GameAnalytics

During the A/B test, it seems that the value to be obtained may be different each time it is started. Please check as it will have strange effects on the results. Unity 2020.3.36 iOS Only

Completely different A/B tests show identical graphs. Like I have two experiments with 4 variations in each of them. They have the same control group but variants are different. I evaluate the winning variant by playtime per user. And BOTH experiments while having completely different variants how to same data on graph! I reported this a while ago but it seems like it did not get any attention. GA login: andrey_melnikov_2002@icloud.com The game is: "AM" -> "AM" -> "CRAZY-SHEEP (Yandex)"

The number of filter items in each menu has been reduced. This has resulted in a number of data items that are no longer viewable. I am hoping that all the non-dimension filters used in previous versions will be available in the latest version. I am particularly troubled by the fact that there is no longer a filter for extracting AB test user group data. Currently when I run the AB test I cannot draw conclusions because I cannot extract the data with explorer, cohot, or funnel. We are hoping for an early resolution to this issue. Thank you in advance for your help.

The documentation mentions there being a way to prevent people that started their game in offline mode from being assigned a new A/B variant when they first connect. However, I cannot seem to find this option anywhere within the A/B interface. The question is: Has this option been removed during the upgrade to the new dashboard version, is this option only available in the SDK, or am I not looking hard enough? Link to doc: https://docs.gameanalytics.com/features/ab-testing/experiments-faq#how-are-offline-users-treated

I can create an A/B test with a config key of "123456789012"(12 characters), but not with a config key of "1234567890123"(13 characters). I can send a request to the server, but I get a 400 error. Since it should be possible to create a config key of up to 16 characters, I think there is something wrong with the validation on the server side.