Re: ARITH(EXTEND), all of our COBOL code has been compiled with that option turned on for several decades now, and I can't imagine a reason NOT to use it, especially if your company deals in banking or financial market data.
Re: performance, I would say rather that there is no performance IMPROVEMENT at ARCH(11) or below. Without the Vector Decimal hardware, you have only the 360-era packed decimal instruction set. Peter From: IBM Mainframe Discussion List <[email protected]> On Behalf Of Schmitt, Michael Sent: Thursday, July 11, 2024 9:41 AM To: [email protected] Subject: Re: Packed Decimal -- Extended(?) Are there disadvantages to ARITH(EXTEND)? Is the only reason to want intermediate results to be compatible with older code is if you need to match previous less-accurate results? Is there a performance impact at ARCH(11) or lower? -----Original Are there disadvantages to ARITH(EXTEND)? Is the only reason to want intermediate results to be compatible with older code is if you need to match previous less-accurate results? Is there a performance impact at ARCH(11) or lower? -----Original Message----- From: IBM Mainframe Discussion List <[email protected]> On Behalf Of Farley, Peter Sent: Thursday, July 11, 2024 1:13 AM To: [email protected] Subject: Re: Packed Decimal -- Extended(?) All true Tom, but as far as I understand it the Vector Decimal instructions still do not provide any more digits of precision than the older, non-vector ones. I believe the OP was asking about more digits of precision, not better CPU usage. COBOL's ARITH(EXTEND) option still provides only up to 31 digits of precision. And we still do not have COBOL access to the Decimal Float capabilities of the hardware. Extended-format Decimal Float provides (if I am reading PoOP correctly) up to 34 significant decimal digits of precision. Peter From: IBM Mainframe Discussion List <[email protected]> On Behalf Of Tom Ross Sent: Thursday, July 11, 2024 12:41 AM To: [email protected] Subject: Packed Decimal -- Extended(?) <Snipped> The IBM z/OS COBOL compiler handled long numbers with packed-decimal instructions for years by using library routines that would process parts of the data nad then combine the results. Modern COBOL compilers on modern hardware (IE: customers who have z14 or later as DR machines) can compile with ARCH(12) which tells the compiler that we can use Vedtor Packed Decimal instructions, that can not only process many digits, but can process pacekd-decimal arithmetic with up to 90% less CPU usage than traditional packed- decimal instructions! In any case, I think the answer is to use a newer COBOL compiler :-) Cheers, TomR >> COBOL is the Language of the Future! << -- This message and any attachments are intended only for the use of the addressee and may contain information that is privileged and confidential. If the reader of the message is not the intended recipient or an authorized representative of the intended recipient, you are hereby notified that any dissemination of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail and delete the message and any attachments from your system. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
