Well, as I wrote in the original reports about the packages creation problem, there is unfortunately no reliable way to reproduce it... Studio behaves unpredictably, the packages creation could crash at any of them during repeating of the same wizard (when it crashed, I simply clicked the Back button to get one wizard screen back and then immediately Next to start the packages creation again), sometimes taking 2 tries, sometimes 15 tries until it was able to create all packages in one run without fail.
It looked like it's happening only if "Create new project TM" option was used during the package creation... and it could be related to the fact that the main TMs (from which the project TM was being created) were rather large - from hundreds of MB to almost 2 GB... but again, there were quite differences between individual languages and the packages creation was failing randomly at various languages, so I don't think there was any direct relation.

The analysis bands problem can be reproduced using the PowerShell function I sent you previously. I also mentioned it in the email (which basically just repeats what was written in the forum posts).

I also can't guarantee that one year after the original reports everything works (or fails) exactly the same as before. Too many variables in the environment has changed since...

Well, as I wrote in the original reports about the packages creation problem, there is unfortunately no reliable way to reproduce it... Studio behaves unpredictably, the packages creation could crash at any of them during repeating of the same wizard (when it crashed, I simply clicked the Back button to get one wizard screen back and then immediately Next to start the packages creation again), sometimes taking 2 tries, sometimes 15 tries until it was able to create all packages in one run without fail.
It looked like it's happening only if "Create new project TM" option was used during the package creation... and it could be related to the fact that the main TMs (from which the project TM was being created) were rather large - from hundreds of MB to almost 2 GB... but again, there were quite differences between individual languages and the packages creation was failing randomly at various languages, so I don't think there was any direct relation.

The analysis bands problem can be reproduced using the PowerShell function I sent you previously. I also mentioned it in the email (which basically just repeats what was written in the forum posts).

I also can't guarantee that one year after the original reports everything works (or fails) exactly the same as before. Too many variables in the environment has changed since...