Re: [Discuss-gnuradio] Gnuradio clock coherence?

[Marcus D. Leech]

On 11/14/2017 02:13 PM, John Ackermann N8UR wrote:
> What is the clock source for a Gnuradio "internal" source such as the 
> Signal Source block?  If there is no hardware device connected, is the 
> block ultimately driven by the PC clock?
>
> Then, if there is a hardware source (SDR) as well as an internal 
> source, does the hardware source provide the master clock for the 
> flowgraph, such that any other software-derived sources are coherent 
> with it?  Or does a Signal Source block still get its clock from the PC?
>
> My real question is whether it's safe to assume that a flowgraph is 
> entirely clocked by the samples from its hardware source, and 
> consequently that processing within the flowgraph (e.g., oscillator 
> and mixer for tuning) does not introduce potential frequency errors.
>
> Thanks,
> John
A flow is clocked by the hardware--either by samples being produced or 
consumed by the hardware.

The exception is when you use a "throttle" block in the absence of any 
hardware.  In that case, the PC clock (very roughly) controls sample rate.

So, in the absence of overruns or underruns, the signal processing in a 
flow-graph cannot introduce timing errors, unless deliberate a programmmed.