PCBUILD Archives

Personal Computer Hardware discussion List

PCBUILD@LISTSERV.ICORS.ORG

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:
David Gillett <[log in to unmask]>
Reply To:
PCBUILD - PC Hardware discussion List <[log in to unmask]>
Date:
Fri, 13 Mar 1998 18:40:58 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (52 lines)
On 12 Mar 98 at 21:24, Rick Poepping wrote:

> First of all Thanks for the replies.
>
> David Gillett wrote:
>
> > On 26 Feb 98 at 22:58, Rick Poepping wrote:
> >
> > > I'm getting ready to do various upgrades to my 586-p133
> >
> >   This is not a very clear designation; no CPU manufacturer really
> > uses "586" as a designation.  [Some motherboard manufacturers use
> > this to identify motherboards designed for the Pentium (Socket 5 or
> > Socket 7).
> >
> >   My first thought was that maybe you have a 5x86-133, which is an
> > AMD version of the 486 with a 4x clock multiplier.
>
> This is correct.  I have an AMD 5x86-p133.  Sorry for the confusion

  There is probably a version of the "64MB cache span limit" that
applies to your system, but the magic number may be less than 64MB.

> >  > And what benefits and pitfalls are there to going beyond that
> > > amount (of RAM).
> >
> >   A chipset limited to caching 64MB of main RAM faces a problem when
> > more is installed.  One simple way out of the jam is to disable the
> > L2 cache entirely....
>
> Is it your opinion that one would be better off disabling the L2 cache
> all together if you install more RAM than the chipset and L2 combination
> can handle?  So I would assume thisa would mean you would get better
> performance by doing so (as opposed to leaving the L2 enabled but having
> more RAM than you can cache)?

  If turning off L2 is the only way the chipset has of dealing with
this problem, it should do that automatically.  You shouldn't need to
manually turn it off unless you see an actual problem with turning it
off.

  In theory, adding memory can reduce swapping to disk.  In practice,
those who've tried this and benchmarked performance indicate that the
execution speed penalty is about 5%, and you may have to be swapping
pretty heavily to make that pay off.

  My own rule of thumb, since DX2 days, is that if your CPU clock is
a multiple (i.e. more than 1x) of the board (RAM) clock, you want L2
cache or your CPU won't deliver the performance you've paid for.

David G

ATOM RSS1 RSS2