home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cs.utexas.edu!ut-emx!tivoli!foraker!taylor
- From: taylor@foraker.NoSubdomain.NoDomain (Eric Taylor)
- Newsgroups: comp.theory
- Subject: Re: Real Numbers vs. Rational Numbers?
- Message-ID: <7111@tivoli.UUCP>
- Date: 19 Dec 92 20:24:25 GMT
- References: <1992Dec16.095412.19570@tom.rz.uni-passau.de>
- Sender: news@tivoli.UUCP
- Lines: 19
-
- In article <1992Dec16.095412.19570@tom.rz.uni-passau.de>, boerncke@kirk.fmi.uni-passau.de (Frank-Roland Boernke) writes:
- |> In lectures we talked about the RAM, a formal machine-model that is proved
- |> to be as powerful as turing-machines(TM) and vice versa, according to the
- |> Church-Turing-Thesis. This seems to be ok until you say, that the RAM allows
- |> only natural numbers in its registers (or rational-numbers, since they are
- |> countable). Our professor told us, that RAMs are allowed to store REAL numbers
- |> too AND they are still as powerful as an ordinary turing machine. I claimed, this
- |> can`t be true, since there isn`t any possibility to work with irrational
- |> numbers in TMs.
-
- REAL numbers are always RATIONAL numbers and vice-versa.
- The terminology "REAL number", I believe, is not a formal
- one, but one borrowed from something like FORTRAN where
- they are simply meant to represent floating point
- numbers.
-
- Any rational number can be represented by the division
- of 2 integers and are therefore easily handled by a
- RAM or Turing machine.
-