Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ncar!ico!vail!rcd
From: rcd@ico.isc.com (Dick Dunn)
Newsgroups: comp.sys.ibm.pc
Subject: Re: 286 clone
Summary: static RAM explained
Keywords: static_RAM, dynamic_RAM
Message-ID: <1989Nov9.234742.1647@ico.isc.com>
Date: 9 Nov 89 23:47:42 GMT
References: <112@oiscola.Columbia.NCR.COM> <7321@ingr.com>
Organization: Interactive Systems Corporation
Lines: 27

In a note about some other stuff, fordke@ingr.com (Keith Ford x8614) writes:
> >Also what is what is the difference between static and dynamic RAM?
> 
> Most simply from an application point of view, the static RAM can maintain
> it's data when connected to a low current source such as a battery.  This
> allows data stored there to remain, even after the machine is turned off.

Not entirely wrong, but off-base.  Information which needs to be maintained
with the power off is often stored in CMOS RAM, which is static RAM.  CMOS
is a semiconductor technology notable for very low power consumption, hence
a battery can maintain it.

However, what the "static" really means is that if you keep power on the
memory, it will "remember".  There are SRAMs which are not at all
low-power.  The "static" contrasts with "dynamic"--DRAMs are what are used
in almost all regular main memory systems--e.g., the memory on your PC.
DRAMs may actually require less power than SRAMs.  The "dynamic" refers to
the fact that the information has to be refreshed periodically.  In effect
a DRAM remembers by storing charge, but the charge tends to leak out.  So
you refresh the memory by reading what's there (before it all leaks out)
and rewriting it.  If you're a software type, you don't have to worry about
any of this...it's all behind the scenes; there's hardware to do the
refresh.  (The interval between refreshes is short--on an AT, each piece of
memory is refreshed about every 4 milliseconds.)
-- 
Dick Dunn     rcd@ico.isc.com    uucp: {ncar,nbires}!ico!rcd     (303)449-2870
   ...Keep your day job 'til your night job pays.