If you actually take the time to look at my test results, you will see that:
1/ The graphs only depict the measured sensitivity figures obtained from
my BCD396XT and
my BCD436HP. I also pointed out that there would be minor production variations between units.
2/ No GRE derived scanners were tested at the time.
3/ My BCD436HP met or exceeded the manufacturers specified sensitivity of 0.3uV or -117dBm (@ FM 12dB sinad) on every band but the 510Mhz band, where the sensitivity did seem low and did not meet the specified level. The 160Mhz band was the other band where sensitivity did seem a little 'down' compared to my BCD396XT, however despite this, the scanner still met the specified 0.3uV sensitivity level (just).
4/ Unless you are interested in receiving extremely weak/distant signals, as I am, it is unlikely that you will notice a difference of 1 or 2dBm in sensitivity. But with P25 in particular, a 2 or 3dBm difference in sensitivity can make a huge difference in decode BER performance of weak signals. Sometimes to the point where signals are either decoded or not.
5/ I have asked that others, owning appropriate test gear, contribute to this thread with their own test results to compare with my results, but sadly to date, nobody has. It is a little strange that considering the number of scanners that are out there, so little independent testing is actually done.
And
6/ Sensitivity is only one factor at play here. There are many other aspects such as IP3, receiver blocking, selectivity, digital decoder performance, audio filtering, squelch operation etc etc etc and they all impact on how well a scanner works