Experimental Physics and
Industrial Control System

Klemen Vodopivec via Tech-talk <[email protected]> · Thu, 30 May 2019 10:11:59 -0400

Hi Hinko,

usually the device driver developer knows best what the update rate is

for the particular device. That said, you'll probably have to write some

code. Using the monitor updates, you can reduce the frequency on the

client side by skipping some updates but you can't get more updates than

the IOC is pushing out. Also, timing of updates depends on the IOC and

record definition, you may not be getting updates at regular intervals.

So if you want to do some regular periodic updates, CA get is your best

friend. Here's proof of concept code using PyEpics for a PV that changes

at 60Hz:

$ cat slow_camonitor.py
#!/usr/bin/env python

import epics
import sys
import time

pv = epics.PV(sys.argv[1])
interval = float(sys.argv[2])
while True:
    print pv.get()
    time.sleep(interval)

$ python slow_camonitor.py BL2:Det:rtdl:Timestamp 1
2019/05/30 09:52:03.200780666
2019/05/30 09:52:04.200765666
2019/05/30 09:52:05.217418666

$ python slow_camonitor.py BL2:Det:rtdl:Timestamp .1
2019/05/30 09:52:08.917366666
2019/05/30 09:52:09.017365666
2019/05/30 09:52:09.117363666
2019/05/30 09:52:09.217362666
2019/05/30 09:52:09.317360666
2019/05/30 09:52:09.417359666
2019/05/30 09:52:09.517358666
2019/05/30 09:52:09.617356666

On 5/29/19 1:21 PM, Hinko Kocevar via Tech-talk wrote:

This might sound like a silly motion, but you are probably used to those from me by now.

A non-EPICS person at my facility wonders if EPICS could support camonitor like PV monitoring, but with CA client dictating the rate of value updates. Imagine 1s update rate, that certain slow client would use, while the PV would update much faster, and the rest of existing CA clients just use regular camonitor’ing.

I can see answers like, just use caget like approach and poll or define a new (set) of PVs for that slow rate, but that is IOC material. My use case revolves around very, very busy IOC devs that have no time to support crazy ideas of individuals who want slow updates.

Any insights appreciated!

Thanks (and have a nice weekend to those living in Sweden),
Hinko

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System