The Problem

500m Interval timing - everybody wants it, but few regattas do it.  Why is that?  After all, we have tablets, hotspots, and RM Timer software for instant timing, right?

The answer is: it takes too much labor.  A regatta wanting to do interval timing for a multi-day regatta just about needs a separate volunteer coordinator just for timing.  Add to that the logistics of getting timing teams to and from the intermediate timing stations, providing amenities and shelter for them (if there is even a fixed place for them to operate), and additional platforms or launches if they have to do timing on the water. And let’s not forget training.  Or the cost of those extra tablets and hotspots and stuff to keep them charged up.

That is a lot of additional overhead for an optional feature.  Too much for most regattas, even the bigger ones.  There are relatively few venues with 500m timing shacks on terra firma, AC power, and reliable wi-fi, and even in those venues interval timing is often not done.

We have a GPS solution, but it is expensive, and still requires extra logistics to mount the transceivers, get them back, charge them up, and send them out again. 


A Solution!

In 2016 we did 500m interval timing for the Rowing Canada Aviron National Rowing Championships, using tablets running RM Timer, with USB internet sticks.  This regatta happened in September, and was hosted at Burnaby Lake, BC for the first of 4 years.  The weather was generally cold and wet.  The Local Organizing Committee had to deal with setting up tents on the 500m platforms, providing battery backups for the computers, and getting the timing teams to and from their positions.  The volunteer coordinator had to really scramble to arrange all the timing volunteers (3 teams, 2 people per team, 3 shifts a day, for 3 days).  The volunteers did a great job, in less than ideal weather conditions; but one wonders how many of them would have been eager to sign up for the job the following year.

And in 2017, the NRC’s would be in November.   With the potential for the weather being even colder and wetter, Simon Litherland, who was the LOC member responsible for timing and results, decided to explore other ways of accomplishing the 500m split timing.

The basic idea was to mount cameras on the 500m platforms, and record times by monitoring them from a central spot at the finish line.  This was not a new concept - many people have talked about doing it this way.  Talking about it in broad conceptual terms, though, and actually executing it are very different things.

The main challenges were:

  • Finding cameras that had the resolution and frame rate to display the boats clearly.
  • Creating a wi-fi network with sufficient bandwidth and speed to carry that data from 4-5 cameras over 2000m with acceptable latency.

Other considerations included protection from the elements (ie, keeping rain/snow etc. off the lenses), using weather resistant equipment, having enough power to run without maintenance for an entire regatta day, portability, ease of deployment, and overall cost.

Simon has written a fairly thorough ‘cookbook’ for the system he put together for the NRC, so I won’t go into the technical details again.  In general, the system consisted of a Swann DVR with wired IP cameras, connecting to Ubiquiti Wifi transmitters supported by batteries.  At the finish line, the DVR output went to a large TV, where a small team of observers and computer operators noted the crossing order at each 500m station and recorded times in RM Timer (our Windows tablet timing software).  Here is the basic design:


How it Worked

In general, it worked really well.  I must admit I was a bit skeptical that they would be able to get the crossing order right at each station, as there was no replay or frame capture to go back to in case of a question.  However, the spotters were experienced referees, and the video quality was good enough that they had no difficulty with this. 

The computer operator had 3 instances of RM Timer running on a laptop with a 2nd (extended desktop) monitor to easily see all three instances at once.  As the boats passed each station, the computer operator recorded a ‘stopwatch’ time in the software for each boat, then applied the bow/lane number to the time from the spotter’s crossing order.  The times then went to the RM Live Timing service for live results.  It went like clockwork.


Lessons Learned

The system worked well, as I said.  But this was the first test, therefore a learning experience for all.  There were a number of things we realized, or thought of during the regatta, that could improve it.

For starters, the cameras were not PTZ; therefore, it took 2 people to align them – Simon on the platforms pointing the cameras and Ian Gordon in the finish tent looking at the monitor and telling Simon, via cell phone, when they were lined up right.  Although this was not a big deal, there is a lot to be said for having the ability to move the cameras and zoom in/out on demand, remotely.

 Next, the timers, though not sitting in igloos on the 500m platforms, were still outdoors.  And it was cold.  And it rained the last 2 days of the regatta.  The tents were really canopies with some sides attached (but open on the course side), so they offered no warmth and didn’t even keep out all the rain.  Hand warmers, many layers of clothing, and electric heat pads were the order of the day every day.  Not the conditions that would have volunteers standing in line to come back next year.

The realization here was that the timers could do their job indoors, or in a weatherproof (ie, warm and dry) tent.  They didn’t need to see the course – they were getting their information from a TV screen. 

Another thing I mentioned earlier is that there was no replay.  That is not to say the system didn’t have the capability, it just required more gear.  Another computer on a private subnet with the DVR (and someone to operate it), and the Swann software, should do the trick.  The software has the ability to export video or even snapshots from replay, which would allow a quick review in case there is a question about crossing order. Even having someone take a short video of the monitor with a cell phone camera would allow greater comfort that it was going to work.


Conclusion

First of all, many thanks to Simon Litherland for his hard work and ingenuity in designing, building, and running this system at the NRC’s.  He was the first to arrive and the last to leave every day, out on the course before first light, setting up the transceivers and power systems, and again after racing taking them back in for the night.  And in between, setting up the system in the finish tent and working with the timing teams all day to make the split times happen.

As for the maiden voyage of the system at Burnaby Lake, it has to be considered a success.  It worked as designed and intended, throughout the regatta, without significant issues.  Applying the experience of the first run will only make it better.  And by sharing this information we hope to encourage other regattas to augment their timing efforts to include 500m interval times.   This provides valuable information to rowers and coaches, and promotes visibility of regattas by giving spectators a better view of races in progress.