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:
Dave Hendricks <[log in to unmask]>
Reply To:
PCBUILD - Personal Computer Hardware discussion List <[log in to unmask]>
Date:
Fri, 26 Feb 1999 10:32:06 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (63 lines)
> -----Original Message-----
> From: David Nasser [mailto:[log in to unmask]]
> Sent: Thursday, February 25, 1999 2:36 PM
> To: [log in to unmask]
> Subject: [PCBUILD] Samsung/Mushkin SDRAM
> 
> 
> Has anyone had experience with Mushkin cas2 SDRAM with Samsung GH
> chips (chip # KM48S8030-GH) TAC 6 ns ???
> 
> I've heard the Samsung chips have an excellent reputation. I had not
> heard of Mushkin 'till this week.

Anand Lal Schimpi of Anandtech (www.anandtech.com) seems to think Mushkin
makes a good DIMM; his test seem to validate that claim as well.

Samsung Has historically been a important DRAM manufacturer; as they were
the first to ship 16Mb/64Mb/256Mb Desity SDRAM TSOPs
http://www.usa.samsungsemi.com/.  I do find it peculiar that Samsung doesn't
have that part number (BTW TSOP is the type of package the chip comes in) in
a 6ns speed grade listed on their webpage.  but things change maybe it just
isn't updated.
Also that is a 64Mb TSOP I'd thing if Samsung was going to make a 6ns part
they'd do it on their 256 Mb Product line.  Especially with the slew of new
Graphics Chips (that are going to require sub 6ns SDRAM; Riva TNT2, Voodoo3,
Savage4, Permedia3) coming to market this quarter.

I think what alot of people forget is that the quality of the PCB board has
as much to do with the quality of the DIMM as the chips on the.  Strangely
enough alot of the troubles that Intel and RAMBUS are having with RIMMs has
to do with PCB Traces having too much impedance and that the PLL chips used
aren't "tight" enough (ie the jitter budget is exceeded because of clock
skew).

A 6ns Modules should qualify for beyond 100 MHz Synchronous speeds.  Its
prolly CAS3 for 133 MHz though if it could even do it.

I do have one caveat though; that is that CAS2 and CAS3 are pretty much a
wash.

My own testing using Linux 2.2.1 and hdparm's timing of Buffered Cache
Readings (hdparm -T /dev/hda) shows me that CAS2 memory can do about 101.6
MB/sec and CAS 3 98.31 MB/sec.  hardly worth the invest in CAS2 SDRAM IMHO.

Micron ( http://www.micron.com/mti )seems to have the same feeling that CAS2
is not worth it. Which is very interesting because they stand to profit huge
if they would say otherwise.
I had a datasheet that I downloaded from Micron (but I can't find it anymore
you know the thing) rating Micron's Inhouse performance data of their -8A,
-8B, -8C, -8D, -8E, Speed grade disignations.

I couldn't find that datasheet I did find though that Micron has put up a
PC133 Spec sheet though :)

¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤øøº°`°º¤¤º°`°º¤ø,¸¸,ø

Dave Hendricks:  Linux-Junkie; MCP; Systems Integrator.  
[log in to unmask]

                Curious about the people moderating your
                   messages? Visit our staff web site:
                     http://nospin.com/pc/staff.html

ATOM RSS1 RSS2