MAPS-L Archives

Maps-L: Map Librarians, etc.

MAPS-L@LISTSERV.UGA.EDU

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Dennis Boone <[log in to unmask]>
Reply To:
Maps and Air Photo Systems Forum <[log in to unmask]>
Date:
Wed, 23 Mar 1994 19:47:36 EST
Content-Type:
text/plain
Parts/Attachments:
text/plain (46 lines)
----------------------------Original message----------------------------
Pardon me for being a bit grouchy, but this is sort of my turf.
 
On Wed, 23 Mar 1994 11:59:54 EST DEBBIE LORDS said:
>     Now that I have your attention.  According to a magazine article I
>recently read (last month), the 66 mhz machine runs at 66 for internal
>processing only, like doing a dir or a cd.  But is that important?  After
>all, how fast can you read a whizzing directory listing?  (And do you want
>to?)
>     When it comes to the workhorse needs, like data i/o, the machine runs at
>a whopping 33 mhz.  Great.  And we all know that what we are interested it is
>using DATA, some of it intensely slow for slower machines.
 
The DIR and CD commands are doing disk I/O just like your application is.
 
The "internal" stuff that IS faster is the execution of individual machine
instructions.  In computational performance, the DX2-66 processors do have a
small advantage over the DX-50 processors.  I/O performance won't be any
worse for a DX2-66 than a DX-50.
 
>     A 50 mhz machine runs at 50 for everything.
 
No.  Most of the I/O you're concerned about is done over the computer's I/O
bus at 8mhz, or perhaps 12mhz, unless you have a local bus disk controller.
Besides, The VESA local bus is not designed to run at faster than 33mhz (the
local bus speed of a DX2-66) anyway, so you aren't gaining anything by having
a DX-50 processor over a DX2-66 processor.
 
>     Also Pentium has not been the "saviour" that was hinted at.  Walk
>carefully, Pat.  I'm not saying that Pentium is bad.  I'm suggesting that
>Pentium has not yet proven itself to those who closely watch hardware
>development.  For all the hoop-lah that preceded it, its been amazingly quiet
>ever since.  That gets me a little nervous.
 
One of my colleagues recently did a comparison of systems.  He compared
machines which are roughly in the same performance class as the Pentium, and
concluded that a P5-based machine has roughly the same performance as a
variety of RISC-based low-end workstations, including low-end Alpha and
PowerPC based systems.  The advantage of the Pentium over these systems is
$200-$4000 in price.  While some of the more extravagant claims about Pentium
performance may well be false, a P5-66 will definitely be a better performer
than a DX2-66.  One drawback of a Pentium is the ever-present heat problem,
but a simple $10 cooling fan cures that.
 
Dennis

ATOM RSS1 RSS2