This site uses different types of cookies, including analytics and functional cookies (its own and from other sites). To change your cookie settings or find out more, click here. If you continue browsing our website, you accept these cookies.
This is more of a warning, and not a question, but I just wasted a good part of the day trying to track down an issue, only to realize my issue was the same bug I reported back in May. This bug has been been accepted as a defect by Alteryx (but frustratingly assigned a low priority.)
The problem is that you may get inconsistent results with the Summary tool depending on if AMP is on vs. AMP being off.
This screenshot above is a good example. The top row represents my workflow outputs when AMP is OFF (header names changed for confidentiality reasons). The bottom row represents the same exact same workflow / data, with the only difference being AMP was ON. I had been cheerily working with the bottom row, when the top row better represents what is truly happening. Ouch.
When I submitted my ticket, I included a version of the attached workflow. Feel free to try it yourself. I just tested using the latest version of Designer, and the problem still exists.
All that being said, please use caution, and double check your outputs to ensure they line up with your expectations if you are using the AMP engine. I've been working with bad data, and reporting to my executive team that same bad data. I would hate for others to be in that same boat.
It looks like the rabbit hole goes deeper with version related issues as well. I'm on 2021.1.4.26400 and I see no differences between any of Amp-Int64, Amp-String, Amp-Double and non-Amp. All of the output files are included in the attached package.