Maxent.jar vs dismo

[Arta]

Dear all

I used to use maxent.jar for many years, now i want to calculate AICc of outputs. For this goal, I switched to dismo package but I've had some problem:

1. the result of dismo is not similar to maxent.jar (variable contribution, I mean). 

2. I used this command but the results, curves,... seem to  related to 'one' replicate only, and it is not referred to replication type at all in html output (opposite of maxent.jar that displays this setting in the html output). maybe my commands has any fault?

> kaiseri.me <- maxent(MacroEnv, kaiseritrain, replicates=30, replicatetype=Bootstrap, writebackgroundpredictions=true)

Many thanks in advance

Arta

[Laura Patterson]

Is the result in dismo very different than the GUI maxent.jar?
If not, might just be due to fluctuations that will naturally occur when building models, ie they change slightly each time.
Some times the order of variables will affect the model building process also.
Do the jackknife graphs look the same?