🚀 go-pugleaf

In article <2nr155$j1p@char.vnet.net> merlin@char.vnet.net (Mystical Wizard) writes:

>I've had my ZEOS 486DX2-66 for about 15 months now.  It would be fair
>to say that I have my fair share of problems with this system.  ...
>(Much deleted regarding problems with cache memory)

A lot of the problems with ZEOS systems appear to deal with the optional cache
memory.  My ZEOS DX2-66 system (no cache except the 8k on the 486 chip) has
been trouble-free for over a year, and plenty fast for my needs.

I can't help but wonder what the extra cache would do for performance, though.

Does anyone have benchmark numbers with and w/o the optional cache,
or with 128k vs. 256k?  Is it really worth the extra ~$250?

In article <jackerman.4.2DA1D278@p01.mn15.honeywell.com>, jackerman@p01.mn15.honeywell.com (John Ackerman) writes:
|> In article <2nr155$j1p@char.vnet.net> merlin@char.vnet.net (Mystical Wizard) writes:
|> 
|> >I've had my ZEOS 486DX2-66 for about 15 months now.  It would be fair
|> >to say that I have my fair share of problems with this system.  ...
|> >(Much deleted regarding problems with cache memory)
|> 
|> A lot of the problems with ZEOS systems appear to deal with the optional cache
|> memory.  My ZEOS DX2-66 system (no cache except the 8k on the 486 chip) has
|> been trouble-free for over a year, and plenty fast for my needs.
|> 
|> I can't help but wonder what the extra cache would do for performance, though.
|> 
|> Does anyone have benchmark numbers with and w/o the optional cache,
|> or with 128k vs. 256k?  Is it really worth the extra ~$250?


	It apperars that we have our first frequently asked 
	question on alt.sys.pc-clone.zeos.  :)  I asked this 
	same thing about a month ago.  And considering the 
	low volume here . . . .

	I recall the Zeos techs telling me that the performance 
	increase would amount to about 10 per-cent in ms-windows,
	if WinTach were used.

	I'm interested in what kind of benefit I'd be getting 
	in X under Linux and XFree.  

	Anyone from Zeos reading this?

	Steve Whitlatch
	swhitlat@prism.nmt.edu

jackerman@p01.mn15.honeywell.com (John Ackerman) writes:

>I can't help but wonder what the extra cache would do for performance, though.

>Does anyone have benchmark numbers with and w/o the optional cache,
>or with 128k vs. 256k?  Is it really worth the extra ~$250?


All my figures are subjective.  I can say that during the diagnostic
phases of getting my "memory" problems fixed, I ran for a while without
the memory and my perceived performance was horrible.  Therefore,
I must assume that you would get a great boost with cache memory.

*IF* you have the newer Martin mainboard, you should be able to purchase
the memory chips from other source, for less.

The part numbers from two different chips are:

TOSHIBA 9348HAK 
TC55328AP-15  Japan


AND

9319 Z USA
MT5C2568-15


You will need 9 chips if you have 32MB of ram or less.
You will need 10 chips if you have MORE than 32MB of ram.
Just plug in the chips and move a few jumpers.
-- 
merlin - mystical wizard - merlin@vnet.net

In article <1994Apr6.041427.7079@nmt.edu>,
Steven Whitlatch <swhitlat@gray.nmt.edu> wrote:

[If a cache were added:]

>	I'm interested in what kind of benefit I'd be getting 
>	in X under Linux and XFree.  

There are other cacheless Zeos Linux users out there wondering the
same thing.  (I am one.)  Wouldn't the improvement be less under
Linux than DOS, though?

--
-Russell Webb
rwebb@panix.com

Steven Whitlatch (swhitlat@gray.nmt.edu) wrote:

: |> Does anyone have benchmark numbers with and w/o the optional cache,
: |> or with 128k vs. 256k?  Is it really worth the extra ~$250?

: 	I recall the Zeos techs telling me that the performance 
: 	increase would amount to about 10 per-cent in ms-windows,
: 	if WinTach were used.

: 	I'm interested in what kind of benefit I'd be getting 
: 	in X under Linux and XFree.  

: 	Anyone from Zeos reading this?

   It is ZEOS policy that we do not quote benchmarks on our systems.
We leave benchmarking to third parties (such as magazine reviews).  I
can say that the more CPU intensive your application, the more benifit
you will get from installing a cache.  If you are using disk intensive
applications, a CPU cache will not increase your performance as much.

-- 
David Mortenson         |                        | ZEOS International, Ltd.  
support@zeos.com   INET | Any opinions expressed | Technical Support Dept.
uunet!zeos!support UUCP |   herein are my own    | 1301 Industrial Blvd N.E.
612-633-4607       FAX  |                        | Minneapolis, MN 55413

In article <2nvotu$8r6@panix2.panix.com>, rwebb@panix.com (Russell Webb) writes:
>In article <1994Apr6.041427.7079@nmt.edu>,
>Steven Whitlatch <swhitlat@gray.nmt.edu> wrote:
>
>[If a cache were added:]
>
>>	I'm interested in what kind of benefit I'd be getting 
>>	in X under Linux and XFree.  
>
>There are other cacheless Zeos Linux users out there wondering the
>same thing.  (I am one.)  Wouldn't the improvement be less under
>Linux than DOS, though?
>

The cache will affect all OS's.  By way of example, a 486 with 32MB RAM
was incorrectly configured to cache only the first 16MB.  We saw
benchmark deterioration on the order of 2 depending on whether the
application was using the cached or uncached RAM.  (this was under OS/2
2.11).  That is, when the OS happened to load the application into the
cached RAM it ran almost twice as fast as when loaded into the uncached
RAM.  These randomly varying results of our benchmark were very
puzzling until we figured out what was really going on.  In any case,
the conclusion is clear, the cache is very important.


-- 
Jon Hacker                         |   OS/2 2.1
Bellcore, Red Bank NJ              |   accept no substitutes 
hacker@patagonia.bellcore.com      |   

In article <CnwCK1.9F9@walter.bellcore.com>, hacker@patagonia.bellcore.com (Jonathan Hacker 21420) writes:
|> In article <2nvotu$8r6@panix2.panix.com>, rwebb@panix.com (Russell Webb) writes:
|> >In article <1994Apr6.041427.7079@nmt.edu>,
|> >Steven Whitlatch <swhitlat@gray.nmt.edu> wrote:
|> >
|> >[If a cache were added:]
|> >
|> >>	I'm interested in what kind of benefit I'd be getting 
|> >>	in X under Linux and XFree.  
|> >
|> >There are other cacheless Zeos Linux users out there wondering the
|> >same thing.  (I am one.)  Wouldn't the improvement be less under
|> >Linux than DOS, though?
|> >
|> 
|> The cache will affect all OS's.  By way of example, a 486 with 32MB RAM
|> was incorrectly configured to cache only the first 16MB.  We saw
|> benchmark deterioration on the order of 2 depending on whether the
|> application was using the cached or uncached RAM.  (this was under OS/2
|> 2.11).  That is, when the OS happened to load the application into the
|> cached RAM it ran almost twice as fast as when loaded into the uncached
|> RAM.  These randomly varying results of our benchmark were very
|> puzzling until we figured out what was really going on.  In any case,
|> the conclusion is clear, the cache is very important.
|> 
|> 
|> -- 
|> Jon Hacker                         |   OS/2 2.1
|> Bellcore, Red Bank NJ              |   accept no substitutes 
|> hacker@patagonia.bellcore.com      |   

	The most compute-intensive task I could find, a SPICE simulation 
running on a RAMdrive (minimal video usage, no hard disk access after 
start of simulation) runs 20% faster with 256K cache installed on my 
486DX-33 (Gosling motherboard).  For most "ordinary" tasks, the difference 
is well under ten percent (that is, unnoticeable unless you get out your 
stopwatch).  For faster processors the difference would presumable be 
greater.

-----
Greg Thoman:  The opinions expressed herein are mine alone, and I am 
              solely irresponsible for them.

In article <jackerman.4.2DA1D278@p01.mn15.honeywell.com> jackerman@p01.mn15.honeywell.com (John Ackerman) writes:
>In article <2nr155$j1p@char.vnet.net> merlin@char.vnet.net (Mystical Wizard) writes:
>
>A lot of the problems with ZEOS systems appear to deal with the optional cache
>memory.  My ZEOS DX2-66 system (no cache except the 8k on the 486 chip) has
>been trouble-free for over a year, and plenty fast for my needs.
>
>I can't help but wonder what the extra cache would do for performance, though.
>
>Does anyone have benchmark numbers with and w/o the optional cache,
>or with 128k vs. 256k?  Is it really worth the extra ~$250?

	Like others, I've seen a 10% increase in general.  I only went
to 128K cache though.

	The newer "martin" motherboards are nicer since you can
by memory from anyone.  That's cheaper.

	But like you I'm stuck with the Gosling.

	Here is a question for Dave:  Will the Gosling support
the same SCSI chip as the Martin, ie. the better SCSI support
and SCSI-II support?


-- 
--
Greg d. Moore           | All that glitters is not gold.
Green Mountain Software |
Troy NY, 12180          | (Also strider@acm.rpi.edu)

First we read that:

|> The cache will affect all OS's.  By way of example, a 486 with 32MB RAM
|> was incorrectly configured to cache only the first 16MB.  We saw
|> benchmark deterioration on the order of 2 depending on whether the
|> application was using the cached or uncached RAM.  (this was under OS/2
|> 2.11).  That is, when the OS happened to load the application into the
|> cached RAM it ran almost twice as fast as when loaded into the uncached
|> RAM.  These randomly varying results of our benchmark were very
|> puzzling until we figured out what was really going on.  In any case,
|> the conclusion is clear, the cache is very important.
|> 
|> 
|> -- 
|> Jon Hacker                         |   OS/2 2.1
|> Bellcore, Red Bank NJ              |   accept no substitutes 
|> hacker@patagonia.bellcore.com      |   

Then we also read that

|> 	The most compute-intensive task I could find, a SPICE simulation 
|> running on a RAMdrive (minimal video usage, no hard disk access after 
|> start of simulation) runs 20% faster with 256K cache installed on my 
|> 486DX-33 (Gosling motherboard).  For most "ordinary" tasks, the difference 
|> is well under ten percent (that is, unnoticeable unless you get out your 
|> stopwatch).  For faster processors the difference would presumable be 
|> greater.
|> 
|> -----
|> Greg Thoman:  The opinions expressed herein are mine alone, and I am 
|>               solely irresponsible for them.


So what's the scoop!?

Well, actually the effect of the L2 cache is MUCH more pronounced in a DX2
then a DX. An intuitive reason is that a cache miss in a DX2 will stall
twice as many CPU cycles and hence takes a much greater performance penality.

On my DX2-66 running Linux, I get about 70% speed increase from enabling the
L2 cache write through cache, and then another 10% increase from switching to
using write back cache (both 256K direct mapped.) 

On the other hand, my brother who has a DX-33 also with 256K L2 cache told me 
that his observed increase is only about 18-20%. 

In both cases, the improvement estimates were real time estimates obtained from
compiling the Linux kernel using gcc, so they include both ram and disk access
overheads as well. 

Obviously different configurations (ram size, disk speed, bus speed etc) will
exhibit different speed up factors. The data points here though, hopefully
would represent typical ball park figures for those who are interested....

-- John

>
>	The newer "martin" motherboards are nicer since you can
>by memory from anyone.  That's cheaper.
>

OK, that's great.  2 questions, thought.  

1.  How do I tell if I've got a martin motherboard?

2.  What memory do I need to buy to go from 128K to 256K cache?

Thanks for the help.

Paul

-- 
--------------------------------------------------------------------------
dCode        | dcode@netcom.com            |    NeXT Mail Happily Accepted
Paul Marcos  | (415) 960-3259 (voice & fax)|