RE: More cycle, yes, more cycle

[Joe Giusto <JGiusto@xxxxxxxxxxxx>]

Does anyone know what the 400 thinks about the cycle?  I mean is it any more
or less efficient in terms of processing resources to use or not use the
cycle method.  

I've heard it said that non-cycle was better on the 38 and the CICS and that
when we got our RISC box that using the cycle again would be better.  I am
not sure of the authority of that statement, I just do what the boss told me
unless I could prove a case otherwise.

Joe Giusto II
Patuxent Publishing Company
 <mailto:jgiusto@patuxent.com> mailto:JGiusto@patuxent.com
http://www.lifegoeson.com <http://lifegoeson.com> 


        -----Original Message-----
        From:   david.kahn@gbwsh.mail.abb.com
[SMTP:david.kahn@gbwsh.mail.abb.com]
        Sent:   Tuesday, April 06, 1999 1:35 PM
        To:     MIDRANGE-L@midrange.com
        Subject:        RE: More cycle, yes, more cycle

        Colin Williams <Williamsc@technocrats.co.uk> wrote:

        >Cycle Processing simple. I only ever find it simple
        >when used to do very straight-forward updates or
        >reports. Introduce level breaks and matching records,
        >and everything goes doo-lally

        Only if you don't do it right. I happen to think that the dreaded
level
        break and matching records features are the real strengths of the
cycle,
        and guarantee that things do _not_ go doo-lally. I've seen a number
of
        examples of real life matching routines written in structured code
that
        hiccupped because the programmer overlooked some of the subtle traps
that
        lie there and can squeeze past testing. Once upon a time all our
master
        files were on tape and every programmer worth his or her salt could
rip off
        a matching records program (having first produced 6 pages of
flowcharts).
        Nowadays, however, programmers are less familiar with the process
and more
        likely to make mistakes. The cycle guarantees that the matching
logic is
        correct.

        I do agree with Colin that the cycle is also good for small simple
        programs, too. I often use it in one off programs that will never
see the
        light of day again. In this case the arguments about confusing the
next
        programmer do not apply. During the Easter weekend just gone we were
doing
        some major conversion work. I must have written a dozen QND programs
during
        the weekend and about a third of them used the cycle. In one case it
was a
        personal challenge. "That could be done in 4 lines of RPG," said my
        colleague John Tisot. In fact it took 5, but he couldn't shorten it
either.
        Without the cycle it would have been a fat software monster of at
least 8.
        :-)

        Dave Kahn, ABB Steward Ltd.


        +---
        | This is the Midrange System Mailing List!
        | To submit a new message, send your mail to
MIDRANGE-L@midrange.com.
        | To subscribe to this list send email to
MIDRANGE-L-SUB@midrange.com.
        | To unsubscribe from this list send email to
MIDRANGE-L-UNSUB@midrange.com.
        | Questions should be directed to the list owner/operator:
david@midrange.com
        +---
+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---