Path: utzoo!mnetor!uunet!lll-winken!lll-lcc!rutgers!husc6!cca!mirror!datacube!dje
From: dje@datacube.UUCP
Newsgroups: sci.electronics
Subject: Re: Looking for memory part
Message-ID: <100500025@datacube>
Date: 30 Dec 87 17:00:00 GMT
References: <1513@mipos3.intel.com>
Lines: 31
Nf-ID: #R:mipos3.intel.com:-151300:datacube:100500025:000:1499
Nf-From: datacube.UUCP!dje    Dec 30 12:00:00 1987


I  understand  that  the  requirements  for parts  differ when you're
building one of something than when you're  designing a  product.  We
use lots of RAMs in image memory applications.  Cost and availability
are prime issues here. We build fast image memories. 

Nibble 256K X 1's are readily available.  Nibble  64K X  4's are not,
though.  We designed them in about 2 years ago.   At the  time we had
three data sheets  and production  parts from  Hitachi.   Then a year
later, the Hitachi source dried up, and no-one else really made them.
Seems we and one other manufacturer were the  only significant users,
so Hitachi stopped making them.   We changed  the design  to use fast
page parts.  This requires presenting addressing on subsequent CAS's.

Regarding Static Column DRAMs, Intel made announcements with a flurry
(Nice glossy  brochures), handed  out samples,  got everyone excited,
and dropped the parts.   (Designus  interruptus).   No-one has really
pushed Static Column hard since, but you may find a  source for them.
Also, static column parts  are CMOS,  meaning they  will usually cost
more.  

Fast page *almost* eliminates the need for those other guys: They're
as fast as nibble and almost as fast as Static column.

 				Dave Erickson
----------------------------------------------
  ------      Datacube Inc. 
 /    /|      4 Dearborn Rd. 
------ |      Peabody, Ma 01960
||  \| |      ihnp4!datacube!dje
||  /|/       Human:(617)535-6644
------        Fax:  (617)535-5643