11:49 <rjo> dpn: This will be an interesting for Pounder but could potentially be worked around with frequency tuning instead

11:53 <mtrbot-ml> [mattermost] <dpn> @rjo: Agreed that both Sayma and Pounder would be good for our use case (stabilised gates), but having them usable in the lab (and at a similar level of stability as suservo!) is still some time out.

11:54 <mtrbot-ml> [mattermost] <dpn> (especially given the lockdown-induced complications)

11:55 <mtrbot-ml> [mattermost] <dpn> We'll probably track the phase accumulator state in SUServo instead and never reset it, instead applying the required relative offset to the POW when switching frequencies

11:59 <mtrbot-ml> [mattermost] <dpn> As far as I can see, that should also be accurate to 16 bits (an extra ~32 bit register per DDS channel and some more logic is needed, but that shouldn't be much of a problem)

13:01 <mtrbot-ml> [mattermost] <sb10q> except for the "inversion" sync bug, and the drtio failures (can be worked around by changing the identifier or making another 'innocuous' change and rebuilding the bitstream), sayma isn't too bad these days

13:04 <mtrbot-ml> [mattermost] <sb10q> and of course the MCH fiasco but that's not an issue if µTCA is ignored as it should

13:05 <mtrbot-ml> [mattermost] <sb10q> and of course the MCH/Metlino fiasco but that's not an issue if µTCA is ignored as it should (use Kasli as master)

13:26 <rjo> dpn: you mentioned you'd want to implement fiber length stabilization with pounder. do you have a timeframe for it?

22:31 <mtrbot-ml> [mattermost] <dpn> rjo: No, or rather not any longer. At some point, we had a potential student project lined up along those lines, but there currently aren't any short-term plans