Thanks for sharing, you show that high sample rates and the resulting decimation ultimately lead to better results. But then you do need a SDR that isn’t susceptible for aliasing and the RTL is.
At the end it is a consideration between the decimation advantage compared to the risk that you receive aliased signals that aren’t really being received on the tuned frequency.
A side note, I have never seen a RTL-SDR successfully sample at higher frequencies then 2.56e6 above that values it starts to drop samples.
rtl_test -d 1 -s 3.2e6
Found 2 device(s):
0: RTLSDRBlog, Blog V4, SN: 40001150
1: Realtek, RTL2838UHIDIR, SN: 00000775
Using device 1: Generic RTL2832U OEM
Found Rafael Micro R820T/2 tuner
Supported gain values (29): 0.0 0.9 1.4 2.7 3.7 7.7 8.7 12.5 14.4 15.7 16.6 19.7 20.7 22.9 25.4 28.0 29.7 32.8 33.8 36.4 37.2 38.6 40.2 42.1 43.4 43.9 44.5 48.0 49.6
Sampling at 3200000 S/s.
Info: This tool will continuously read from the device, and report if
samples get lost. If you observe no further output, everything is fine.
Reading samples in async mode...
Allocating 15 zero-copy buffers
lost at least 824 bytes in buffer 0
lost at least 972 bytes in buffer 1
lost at least 1024 bytes in buffer 2
lost at least 612 bytes in buffer 3
lost at least 648 bytes in buffer 4
lost at least 1244 bytes in buffer 5
lost at least 972 bytes in buffer 6
lost at least 648 bytes in buffer 7
lost at least 408 bytes in buffer 8
lost at least 1280 bytes in buffer 9
lost at least 648 bytes in buffer 10
lost at least 1092 bytes in buffer 11
lost at least 904 bytes in buffer 12
lost at least 1484 bytes in buffer 13
lost at least 1296 bytes in buffer 14
lost at least 800 bytes in buffer 15
lost at least 852 bytes in buffer 16
lost at least 784 bytes in buffer 17
lost at least 800 bytes in buffer 18
lost at least 972 bytes in buffer 19
lost at least 1160 bytes in buffer 20
lost at least 664 bytes in buffer 21
lost at least 1040 bytes in buffer 22
^C
Signal caught, exiting!
User cancel after 23 buffers, exiting...
Samples per million lost (minimum): 3504
Jan | PE0SAT