axiom-developer
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Axiom-developer] [#167 Infinite Floats Domain]


From: wyscc
Subject: [Axiom-developer] [#167 Infinite Floats Domain]
Date: Mon, 13 Jun 2005 09:13:07 -0500

Changes 
http://page.axiom-developer.org/zope/mathaction/167InfiniteFloatsDomain/diff
--

??changed:
-Most people would expect the answer of 'z-x' to be '0.16887242 E-20' but this 
ignores the fact that the display is converted from an internal binary 
representation to a decimal one. During the conversion there is truncation 
error (think of 0.1 in base 3 converted to decimal 0.3333... with output at any 
finite precision). So 'x' is not what it seems, and neither is 'z'. Below, the 
constants are converted to binary internally before computation, at a higher 
precision than x (resp. z) to bring out the differences. We can now see that 
'x' is larger than 'z'. So Axiom is correct and the expectated answer is wrong.
Most people would expect the answer of 'z-x' to be '0.16887242 E-20' but this 
ignores the fact that the display is converted from an internal binary 
representation to a decimal one. During the conversion there is truncation 
error (think of 0.1 in base 3 converted to decimal 0.3333... with output at any 
finite precision). So 'x' is not what it seems, and neither is 'z'. Below, the 
constants are converted to binary internally before computation, at a higher 
precision than x (resp. z) to bring out the differences. We can now see that 
'x' is larger than 'z'. So Axiom is correct and the expected answer is wrong.

--
forwarded from http://page.axiom-developer.org/zope/mathaction/address@hidden




reply via email to

[Prev in Thread] Current Thread [Next in Thread]