🚀 go-pugleaf

halil@ittpub.nl writes:

>to me (I prefer e-mail).

>Thanks,
>Halil Ozmen		halil@ittpub.nl

Hmmmm, that's interesting, the same thing happens to  my friend's computer
yesterday.

He upgraded from 16mb to 32 mb and the system doesn't seem any faster,
he's running memory intensive compilations under Linux.
His first 16mb is 70ns and the other 16mb is 80ns.
Could that pose a problem ??

Also, he has only 64kb of internal cache on the motherboard, would
upgrading to 256kb helps ?

Ben.

In <boonlow.765828469@sfu.ca> boonlow@kits.sfu.ca (Boon Chong Benjamin Low) writes:

>he's running memory intensive compilations under Linux.
>His first 16mb is 70ns and the other 16mb is 80ns.
>Could that pose a problem ??
Very unlikly. If RAM is too slow, he will experiance problems like parity
errors or kernel panics, not slowdown.

>Also, he has only 64kb of internal cache on the motherboard, would
>upgrading to 256kb helps ?
He needs 128kb cache for 32Mb ram. Each 32k bank of cache supports 8Mb of
Ram. This is only true for a 486, on a 386 the cache controller only uses
8k banks for 8Mb each.
If he has a 486 and does not encrease the cache it is very likely that his
system will be slow as a dog. If I got it right, Linux is one of the low-
loading operating systems, so the slowdown will not be so noticeable until
he loads enough tasks to fill up the first 16Mb. SCO Unix for instance
loads high, if your kernel resides within a non cached area of memory, you
will get the slowdown immediatly.

	MaM

-- 
* /\/\ichael /\ndreas /\/\eiszl | Tomsonstr 25 | 59071 Hamm | Tel+492381980131 *
* mam@mamunx.garmhausen.de	<EUNET     X.400      EUNET>    mam@werries.de *
*       \C=US \A=ATTMAIL \P=Garmhausen \O=GuPS \G=Michael \I=A \S=Meiszl       *

If the memory speeds are different (70ns and 80ns, for example), then 
you're asking for trouble.  Get the manual for the motherboard and 
install the speed memory that it specifies.  Mixing speeds leads to 
memory parity errors, which leads to all sorts of wierd things 
(spontaneous reboots and hangs, for example).  A 10ns difference may not 
seem like a lot to you, but since light travels about 9.3 inches in a 
nanosecond, you'd agree that 93 inches of light travel is quite a bit of 
distance in that 10ns time frame, especially considering the physical 
dimension of your RAM SIMM's.

Installing more memory will slow down your boot up time because your 
system will want to check all this memory (and so will HIMEM.SYS in 
MS-DOS 6.2).  Slower there, faster elsewhere.

Oh, the rough rule for external cache RAM is 128k for a 386, 256k for a 
486, and 512k for a Penitum.  That's a rough rule, though.  The average 
user will do okay with that, but anyone reading this ain't a normal user, 
so y'all might pop over the line.

Scott Purl
Calamity Factors -- Terra Simulations and Computing
sdpcart@selway.umt.edu

>If the memory speeds are different (70ns and 80ns, for example), then 
>you're asking for trouble.  Get the manual for the motherboard and 
>install the speed memory that it specifies.  Mixing speeds leads to 
>memory parity errors, which leads to all sorts of wierd things 
>(spontaneous reboots and hangs, for example). 

Excuse me, but this is nonsense.

Memory "speed" is not that at all. It is instead 
	The _maximum_ time required until data output is
	stable and valid, after an address change.

In other words:
	CPU places an address 0FE80786 on the address bus.
	x amount of time passes (say 80 ns)
	The output of that chip for that address is known to be valid.

(Yes, I left out lots of other factors such as decode time, etc, etc.
-- In this discussion they will just obscure the issue...)

Now, there is nothing that says the data can not be there BEFORE then,
say 79 ns. Nor is there any reason that you cannot go back & read the
data later, say 85, or 100 ns. The spec is the maximum; many will do
better.

In fact, the manufacturer selects out the faster chips & sells them as
such, so there will not be a large number of 70 ns one in 100 ns
packages. But that is because they can get you to pay more for them...

So long as ALL chips are at LEAST fast enough, you will be ok.
-- 
A host is a host from coast to coast.................wb8foz@nrk.com
& no one will talk to a host that's close...........(v)301 56 LINUX
Unless the host (that isn't close)....kibo# 777............pob 1433
is busy, hung or dead..............vr....................20915-1433

In article <1994Apr11.062202.14307@selway.umt.edu>, sdpcart@selway.umt.edu (Scott Purl) writes:
|> If the memory speeds are different (70ns and 80ns, for example), then 
|> you're asking for trouble.  Get the manual for the motherboard and 
|> install the speed memory that it specifies.  Mixing speeds leads to 
|> memory parity errors, which leads to all sorts of wierd things 
|> (spontaneous reboots and hangs, for example).  A 10ns difference may not 
|> seem like a lot to you, but since light travels about 9.3 inches in a 
|> nanosecond, you'd agree that 93 inches of light travel is quite a bit of 

Huh?? what does light have to do with anything?  does he have the first fibre-optic computer in the world or what.

If I recall correctly the speed of a simm is the MAX speed it has been tested good for.  that means that it can be run as a slower rate. i.e. a 70ns SIMM can run fine at 80ns.  but that is just what I have heard.



|> distance in that 10ns time frame, especially considering the physical 
|> dimension of your RAM SIMM's.
|> 
|> Installing more memory will slow down your boot up time because your 
|> system will want to check all this memory (and so will HIMEM.SYS in 
|> MS-DOS 6.2).  Slower there, faster elsewhere.
|> 
|> Oh, the rough rule for external cache RAM is 128k for a 386, 256k for a 
|> 486, and 512k for a Penitum.  That's a rough rule, though.  The average 
|> user will do okay with that, but anyone reading this ain't a normal user, 
|> so y'all might pop over the line.


big smile  : ^ )

In article <1994Apr11.062202.14307@selway.umt.edu>,
which was posted to comp.sys.ibm.pc.hardware, which has been obsolete for
half a year, and comp.ibm.pc.hardware, which doesn't exist on any site I
have access to, as well as to biz.comp.hardware,
Scott Purl <sdpcart@selway.umt.edu> repeated this time-worn
canard:
>If the memory speeds are different (70ns and 80ns, for example), then
>you're asking for trouble.  Get the manual for the motherboard and
>install the speed memory that it specifies.  Mixing speeds leads to
>memory parity errors, which leads to all sorts of wierd things
>(spontaneous reboots and hangs, for example).

Email to sdpcart@selway.umt.edu went unanswered.

There is no such thing as "matched" SIMMs.  Identically marked
SIMMs can be 20 ns different, and are likely to be at least 5 ns
different.  SIMMs marked a grade apart can be within a nanosecond
of each other.  That's because DRAMs are *always* faster than
they are marked.  They can be *much* faster if, for example,
the manufacturer's process is turning out nothing but 60 ns parts
but customers are ordering mostly 80 ns parts.
(This was explained to me by the man who manages the largest DRAM
plant in the world, who was visiting to try to sell me ASICs made
in the same plant.  You may have "heard" a different story from the
guy in the corner clone store.)

As I've said before, I believe what is really going on here is
a big scam.  Dealers are selling motherboards without testing them
properly.  Many of these low-cost boards cut corners and do not
meet the *highly standardized* requirements of DRAM SIMMs.
Their strobe transitions aren't monotonic.  The strobes ring.
The supply rails are noisy.  Pulsewidths aren't met.
Access times are cheated.  (I have interviewed with two
motherboard companies where I was asked which DRAM specs can
be cheated and which must be observed.  Scum!  Schlockers!)

These boards will usually work when the DRAM array is not fully
stuffed.  The problems begin when the user adds more RAM,
because that's when the signal integrity and time cheats show up.
Such a dog board often can be made to work by "cherrypicking" DRAMs,
that is by trying a few pieces until a "golden" set can be found
that can tolerate that particular motherboard's problems.

The dealer then faces a problem, how to get the customer to
pick cherries at the customer's expense.
Tell them "these SIMMs aren't matched, you need a matched set"
is often the answer.

This has been going on for so long that it's creeping into
OS documentation.  At least one commercial OS product prints
"mismatched RAMs" when its memory initialization fails.

A digital system well-designed for volume production will always
accept *faster* chips than were specified.  You can "mix" them in
the same "bank" or between "banks".
The exception to this rule is for motherboards which sense the
grade of 72-lead SIMMs and automatically configure "wait states" or
other timing behavior: these should have the slowest parts in
the socket that does the sensing.

Don't buy *anything* from a dealer who tells you stories about
"matched" SIMMs.  Ask them to show you oscilloscope traces.

>Scott Purl
>Calamity Factors -- Terra Simulations and Computing
>sdpcart@selway.umt.edu

I'd like to know *what* Scott simulates.  I hope it's nothing electronic.
Scott's advice to get the motherboard manual is good, though.

Cameron in San Jose, speaking only for myself.


-- 
Cameron Spitzer, San Jose CA   cls@truffula.sj.ca.us