On this page, we share the results of tests “in the field” with real users…

• Time: 15h00 - 17h00
• Weather: Gorgeous
• User-Group: 8 participants incl. 2 computer scientists, two boys (age of 8-12), two primary school teachers, 1 catholic priest
• Devices: 2,3,4
• Test Location: Inner city of Bonn

• The Internet connection was fairly stable on all devices
• The server application didn't make any trouble
• The client software was stable
• Observer application on devices ran without problems
• Creativity in flexible use of the game
• Devices run longer than expected (app. 2 hours)

• GPS was the big troublemaker (not the connection to the devices, but the delay until the first fix)
  • GPS 4 found it's first fix after about 15 minutes in the Beethovenstraße
  • GPS 3 found it's first fix after about 40 minutes in the Poppelsdorfer Allee area
  • GPS 2 found it's first fix after about 60 minutes at the Hofgarten in front of the university
  • Device 1 didn't found a fix at all, so we decided not to use device 1

(Maybe all the systems updated their Almanach data, which took so lang - but they all were in use the day before and fully charged, which means the should have had already the required orbit data)

• It was too sunny to read anything on device 1's display
• You have to mention that Mr. X's role has to be selected in the “join” dialog in order to have his functionality available.
• Coloring of participants on the map was a bit unreproducible
• Its hard to look all the time on the device's screen

• The game starts 10 minutes after gathering, such that the players have enough time to swarm out in the playing area.
• There should be at least 2 detectives, better would be 3.
• The show up interval for Mr. X should be dependent on the area (e.g. around b-it 10 minutes, in the inner city 5 minutes)
• The independent variable for the test is:
  • Show-up interval

• For the first tests we agree on moving at walking speed to eliminate the effects of personal fitness. The strength of Mr. X vs. detectives should be determined by evaluating the plain game result only.

• After each test run we decide on adjusting our variable depending on the outcome of the game. If the result is strongly suggesting that one side is in advantage, the variable is changed, otherwise the next game will be played with the same setup.