Open danielballan opened 7 years ago
Stray thought: a simple way to do this now and be right 90% of the time would involve RemoteDispatcher
, tqdm
, and total=start.get('num_points')
.
So just a quick pint from my end. This does not need to be perfectly accurate. In fact I think that for adaptive scans it will always be somewhat less reliable.
So I would just use the process Dan has mentioned which is close most of the time and gets more accurate the further into the scan you get. This is enough to let the user know a rough eta for the end of the scan which is all that is needed.
Cheers Andrew
Sent from my iPhone
On 20 Aug 2017, at 12:56 pm, Dan Allan notifications@github.com<mailto:notifications@github.com> wrote:
Stray thought: a simple way to do this now and be right 90% of the time would involve RemoteDispatcher, tqdm, and total=start.get('num_points').
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/NSLS-II/bluesky/issues/776#issuecomment-323597491, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AYlCpYAll0vv3d457AuLGx0appAwbY8jks5saGVDgaJpZM4O8W4Z.
Thanks, @awalter-bnl. We'll take this idea for a spin at ESM when we visit to upgrade the software.
Any update on this?
This can be done using existing tools -- no changes to bluesky needed. I will write up an example tomorrow.
I had forgotten that @danielballan had already a base idea to use for this, once he is back I will discuss with him. For plans with multiple 'yields' inside them we could probably couple with #713 (time for scan compelte) to get an eta (without progress bar) at the end of each 'yield'.
thoughts?
Notes on a conversation with @awalter-bnl:
And on a conversation with @tacaswell: