In May, we tested ToePunch in Finland at the TRaVilO event and encountered a problem with box overheating. The problem was caused by a hardware error where from time to time the NFC reader module (chip) forgets its configuration after waking up and stops reading. We had to redesign the firmware so that it resets the reader chip configuration after each wake-up. With this new version we went to FinTrailO.
The test of boxes in the real event was successful, all the boxes worked well and all the chips that successfully passed the Clear/Check procedure punched well during the whole race.
We've tested also the readout function of the ANT application and it worked also quite well with two exceptions:
The resume of this test in-vivo was: hardware works well, we should tune better the result processing system.
The system configuration sounded wery complicated before the event. I tried to read all instructions but I felt like there is something that I forget to do or something that I do not realize that should be done. During the event, I noticed I was wrong. The configuration of the units, etc. was really easy. Of course, I did not see all the work that was done, like the event pre-configuration with ANT and the creation of the event chips, but I guess those steps are quite simple too and maybe could be done by the event organizer in future.
Our response: Admin chips are made using a tiny Android application (ToePunch Config Tool). You simply select Event chip making, fill in an event ID in preoresultat and start date/time. Then you put a chip to the smartphone. That's all.
The data-read in had to be done with a mobile phone and ANT, since there was no WiFi available for all units and few marshals. The process was very easy, but starting to use the application wasn't. However, the ANT version was very new and it is being developed all the time. Automatic read-in of punches from units to server would be ideal. In our event, the successful delivery of punches to preoresultat software was really useful. That software was familiar to our marshals and we had no problems in the second event when the data-read in did not succeed due to the configuration that was made too early. But in the two events when it succeeded, it really made the results calculation and publication faster and was very helpful.
The equipment was a problem in our event, as SportIdent holders weren't available. I didn't want to connect them into Emit holders, since it might need a power drill to bind them properly, however it might be good when you have several events. We managed the problem by taping the units to sticks, but that wasn't very successful. When I picked up the Toe units from the course in my car to bring those to the next course, many of the tapes were detached. It does not make much sense to use such the system for several events if it's needed to tape the units again between the events. But the problem might only have been the quality of the tape, with stronger tape it might work.