Experimental Physics and
Industrial Control System

[email protected] · Tue, 16 Apr 2002 10:02:41 +0200

  > if someone likes to change the ao record: What do you think about adding two
  > new fields? These fields contain binary values which indicates that the high
  > (low) limit is reached (limited or rejejected). This would be useful for
  > other application parts like digital records or graphic displays.

I would strongly oppose against such kind of suggestions. For a record
that is used as widely and in high numbers as the ao, there must be more
than good reason to blow up its code and storage.

Also having multiple single bit fields for conditions that are exclusive
seems a bit inefficient.

In this case the intended functionality can be achieved in many ways
using the ao's alarm limits and/or a second (e.g. calcout) record that
might even write to another mbbi if translation into status strings is
needed. Such solutions using an additional calcout can also be used for
trend indication, rate-of-change calculation etc.

One of the main advantages of the EPICS database is the approach of
small atomic records that are connected by powerful links to build
complex applications. Introducing all possible features and complexity
into the atomic components breaks this concept.

Ralph

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System