There could also be FPGA constraints that make the clock architecture on the 
E320 necessary. 

Sent from my iPhone

> On Jun 6, 2022, at 7:57 PM, David Raeman <[email protected]> 
> wrote:
> 
> Hi Jon, I did some poking around in the code, and I don't believe the E320 
> supports that feature. On B2xx radios, if you don't specify an explicit 
> master clock rate it has logic to determine an ideal rate based on the 
> sampling rate, and it exposes an auto_tick_rate property to toggle that 
> behavior from the application. Conversely, on the E320 radio if you don't 
> specify an explicit master clock rate it always uses a fixed value of 16 MHz, 
> and the implementation has no logic that I see to perform the dynamic tick 
> rate calculations.
> 
> Since the B2xx and E320 radios are both based on the AD9361 RFIC, 
> theoretically I think the logic could be placed somewhere such that the E320 
> and B2xx radios could both provide the feature. However, the implementations 
> are pretty diverged - the E320 is based on the newer MPM code architecture 
> and the B2xx is not.
> 
> Short story is the E320 doesn't seem to support this, but I think it's just a 
> matter of missing bits in the software.
> 
> Best,
> -David
> 
> 
>> -----Original Message-----
>> From: Jon Beniston <[email protected]>
>> Sent: Monday, June 6, 2022 6:46 PM
>> To: 'Marcus D. Leech' <[email protected]>; usrp-
>> [email protected]
>> Subject: [USRP-users] Re: E320 Automatic master clock
>> 
>> Hi Marcus,
>> 
>>>> If I just try to re-make the device, I get an exception. Eg:
>>>> 
>>>>                m_dev = uhd::usrp::multi_usrp::make(device_args);
>>>>               m_dev->set_master_clock_rate(61.44e6);
>>>>             // How to restart a session here?
>>>>                 m_dev = uhd::usrp::multi_usrp::make(device_args);
>>>> 
>>>> ...
>> 
>>> However, there is a set_master_clock_rate() API call:
>>> 
>>> https://files.ettus.com/manual/classuhd_1_1usrp_1_1multi__usrp.html#a
>> 99
>>> 254abfa5259b70a020e667eee619b9
>> 
>> Yes, that's what I'm calling above.
>> 
>>> What the consequences are for changing this within a session
>>> is necessarily device dependent.   If I were going to do this, I'd tear down
>> the streamers, set the
>>>  master_clock to the new rate, and then re-create the streamers.
>> 
>> The problem is, that method doesn't appear to support an "automatic" rate,
>> unless I'm missing something.
>> 
>> Thanks,
>> Jon
>> 
>> 
>> 
>> _______________________________________________
>> USRP-users mailing list -- [email protected] To unsubscribe send an
>> email to [email protected]
_______________________________________________
USRP-users mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to