Instruction timing varies drastically due to multi-level cache, parallel and out of order execution, model dependent facilities, etc. You could have best-case and worst-case formulas. But I expect an average-case formula would be too data dependent to be useful.
-----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Shmuel Metz (Seymour J.) Sent: Tuesday, December 18, 2012 3:05 PM To: [email protected] Subject: Re: 64 Bit Common Storage (was Common Data Space Basics) In <[email protected]>, on 12/18/2012 at 06:11 AM, Donald Likens <[email protected]> said: >On the other hand it would be nice to know the relative speed of every >instruction. I doubt that there is such a thing. The last time that IBM published timing formulæ, simple numbers had given way to expressions; I'm sure that the relevant formulæ have grown more complex for current hardware rather than simpler. Even in the old days, you'd have to know the model to rank execution time of, e.g., SLR, SR, XR. -- Shmuel (Seymour J.) Metz, SysProg and JOAT Atid/2 <http://patriot.net/~shmuel> We don't care. We don't have to care, we're Congress. (S877: The Shut up and Eat Your spam act of 2003) ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
