Points on the graph showing concurrent users for the past hour in the Analytics dashboard do not always line up correctly

Bug Reports Website Bugs
#1

Explanation

When viewing the concurrent users for the past hour in the analytics dashboard, the marked points on the graph (where the coloured circles are) do not match the height of the marking lines, and are not consistant in their heights.

For example, the 2 points shown in the image below both have a value of exactly 3 (not fractional, as it is CCU), however the first point is lower in height than the second point, and isn’t on the dotted line (which shows a height of 3).

image

Complete graphs of 2 seperate games (unable to test for games with higher CCU, as my games are currently performing poorly) are below:

image
image1039×615 69.2 KB

image
image1090×635 84.4 KB

Reproduction

  1. Access the ‘Performance’ catagory of an experience’s analytics.
  2. Set the timespan to the previous hour
  3. Observe the Concurrent users graph, where points do not match up with the marking lines

System Information

Operating System: Windows 10
CPU: Ryzen 5 3600
GPU: GTX 1650 Super
RAM: 16GB @ 3600MHz
Screen Resolutions Tested: 2560 x 1440; 1920 x 1080
Browsers Tested: Google Chrome; Microsoft Edge

Operating System: iOS 16.6
Device: iPhone 13
Browsers Tested: Safari

1 Like
#2

We’ve filed a ticket into our internal database for this issue and will start investigating, we will update you when we have further information.

Thanks for flagging!

#3

both have a value of exactly 3 (not fractional, as it is CCU)

The value could be fractional as the CCU data is collected multiple times per minute however the granularity at which we show the CCU value is on a per minute level. If you have 3 people in your experience and 1 of them leave 46s into the minute and we collect stats every 15 seconds then you have 4 data points looking like [3, 3, 3, 2] = 2.75 for the entire minute.

I believe we will do a rounding on the display value so that it’s always a whole number, but the underlying data would be fractional. When you CCU number goes up, the Y-axis will self-adjust and you won’t notice the fractional vs rounded data point.

LMK if this makes sense to you or if you have any other questions.

1 Like