home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!stanford.edu!agate!spool.mu.edu!vms.csd.mu.edu!5916RAHMANK
- From: 5916rahmank@vms.csd.mu.edu ( )
- Newsgroups: comp.os.ms-windows.misc
- Subject: Re: DOES A 387 HELP WIN/3.1?
- Date: 3 Jan 1993 17:59:03 GMT
- Organization: Department of EECE, Marquette University
- Lines: 65
- Message-ID: <009660E6.DDCF66E0@vms.csd.mu.edu>
- References: <michael.726006583@glia>,<iH6swB2w165w@salzo.Cary.NC.US>
- Reply-To: 5916rahmank@vms.csd.mu.edu
- NNTP-Posting-Host: vmsf.csd.mu.edu
-
- In article <iH6swB2w165w@salzo.Cary.NC.US>, charvel@salzo.Cary.NC.US (Jeff Cogswell) writes:
- ->michael@glia.biostr.washington.edu (Michael) writes:
- ->
- ->> In <1992Dec29.151718.6115@mksol.dseg.ti.com> mccall@mksol.dseg.ti.com (fred j
- ->> >In <1992Dec29.004615.20098@netcom.com> ergo@netcom.com (Isaac Rabinovitch) w
- ->> >>In <1992Dec28.185555.16798@mksol.dseg.ti.com> mccall@mksol.dseg.ti.com (fre
- ->> >>>In <1992Dec22.194817.25666@kodak.kodak.com> thomas@acadia.Kodak.COM (Thoma
- ->> >>>> "A Friend" told me that an 80387 would help Windows/3.1 run faster
- ->> >>>> because there are a lot of floating point calculations involved.
- ->> >>>> Can anyone verify this?
- ->> >>>No, because it isn't true. A math coprocessor will do absolutely
- ->> >>>nothing to improve Windows performance.
- ->>
- ->> >>*Almost* true.
- ->>
- ->> >Well, no. What I said is exactly true.
- ->>
- ->> Ah but he was referring to what the first person said. And he meant that
- ->> the idea had merit but didn't apply in this case. You are being over
- ->> defensive.
- ->>
- ->> True, but this explanation does serve to explain why the reasoning may
- ->> be sound in many cases but not in this one. Relax. Just because someone
- ->> answers a question differently, doesn't mean that you need to defend your
- ->> first answer. No one attacked you. :)
- ->>
- ->> Michael Stanley (michael@glia.biostr.washington.edu)
- ->
- ->I'm posting this rather than mail it so everyone can see. I'm sure
- ->there are many other people who get sick of reading such defensive
- ->postings. Michael's right. "No one attacked you." The fact is,
- ->Windows is so large, even Petzold doesn't know everything there is
- ->about it. When someone says "I'm right" he/she is just asking for
- ->trouble. There's always exceptions to the rule, especially in a
- ->complex multitasking environment. Besides, unless the people claiming
- ->to be exactly right have completely traced through the assembly code
- ->comprising Kernal, GDI, User, and all other modules, then they can
- ->not say with certainty that they are exactly right.
- ->
- ->Come on, everyone. We all have weaknesses.
- ->
- ->Jeff Cogswell
- ->charvel@salzo.Cary.NC.US
- ->
- --------------------------------------------------
-
- The issue of math coprocessor should not be the case of
- a square peg and round hole situation. BTW, this has been
- discussed in a long thread before and I recollect the outcome
- was that - a coprocessor does help speed up windows performance.
- A very simple experiment that I did to find out this fact is
- the following:
-
- a) Take a 386 pc/clone with a math coprocessor installed. Load
- windows and measure the time it takes to load. Repeat the
- measurement of time for few other of your favourite applications
- WfW, excel, CAD, ... Don't forget to note them down.
-
- b) Now open the box and pull the coprocessor off (make sure you
- ground yourself properly). Run the windows again and repeat the
- steps of a) by noting the time taken in the respective application.
-
- The role of the coprocessor becomes obvious!
-
- -Anis.
-