The following summarizes improvements and bug fixes made in version 5.03 of Derive 5 released on 15 January 2001:

Algebra window items:

•        Derive 5 is now fully compatible with Windows 2000.  The problems displaying and printing Derive text objects when operating under Windows 2000 have been resolved.

•        The minimum percentage of memory allocated by Derive and allowed by the Options > Startup command is now computed and is based on the current physical memory of the host machine.  This makes it possible to allocate less than 20% of physical memory to Derive on computers in a multi-tasking environment with a large amount of memory.

•        Matrices are now printed correctly, even if both the length and width of the matrix spans more than one printed page.

•        The plot color of point list plots and parametric plots in 2D-plot objects is now saved in DfW files.

•        After the user pressed the Esc key to abort a long computation, on very rare occasions Derive would be abnormally terminated by a system zero-divide error.  This problem has been resolved.

•        A difficulty loading utility files with very long file names from a drive other than the current one has been resolved.

•        When formatting text in a text box, version 5.03 no longer scrolls the worksheet to make the top of the text box visible after formatting the desired selection.

Plot window items:

•        The intersection of 3D plots is now displayed with greater accuracy without significant speed degradation.

•        Choosing the top left white color in the color palette as the background color in the 3D plot-window no longer results in a background color of yellow on some machines. 

•        A problem that prevented making 3D plots of some equations involving two variables has been resolved.  For example, z=x and z-x=0 both plot as a plane in the 3D-plot window.

Mathematical items:

•        The ability to solve equations involving the elementary math functions has been enhanced by the addition of several new rules.

•        Version 5.03 is able to efficiently and accurately approximate huge magnitude numbers, and then display them in scientific notation.  For example, try 7^1000000 approximates to 1.096514191·10^845098, and 1000000! approximates to 8.263934043·10^5565708.

•        The accuracy of the real and imaginary parts when approximating the error and complementary error functions has been improved.  For example, at 10 digits of precision, version 5.03 correctly approximates ERFC(10*î) to 1 - 1.524307422·10^42·î.

•        Several new definite and indefinite integration rules have been added.  For example, version 5.03 can integrate LN(SIN(x)) from x=0 to π/2.

•        The ` transpose operator now distributes over matrices whose elements happen to be equations and inequalities.  For example, version 5.03 simplifies [x=1, x=2; x=3, x=4]` to [x=1, x=3; x=2, x=4].

•        User functions defined using the APPROX, EXPAND, and FACTOR functions now work as expected.  For example, if FOO(x) is defined as APPROX(x), version 5.03 simplifies FOO(π) to the rational approximation 1146408/364913 instead of the symbol π.

•        A problem that prevented solving equations involving absolute values for real values of the solution variables has been resolved.  For example, version 5.03 can find both real solutions to the equation |x| = 5.


Other Revision SummariesRevision_Summary 

Created with the Personal Edition of HelpNDoc: Easily create HTML Help documents