Experimental Physics and
Industrial Control System

Ralph Lange <[email protected]> · Mon, 1 Mar 1999 11:15:47 +0100 (MET)

> I am really confused about the initial values of ASLO for ai and ao
> records.

You are right - the situation is a bit confusing.

> In the record reference manual it says that the
> initial value for ai and ao are 1.0.
> [...]
> The results are:
> ASLO has a default of 0   for ao records and
> 	 a default of 1.0 for ai records.

Right. But a glance at the code might clarify the situation:

aoRecord.c (~line 192):

       case(0): /* convert */
            value = (double)pao->rval + (double)pao->roff;
            if(pao->aslo!=0.0) value *= pao->aslo;
            value += pao->aoff;

aiRecord.c (~line 365):

        val = (double)pai->rval + (double)pai->roff;
        /* adjust slope and offset */
        if(pai->aslo!=0.0) val*=pai->aslo;
        val+=pai->aoff;

So there's no real difference between ASLO being not set, 0.0 or 1.0.
(Actually ASLO being 0 saves a few usec for the multiplication.)

> I am putting 1.0 as the value of ASLO in all the capfast symbols
> that we are using for both ai and ao records, which seems to be the
> right way to go to me.

Generally agreed.

We could actually have a default of 1.0 for ao records in the dbd file and
get rid of the test against zero in the code.
Or we could have a default of 0 for both record types and define the value
0 for ASLO as "don't use it". (This would need a word in the doc.)

Should we do something? Is this worth the effort? *Shrug.*

Ralph

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System