Additional optional sample rates in SoapySDRPlay driver

[fventuri]

@SDRplay Andy, I am creating this new issue in the SoapySDRPlay driver to ask your opinion about the following question.

Franco Spinelli IW2DHW (@frspin) and I have been working for some time to get John Melton's G0ORX linphsdr program to work with the SDRplay RSPs using the SoapySDRPlay driver (see my pull request to Johns' repository here: https://github.com/g0orx/linhpsdr/pull/84) John's program uses a DSP library called WDSP (https://github.com/g0orx/wdsp) by Warren Pratt NR0V, that has been specifically designed for the HPSDR class of SDRs; these SDRs work with sample rates in the 48kHz 'family', i.e. 48kHz, 96kHz, 192kHz, and 384kHz.

Based on the discussions with you and Vincent a while ago (and based on what SDRuno presents to the user), I was under the impression that the RSPs would work best with output sample rates of 2MHz (and 1M, 500k, 250k, 125k, 62.5k using decimation) with an IF of 1620kHz, and with sample rates > 2MHz with a zero IF.

What we would be interested instead is to be able to use them with a sample rate of say 1536kHz and a zero IF (which with decimation would yield all the sample rates above, since 1536/4=384, 1536/8=192, and so on).

Franco (@frspin) modified his own copy of the SoapySDRPlay driver, ran same tests, and it looks like this configuration would work, but, not being familiar at all with the internals of the RSPs and the libsdrplay_api library, I wanted to ask your counsel before moving forward.

If you think these sample rate values are acceptable and do not cause other problems, I thought they could be added to the SoapySDRPlay driver, either in the main code, or perhaps using a 'cmake' compilation flag (like we used to to have for the RF gain thing a while ago), or perhaps in a different repo branch (which might make things a little more complicated since the two branches would have to be kept in sync after that). Another option would be for us to create a fork somewhere with these additional optional values of the sample rate, but I am afraid it could cause unnecessary confusion among the customers.

Thanks, Franco