Experimental Physics and
Industrial Control System

Kay-Uwe Kasemir <[email protected]> · Fri, 15 Sep 2006 16:22:28 -0400

On Sep 15, 2006, at 15:56 , Bill Nolan wrote:
Hi all,

It looks like I am going to have to connect a number of devices in

to my system that only offer active-X hooks.

I have two device types with widows control software that offer

Active-X interfaces, the remainder of my controls are Epics on
Linux hosts.

Is the Active_X interface form the CA_interfaces page still the
best way

to approach getting the two systems to talk ?

In case you're referring to the ActiveX CA server & client stuff that

one can still get from http://ics-web1.sns.ornl.gov/~kasemir/axca/
index.html :

Bad idea. It was primarily developed to allow Labview to serve

data via Channel Access, but for LabView there are now better

solutions, so this has not been maintained for years.
Besides, if you wanted to use _that_ to connect your ActiveX stuff

to EPICS, it would mean:

You write a gateway program in for example Visual Basic that uses the
Active X

controls that you got for your devices and passes data back and forth

between them and the Active X CA server or client.
Would making a soft IOC and linking the vendor supplied Active X/dll's
in to sub() records be a better way to go ?

Yes, that would be a much more direct way to get from an IOC database

to your devices. And I agree with Mark that writing 'proper' device
support

is probably better than routines for sub-records, unless this is only

a 'one-of' thing that you need to handle.
-Kay

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System