Interesting @juiceme, but this raises 2 questions for me:
- as I didn't run any "real" experiment before the update, if I do this now, I don't how I will be able to use this results, any idea?
- could you please describe the process to run such measures?

Thanks.