Re: This calculation is just wrong / computer can't count!

I can only throw up my hands and say "This person is obviously stupid". You persist in
using the nonsensical phrase "unreliable" to describe the low-order digit. It is
understood by anyone who has ever done any computation in any form of arithmetic,
including paper-and-pencil, that when a finite precision is involved, the low-order digit
of a computation involving decimal computations is only accurate to ±1/2 digit. The
computations you are getting ARE MATHEMATICALLY RELIABLE!!!!! THEY ARE NOT INACCURATE!!!!
To within ±1/2LSB they are as precise as you are ever going to get. You will NEVER get
accuracy better than ±1/2 least signficant digit in ANY decimal computation that you do,
EVER. I don't care whether you are using binary or decimal arithmetic, as soon as you put
a limit on the number of digits, you are going to see a statistical error of ±1/2 least
significant digit.

I, and a lot of other people, have been trying to explain fundamental concepts of
mathematics to you for days. You seem to be hung up on grasping even the most elementary
fundamental concepts of arithmetic, thinking instead that decimal representation is the
only possible representation of numbers. It isn't, and as long as you persist in thinking
that text strings representing 64-bit-binary-floating-point have to correspond to your
delusional thinking about decimal arithmetic, you are doomed.

Computers do not implement abstract mathetmatical concepts. They also implement
finite-precision arithmetic, and they implement it using binary representation. All your
problems stem from one simple fact which you seem completely unable to grasp: FINITE
PRECISION ARITHMETIC INTRODUCES ERRORS. The rest follows naturally, and the issues about
representaiton (binary or decimal) are actually second-order effects, until you confuse
the fact that your complaints are in the decimal domain, which is irrelevant. You must
limit all future complaints to the binary domain, whcih is the domain of implementation.

I don't know if I can add anything useful to someone who clearly has no grasp of reality
and refuses to be educated in spite of the efforts of a dozen or more people to do so.
joe

On Mon, 8 Oct 2007 10:45:46 +0100, "GT" <ContactGT_remove_@xxxxxxxxxxx> wrote:

"Joseph M. Newcomer" <newcomer@xxxxxxxxxxxx> wrote in message
news:mpigg3ldi9556qdl6sih5nffu1gi31t3p8@xxxxxxxxxx
I have no idea whatsoever why you are using the word "wrong". Now you've
introduced the
word "unreliable". WHAT PART OF WHAT WE HAVE ALL BEEN TELLING YOU ARE YOU
MISSING?

I have no problem understanding binary to decimal conversion and why the
final digit is unreliable in on-going calculations, but what we are missing
here is an answer to my question - how do I get the computer to work to a
number of decimal places that are mathematically reliable! Please re-read
the thread and reply when you have something useful to add.

Joseph M. Newcomer [MVP]
email: newcomer@xxxxxxxxxxxx
Web: http://www.flounder.com
MVP Tips: http://www.flounder.com/mvp_tips.htm
.

Relevant Pages

  • Re: How to convert a number to hex number?
    ... Originally two's compliment representations were used to efficiently ... the same could be said for any ascii representation of numbers. ... The idea of base-complement is that the first digit is the zero digit for ... Also I wanted to see how much slower using strings instead of ints would be. ...
    (comp.lang.python)
  • Re: algorithms + data structures = programs
    ... Richard Heathfield wrote: ... Yes, okay, I agree. ... And dis the digit that will be displayed at ... Wirth ex machina again - he presumes the last digit of the representation ...
    (comp.programming)
  • Re: This calculation is just wrong / computer cant count!
    ... continuing refusal to accept reality is the problem. ... about accuracy and reliability, are simply asinine. ... The final digit is NOT random, ... representation is the 64-bit binary representation; anything else you see is irrelevant. ...
    (microsoft.public.vc.mfc)
  • Re: Numeric literals in other than base 10 - was Annoying octal notation
    ... Systems are not defined for radix 1. ... and every number has an infinitely long representation. ... take a 1 digit instead then it becomes workable. ... or the only number that can be represented is infinity. ...
    (comp.lang.python)
  • Re: Chipmunk Basic arrays (Attn: Ron Nicholson)
    ... They do have exact values. ... There is no "last" digit in the square root of 2. ... "I have finished writing down the exact decimal representation ...
    (comp.lang.basic.misc)