It's a set of rather large .dx files (hence maybe why the problem was happening). I can probably get those to you after this weekend


On Sat, Feb 16, 2013 at 6:48 PM, Baker, Nathan <> wrote:
Hello --

Yes, that sounds like a bug.  Would you be willing to share the OpenDX files that you were trying this with?


Nathan Baker
Pacific Northwest National Laboratory
Phone: +1-509-375-3997

From: Lane Votapka <<>>
Date: Friday, February 15, 2013 4:04 PM
To: APBS-USERS mailing list <<>>
Subject: [apbs-users] mergedx2 resolution

Hello APBS users/developers,

I believe that there is a bug in the mergedx2 program when the -r (resolution) argument is used.

When I type this command:

mergedx2 -r 0.5 -o gridmerged2.dx *_-PE0.dx

I recieve this error:

# main:   Reading Headers...
WARNING: The specified output resolution is greater than the
         resolution of the input DX files. Upsampling.......
# main:   Dimensions of the merged grid
# main:   nx = 2261, ny = -2147483648<tel:2147483648>, nz = -2147483648<tel:2147483648>
# main:   hx = 0.499895, hy = 0.000001, hz = 0.000001
# main:   xmin = -575.530000, ymin = -560.902000, zmin = -589.033000
# main:   xmax = 554.232005, ymax = 548.614121, zmax = 594511896
VASSERT: ASSERTION FAILURE!  filename vmem.c, line 241, ((num > 0) && (size > 0))

The program will even fail if I use -r 1.0 (the default). I'll bet it's somethin to do with the ny, and nz parameters output near the top.

mergedx2 without the -r argument works just fine, but that does me no good since I do want to change the resolution of my grid.