Xref: utzoo comp.lang.misc:3504 comp.arch:11489
Path: utzoo!attcan!uunet!ginosko!gem.mps.ohio-state.edu!wuarchive!udel!rochester!dibble
From: dibble@cs.rochester.edu (Peter C. Dibble)
Newsgroups: comp.lang.misc,comp.arch
Subject: Re: Fast conversions, another urban myth?
Message-ID: <1989Sep24.014748.4245@cs.rochester.edu>
Date: 24 Sep 89 01:47:48 GMT
References: <832@dms.UUCP> <688@UALTAVM.BITNET>
Reply-To: dibble@cs.rochester.edu.UUCP (Peter C. Dibble)
Followup-To: comp.lang.misc
Organization: U of Rochester, CS Dept, Rochester, NY
Lines: 19

In article <688@UALTAVM.BITNET> ECULHAM@UALTAVM.BITNET writes:
>In article <832@dms.UUCP>, albaugh@dms.UUCP (Mike Albaugh) writes:
> <<maybe hardware decimal support would go 10% faster than software>>

>I've clearly exaggerated the performance gain of the decimal hardware.
>Yet even with that, all that extra hardware cannot even double the
>performance of the system.
> 
>Clearly, decimal arithmetic is one of those high cost, low payback
>extensions. We should direct our efforts elsewhere.

How much would that extra hardware cost?  Would it double the cost of
a microprocessor?  Add 10 % to the cost of a mainframe's CPU?

Computers that run Cobol are often heavily loaded with expensive I/O (and
memory and ...).  The processor would have to go up in price a _lot_ to 
make as much difference as a 10 % improvement in decimal speed.

Peter Dibble