Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site brl-tgr.ARPA
Path: utzoo!linus!philabs!cmcl2!seismo!brl-tgr!tgr!cdh@BBNCD3.ARPA
From: cdh@BBNCD3.ARPA (Carl D. Howe)
Newsgroups: net.micro
Subject: a useful statistic to remember regarding memory size
Message-ID: <10320@brl-tgr.ARPA>
Date: Wed, 1-May-85 10:04:45 EDT
Article-I.D.: brl-tgr.10320
Posted: Wed May  1 10:04:45 1985
Date-Received: Fri, 3-May-85 08:23:34 EDT
Sender: news@brl-tgr.ARPA
Lines: 17

Here's a useful statistic for people trying to guage whether "large"
RAMs are cost effective:  Assuming $5 per 256K RAM, you can now
fully populate the address space of a 68000 (i.e. have a 16 MByte
main memory) for a chip cost of $2560 (i.e. less than a MAC XL).
If you want byte parity, the price goes up to $2880.

Remember when 4K cost more than that?

If you are now designing a general-purpose microcomputer product for sale in
the next few years, and you have no way to use that amount of memory, perhaps
it is time to think about a more modern approach.  Old measurements of what is
cost-effective may no longer apply.

Carl

P.S.  I understand chip cost is not the only factor relating to cost of memory,
but until recently, it certainly dominated.