Hello, I just want to make sure I understand the different API's and when to use them. I've been using the multi_usrp API for most development using the standard FPGA load on an N320.
I recently added in an FFT RFNoC block into the FPGA load, and will soon add 1 or more custom RFNoC blocks. I switched to the RFNoC API for my application code so that I can create my graph and directly interface with the blocks. I believe I "needed" to do this, correct? Meaning if I wanted to create a graph with anything other than the default statically connected radio->ddc (for RX path) and if I want to interface with those blocks, then I need to use the RFNoC API. Please let me know if I'm understanding this right. I've already done this, and things are working -- but I want to make sure I couldn't have done this with the multi_usrp API. We have a lot of pre-existing code using the multi_usrp API. If I add and start using different RFNoC blocks, I want to be sure that I need to switch over to the RFNoC API before modifying all that code. One thing that confuses me a bit is that I just found the multi_usrp_rfnoc.cpp file, which contains the code for the multi_usrp_rfnoc class. I'm not clear if or how this class is used. I see that it is a subclass of multi_usrp, but if I "grep" UHD source for "multi_usrp_rfnoc", I cannot see where an instance of multi_usrp_rfnoc is created anywhere. I do see that multi_usrp_rfnoc.cpp is included under the LIBUHD_APPEND_SOURCES in "/lib/usrp/CMakeLists.txt". I'm wondering if I could/should be creating a multi_usrp_rfnoc class, or if this somehow happens behind the scenes when using the multi_usrp API with an RFNoC-based radio. Again, I don't see where it's being used at all, but I am far, far, far from a C++ expert and am likely not understanding something. Thanks for any help, Jim
_______________________________________________ USRP-users mailing list -- usrp-users@lists.ettus.com To unsubscribe send an email to usrp-users-le...@lists.ettus.com
