Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!uunet!mcsun!inesc!ajr
From: ajr@inesc.UUCP (Julio Raposo)
Newsgroups: comp.sys.ibm.pc
Subject: memory tests
Keywords: finding RAM
Message-ID: <209@inesc.UUCP>
Date: 22 Jan 90 16:06:09 GMT
Organization: INESC - Inst. Eng. Sistemas e Computadores, LISBOA. PORTUGAL.
Lines: 18


	I've seen an intresting algorithm to test the presence of memory
    (RAM). I think it supresses the problems associated with reading the
    same value in sequence and other bus problems.
	Here goes:

	- Store 2 in that memory location
	- decrement that value (8086 -> DEC Word Ptr[xxxx])
	- see if it is 0 (test the zero flag)
	- if it is 0 then the adress has no RAM
	- decrement again
	- see if it is 0
	- if it is 0 then the adress has valid RAM
	- else th adress has no RAM.

	It seem to have no problems, except the time it takes.

	Any comments?