search

wright-group / yaqc-cmds

A qt-based graphical client for yaq with a focus on coherent multidimensional spectroscopy in the Wright Group.
MIT License
7 stars 4 forks source link

sensor sleep time configurable? #345

Closed ddkohler closed 3 years ago

ddkohler commented 3 years ago

We have a sleep command while waiting for the measurement:

https://github.com/wright-group/yaqc-cmds/blob/45b2ac0f67e5a139daac532fbce3ddc6d2bc5e56/yaqc_cmds/sensors/_sensors.py#L166

It sounds like putting a large wait time (i.e. 0.1 s) is beneficial for fake hardwares, while we want it shorter (<= 0.01 s) for actual hardware to minimize dead time. Large wait times cause significant lag for few-shot acquisitions e.g.: https://gitlab.com/yaq/yaqd-ni/-/issues/2

Given it seems a parameter that changes, should we make the wait time a parameter of config?

untzag commented 3 years ago

My vote is just to take the wait time out. We can fix the "fake sensors are too fast" problem by adding in waiting to our fake sensors, if necessary. I don't think yaqc-cmds should attempt to rate-limit fast sensors.

If the client is somehow brought to it's knees by too much sensor data that's another thing... but I don't think that's the problem that we were addressing by adding that wait.

untzag commented 3 years ago

closed by #349