Re: Cycles per ASM instruction

In
Message-ID:<1172092795.418010.174730@[EMAIL PROTECTED]
>,
hancock4@[EMAIL PROTECTED]
 wrote: 

>>      I remember seeing them for the 370/115.  It's hard to forget
>> that moving from a 360/22 to a 370/115 gave us a CVD timing worse
>> by one to two orders of magnitude.  IIRC, the number of cycles for
>> that instruction was approx 250.
>
>
>Are you saying the 370-115 was slower than the 360-22?  (I didn't even
>know there was a model 22).

     The 22 was effectively a 30, but it had a max of 16KB, where
the 30 would go to either 32 or 64.  It's possible that there were
other options that couldn't be added to it (such as floating
point).

     To answer your question:  Yes and no.  For almost all
instructions, the 115 was faster than the 22, but not as fast as a
50 (as I recall).  However, for some of the instructions in the
slow microcode (such as the CVD), it was *much* slower.

     I was the applications programmer dubbed "sysprog" for the
upgrade, so I got to do a lot of this research.  Of course, it was
all ignored by management.

>To this day proper definition of subscripts and SEARCH arguments in a
>COBOL program will save processing time if the files have any length
>to them.  We learned the im****tance of which field to use, but I'm not
>sure all programmers were taught it or remembered it from class.
>(Many were taught but probably forgot.)

     How many times do we see a COBOL program with an array index
defined as COMP-3 or, worse, DISPLAY?

-- 
Arthur T. - ar23hur "at" intergate "dot" com
Looking for a z/OS (IBM mainframe) systems programmer position