it would result it numbers at least as good.
Previous to the 7D2 and 6D, pattern noise was a real limitation in Canon cameras (the 1D4 and 1DX are also pretty good, but not a good as the 7D2 and 6D). That pattern noise produced poor shadow areas compared to what could be extracted from Nikon sensors, especially at some ISOs (like 200 and 400 on many canon cameras). But Nikon's raw files look "wormy" in the shadow areas from the in camera filtering of the raw data. The Canon 7D2 raw data looks much more random, as it should be.
The 7D2 sensor dynamic range should be capable of 14.3 stops. Downstream electronics are still limiting that, so the per pixel dynamic range maxes at 11.2 stops at ISO 200. Note the DXO dynamic range is a pixel averaged 20x30 inch print dynamic range. I compute the DXO normalized dynamic range at 12.4 stops at ISO 200, 12.3 at ISO100, though I see DXO is now reporting 11.8 for the 7D2. I'm not sure why their numbers are low compared to mine.
I've also done some preliminary tests of the AF speed of the 7D2 compared to my 1DIV with a 300 f/2.8 version II. Results look very encouraging with similar performance. However, in some restricted cases, like focus moving from very close to very far, the 7D2 +300(vII) +2x (vIII)a is much slower and fails to find focus without hunting whereas the 1DIV nails it most of the time, but sometimes reacts as slow as the 7D2 (maybe 10% of the tries). If the focus range is limited, the 7D2 performs much better. AF from near to far is usually slow compared to far to near so if you prefocus, prefocus on the far side, so when you acquire the subject, the camera uses its fast rate direction.
Other sensors I have reviewed are here: http://www.clarkvision.com/articles/index.html#reviews_1"
Source:
http://www.naturescapes.net/forums/viewtopic.php?f=57&t=249565&p=2354307#p2354307 ”