Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!harpo!decvax!cca!ima!haddock!johna
From: johna@haddock.UUCP
Newsgroups: net.micro
Subject: Re: 4 -> 8 -> 8/16 -> 16 -> 16/32 -> 32 - (nf)
Message-ID: <133@haddock.UUCP>
Date: Mon, 19-Mar-84 23:37:39 EST
Article-I.D.: haddock.133
Posted: Mon Mar 19 23:37:39 1984
Date-Received: Wed, 21-Mar-84 01:39:54 EST
Lines: 13

#R:sdccs5:-116100:haddock:13200010:000:449
haddock!johna    Mar 19 12:32:00 1984

>>      Floating point ?  I don't do too much crunching, but I would expect
>>      64 bits of double precision on a 32 bit machine would give you plenty
>>      of precision.

 Floating point gives great RANGE but precision is not its strong suit.
64 bit IEEE floating point has a 53 bit mantissa, giving approx. 17 digits.
Any number  10^-17 < x < 10^17 will be exact, but anything else will only be
an approximation.

		 decvax!ima!haddock!johna