[Camino-users] mesd runtime

[Ian Malone]

Hi, I'm following the lookup table step of the probabilistic fibre 
tracking tutorial:
http://web4.cs.ucl.ac.uk/research/medic/camino/pmwiki/pmwiki.php?n=Tutorials.MultiFiberTrack

How long should the following take?
mesd -schemefile 01719-100-1.scheme2 -filter PAS 1.4 -fastmesd 
-mepointset 15 < B1000_CalbData.Bfloat > B1000_CalbDataPAS15.Bdouble
It's been running for about three days now on a 3.2GHz Xeon. Camino is 
v2 revision 880, the process is definitely doing something:
PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
18   0 1146m  58m 2472 S  100  1.9   4416:20 java

The scheme file has 137 entries (9 b=0 and 2 sets of the same 64 
directions which will have some small differences as we're applying 
b-vector re-alignment).

cat B1000_CalibData.info
VERSION 1.0
ROTATION_SEED   0
ONE_DT_BLOCK_SIZE       3401
MIN_DT2_ROT_ANGLE       0.0
MAX_DT2_ROT_ANGLE       0.7853981633974483
DT2_ROT_ANGLE_STEP      0.08726646259971647
TWO_DT_BLOCK_SIZE       24583
NUM_TWO_DT_BLOCKS       10
DT1_E1_THETA    0.0
DT1_E1_PHI      0.0
DT2_E1_THETA    1.5707963267948966
DT2_E1_PHI      1.5707963267948966
DT2_ROT_AXIS_THETA      1.5707963267948966
DT2_ROT_AXIS_PHI        0.0

ls B1000_CalbData.Bfloat -lh
131M 2011-07-18 16:04 B1000_CalbData.Bfloat
(yes, there is a typo in the filename)

On the possibility I'm using a weird number of basis functions I've 
started a separate run with RE=16, that one has only being going half a 
day so far.

Thanks for your time.
Ian