Different Results for different versions

[Nathan Dalbec]

So am I the only one who has this problem? Honestly I can generate a different DAG with the exact same data with any version I use. Is the issue that the PC and GES algorithms are being improved with each update or is this program that unstable that it will produce a new graph with every version?  How are can we ask the scientific community to except our results if they cant be easily reproduced using the same software? I would pay for a premium version just to have consistent results. If there is a version out there that actually can give accurate PC and GES outputs I would love to find it.