The programmer of Linrad (aka Leif sm5bsz) has uploaded a video to YouTube that compares several software defined radios on dynamic range and compression performance in the presence of strong nearby signals. In the video Leif tests the Airspy, BladeRF with B200, FDM-S1, Funcube Pro+, rtlsdr/E4000, rtlsdr/FC0013, rtlsdr/R820T, SDR-14 and SDRplay.
The main test works by tuning to a broadcast band FM frequency and then injecting a strong carrier signal at distances of 500 kHz, 1 MHz, 2 MHz and 5 MHz from the center frequency. The carrier signal strength is slowly increased until the SDR shows signs of complete degradation of reception of the FM signal. Better SDRs will tolerate stronger nearby signals without degradation.
The results are summarized at 34:20, 1:21:38 and 1:48:30. We have also taken screencaps of the results at 1:21:38 and 1:48:30 and they are shown below. The first column is when a higher gain is used, and the second column is when a lower but still barely copyable gain level is used. In the Levels for loss of performance columns smaller numbers are better and in the Dynamic range columns larger numbers are better. Finally, at the end of the video starting at 1:45:55 Leif also tests the spur performance of the SDRs.
A few days ago we posted a video by sm5bsz showing some comparisons between the E4000, R820T and FC0013 tuners, and also a comparison between the special linearity gain mode driver in Linrad and standard Osmocom driver in SDRSharp.
Now sm5bsz, programmer of Linrad and the special gain modes for the E4000 has done another test using only Linrad, which more fairly demonstrates the difference between the various tuners, and the effect of the special gain drivers in Linearity mode. He writes
In this video RTL2832 dongles are compared for sensitivity, spurs and intermodulation. The difference between the Linrad linearity mode and the original Osmocom gain setting is demonstrated as well as spurs in R820T and FC0013.
Which one to prefer depends on the local RF Environment and whether a selective filter is used between the antenna and the dongle.
Note: The Linrad vs SDRSharp video has been removed by the uploader.
Finally in this video, he also compares the standard Osmocom driver to the sensitivity mode available in the modified gain profile drivers. He writes
The sensitivity mode has very poor performance for signals far away from the passband, but it allows about 10 dB better dynamic range for interferences within the passband. Sensitivity mode is for usage with a selective preamplifier while the Osmocom gain mode is a reasonable compromise. The Linrad linearity gain mode is for use without filters in difficult RF Environments.
On YouTube sm5bsz has uploaded a video showing a comparison between the E4000, R820T and FC0013 tuners, and also comparing the receive performance of SDRSharp and Linrad. In the video Linrad showed superior receive performance with the E4000 when compared to SDRSharp due to some custom gain profiles which are enabled in Linrad only (but can also be enabled in SDRSharp with a plugin/mod).
Linrad is another software defined radio program which is much more difficult to use, but was the first program to support the modified librtlsdr. Some people prefer Linrad due to it’s advanced GUI which has a lot of signal information on display.