Experimental Physics and
Industrial Control System

[email protected] (Tim Mooney) · Thu, 9 Jun 94 16:10:21 CDT

re...

> We are preparing to convert our control system to EPICS.  A large percentage
> of the control is done using the BITBUS as the field bus and we have a large
> investment in this architecture, using 486 PC's as the interface to the 
> control system.  Is anyone thinking about using PC's as an IOC?  VxWorks is 
> testing the 486 version at this time.  We presently have 15 BITBUS Master nodes
> installed, and 4 more required.  The present Master's run on 10 PC's, and would
> be replaced by at least 6 VME Crates using probably 68040's.  The cost of 
> purchasing this added equipment make the PC solution look attractive.  Any
> comments about using PC's? dependabilty, reliability, etc?

Every week I hear from someone wanting to use a PC as an IOC, either to
preserve an existing hardware/software investment, or to take advantage
of EISA boards that are cheaper than the equivalent VME board or that
have no equivalent on an EPICS supported platform.  In fact, several
APS-affiliated developers are talking about limiting their use of EPICS
at APS to those features they can mimic on PC's at their home labs, so
they can run approximately the same software in both places (a very big
deal to many developers here).

Scary?  It reduces EPICS, in such applications, to small pieces of the
channel-access library.  But I can't argue with someone solving their
group's problems the only way they can.

What would it cost to port EPICS to VxWorks on a PC?

Tim Mooney
APS XFD

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System