MathGroup Archive 1992

[Date Index] [Thread Index] [Author Index]

Search the Archive

Problems plotting arrays


I've noticed what appear to be two bugs in Mma 2.1, the 
first involving contouring and the second in the new 
package ListShadowPlot, when plotting arrays.  I'm using a NeXT.

**BUG(?) #1**
In a Notebook I wrote for Mma 2.0, I read in external data
from a file MyDat. The data was written into this file as 
numbers ((a(i,j),j=1,numx),i=1,numy), and read into Mathematica via

NumX=Read[MyDat, Number]  <= Number of elements in row
NumY=Read[MyDat, Number]  <= Number of elements in column
a=ReadList[MyDat, Table[Real] ]; <= Read in the array as a
                                    flat sequence of numbers
b=Partition[a,NumX]; <= Break it logically into rows

Under Mma 2.1, , e.g., for  36 rows and 26 columns in the
array b, surface plots like

ListPlot3D[b, ...directives]

continue to work fine.  However, an attempt to use the 
*new* package ListShadowPlot3D [in Graphics`Graphics3d`] 
to plot the same data, via

ListShadowPlot3D[b]

yields (for the same 36 row by 26 column array b above)

Part::partw: Part 27 of {<<26>>} does not exist.
Part::partw: Part 27. of {<<26>>} does not exist.
Part::partw: Part 27 of {<<26>>} does not exist.
General::stop: 
   Further output of Part::partw
     will be suppressed during this calculation.
     
Then generally the swap file begins to grow....The
object b looks precisely as it should.

**BUG(?) #2**
The thickness directives in

ListContourPlot[b,
ContourSmoothing->Automatic, 
ContourStyle->{{Thickness[.002]}, {Thickness[.002]},
 {Thickness[.002]}, {Thickness[.002]}, {Thickness[.006]}},
Ticks->Automatic]

which under Mma 2.0  made every fifth contour 3 times as thick, 
no longer work, giving the error message

ContourStyle::style: 
   -- Message text not found -- (
    {{Thickness[0.002]}, <<3>>, {Thickness[0.006]}})
  
and all contours come out the same thickness.  [But boy! -Is 
the contouring faster than it was under Mma 2.0!]  Apart from this, 
the contour plots are correct.

Can anyone confirm these problems and/or suggest workarounds?





  • Prev by Date: changing the font of "input"
  • Next by Date: Re: Elliptic integral problems in mma
  • Previous by thread: changing the font of "input"
  • Next by thread: Mma on RISC