[igraph] Inconsistency in measuring using iGraph GUI (tkigraph() ) and command line R - What did I miss?

[shahab]

Hi,

I faced a strange anomaly while measuring some certain metrics (e.g.
Average path Length, Power.-law degree exponent) using iGraph GUI and
alternatively  using same iGraph commands in R command-line  window. I
am sure that I have missed something, but I couldn't find what? and
how to make both consistent.
e.g.
In R command line:
 g <- erdos.renyi.game(4050, 157411, type="gnm", directed=TRUE)
average.path.length(g, TRUE)
[1] 2.670126

while using igraph GUI, I create same random graph based on
erdos&renyi model using same parameters,but I get a different value
for average path length : 2.205454

I am sure I am missing something, but what? any idea for this
inconsistency? How can I get same value using both?

best,
/Shahab