Has anyone out there ever GPS'd a 24 or 12 hour race? Not something you'd think of doing, I know. But after finishing this past weekend's 12 hour race I realized that I could do something interesting with the data.

Pacing is, of course, key in these events. But normally all you get is your 'lap times' from the race officials which include your transition times as well as riding time. If you are like me, you don't want to worry about pushing buttons during a race, so I don't even time my own laps with a watch, let alone try to time my transitions as well.

However, I stuck a Garmin Foretrex unit on my handlebars (weighs about as much as a HR monitor) and didn't touch it during the race. Afterwards I was able to produce the following chart of my *riding* lap times and transition times:

<img src=http://www.topofusion.com/photofusion/insomnia/laps.gif>

What this tells me is that I rode fairly steadily. Looking at my official lap times (the sum of transition+riding lap time), however, doesn't look quite as steady. I wouldn't have much of an idea if I was just riding slowly or if I just took a longer transitions. Some of my laps were timed at around a half hour, but I never rode that slowly.

I can also figure out how much time I spent stopped (flats, breaks to use the facilities, etc) during each lap.

If anyone is interested in this (I know it's only for the truly obsessed), post here or PM me. I'm one of the authors of a GPS mapping app called TopoFusion. Right now this lap feature is only in my development version, but we will release it in the beta/full versions soon. It's all automatic--you just give it your tracklog of the race and it spits out the lap/transition times.

BTW - the race was the "12 Hours of Insomnia" in Hermosillo, Mexico, a night-only 12 hour race. I managed to win the solo category. More on the race at:

http://www.topofusion.com/news.php

I use the Timex.

Works similar, except you have to carry a little GPS device in the back pocket.
The only problem I have had is that the GPS signal is sometimes misrepresented in areas where you make constant turns. The device does not take constant calculations of your position (or doesnt seem to) so it ends up short of data. This is mostly in extreme wooded and mountainous areas.
In areas that you have less obstructions to the sky it works very accurate.

I use the Timex.

Works similar, except you have to carry a little GPS device in the back pocket.
The only problem I have had is that the GPS signal is sometimes misrepresented in areas where you make constant turns. The device does not take constant calculations of your position (or doesnt seem to) so it ends up short of data. This is mostly in extreme wooded and mountainous areas.
In areas that you have less obstructions to the sky it works very accurate.

I haven't looked much at the timex systems. What sort of post-race analysis does their software do? Anything like what I've done above?

In order to find lap times the only requirement is that you're getting a good signal near the start/finish area. Of course if you want distance, total climbing, et cetera, signal can be a big problem. The race was in Mexico at about 1000 feet, so I imagine I had a lock throughout the race. Even still, the GPS had about 3 miles less than my cyclometer (2% error) since there were quite a few switchbacks on the course.

Another question: how long does the battery last on the timex? Have you been able to get a tracklog for an entire 12 or 24 race? I'd be curious to see any data you have...

Timex has some software that works pretty good. Really simple such as what you have below. only graphs miles, heartrate, and I think altitude (havnt played with that on the timex, using polar for that).

I am playing with a few other software tweaks that have similar graphical analysis to what you have but with some other gps features.. This has been kinda unreliable but getting better.
I noticed on the Timex site they have some new models and might look at them as they are allowing software developers to write code for the features.

I have had the pack about a year and I think three 24 hour races and maybe 4 12 hour races... No battery changes needed on the watch, but on the GPS device it works well over a 24 hour race possibly 3 or more?.

Timex has some software that works pretty good. Really simple such as what you have below. only graphs miles, heartrate, and I think altitude (havnt played with that on the timex, using polar for that).

I am playing with a few other software tweaks that have similar graphical analysis to what you have but with some other gps features.. This has been kinda unreliable but getting better.
I noticed on the Timex site they have some new models and might look at them as they are allowing software developers to write code for the features.

I have had the pack about a year and I think three 24 hour races and maybe 4 12 hour races... No battery changes needed on the watch, but on the GPS device it works well over a 24 hour race possibly 3 or more?.

Thanks for the info. I had another look at the timex system. They claim 12 hours of battery life for the GPS recorder. But with rechargeables I could see you having to change batteries 3 times over a 24. The foretrex worked well for the 12 hour race (no charging or battery changes). Garmin claims 15 hours of battery life.

I see that the timex only sends distance/speed at each time interval. So I don't think there is anyway to do the sort of lap analysis I'm looking at... at least not automatically. I am doing it by looking at the lat/lon coordinates. When you near the start/finish area again I know that the lap is complete and can look at the timestamps. You might be able to do this with only time/distance, but it would be very unstable to GPS errors.

Anyway, it's interesting to see I'm not the only one out there racing and GPS'ing.

Anyway, it's interesting to see I'm not the only one out there racing and GPS'ing.
Of course, you only get the up / down, and not the left / right, but it's really easy to pick out the laps just from the curve.

The Polar's not cheaper than a GPS, though... :mad:

Of course, you only get the up / down, and not the left / right, but it's really easy to pick out the laps just from the curve.

The Polar's not cheaper than a GPS, though... :mad:

Right, it should be relatively easy to pick out the laps from an elevation profile, but are you able to accurately determine how long you spent in transition as well?

I guess what I like about my analysis is that it tells you how long you were in the pits and is also completely automatic.

One of these days we will get around to adding Polar heart rate data to TopoFusion. I believe the file format has been published.

are you able to accurately determine how long you spent in transition as well?
I assume by transition you mean hanging around in the pits, at checkin / check out, well, yes, it's that flaaaaaattttt line that lasted up to a freakin hour at my last 12 hour race.

I guess what I like about my analysis is that it tells you how long you were in the pits and is also completely automatic.
Automatic, the Polar is not, but again, you can pick out the flat line really easily by eye. The Polar data analysis program gives you "quicktips" at the cursor, so you can just put the cursor over it and do the math to see the time in transition. But, as you note, this is not automatic by any stretch of the imagination...

One of these days we will get around to adding Polar heart rate data to TopoFusion. I believe the file format has been published.
I assume you mean the Polar heart rate file format? It's a trivial format, if you want me to send you one I'll send you mine from Willamette Pass. PM me with an email address and give me a couple of days, I'm kind of slammed of late. I believe that Polar is happy to support companies who want to work with their products, but I haven't gone down that path so I can't say for sure.

One of these days, when I quit one or the other of both my day jobs, I might just get around to writing a better analysis tool than Polar's. It's pretty straightforward, especially if you have a hint of the lap time, to pick out the laps, detect the transitions, etc., and it would be a cool feature.

What I _really_ want is a wireless GPS/HRM combo still with a wheel pickup to backup the GPS in the woods, and with a http://www.fitsense.com/tour5.asp#tourtop quality radio on both the wheel pickup and the HRM.

My main beef with my Polar is how trivially easily it will loose both the HR signal and the wheel signal, and it won't re-acquire unless you stop it and re-start it, which is a PITA and destroys the continuity of the data, unless you hand-edit back together, and frankly, I mostly just like to look at the pretty pictures of the altitude, distance, HR, etc.

While I don't really want to slam Polar, their low-pass signal scheme was understandable ten or fifteen years ago, but it's time for something reliable, ya know?


Rolland

I assume by transition you mean hanging around in the pits, at checkin / check out, well, yes, it's that flaaaaaattttt line that lasted up to a freakin hour at my last 12 hour race.

:)


Automatic, the Polar is not, but again, you can pick out the flat line really easily by eye. The Polar data analysis program gives you "quicktips" at the cursor, so you can just put the cursor over it and do the math to see the time in transition. But, as you note, this is not automatic by any stretch of the imagination...


Right. Well, that's useful nonetheless. One of the odd things about the Insomnia race was that the lap was so short, so picking out 30 transition times wasn't very appealing to me. We have quite a few users who are runners (using Forerunners), so I'm tweaking the code a bit to work for even shorter laps right now.

I assume you have one of the polars that records data every 2 seconds or so. I suppose one problem would be if the start or end of a lap is very flat--you wouldn't be able to tell exactly where you started moving again. I'm not sure how the stable to fluctuations the barometer in the Polar is.


I assume you mean the Polar heart rate file format? It's a trivial format, if you want me to send you one I'll send you mine from Willamette Pass. PM me with an email address and give me a couple of days, I'm kind of slammed of late. I believe that Polar is happy to support companies who want to work with their products, but I haven't gone down that path so I can't say for sure.


PM sent, thanks. All I'd need is to be able to read the formats, then correlate that with the time stamps on the GPS data. It should be pretty cool, but as you said, finding the time to write the code is another thing...


What I _really_ want is a wireless GPS/HRM combo still with a wheel pickup to backup the GPS in the woods, and with a http://www.fitsense.com/tour5.asp#tourtop quality radio on both the wheel pickup and the HRM.

My main beef with my Polar is how trivially easily it will loose both the HR signal and the wheel signal, and it won't re-acquire unless you stop it and re-start it, which is a PITA and destroys the continuity of the data, unless you hand-edit back together, and frankly, I mostly just like to look at the pretty pictures of the altitude, distance, HR, etc.

While I don't really want to slam Polar, their low-pass signal scheme was understandable ten or fifteen years ago, but it's time for something reliable, ya know?

Rolland

Maybe 10-15 years more and we'll have what you describe. There are similar problems in the GPS world (where things are far behind the times). One example is in tracklog capacity--units with over 50MB of memory are still limited to a 10,000 point tracklog. This sorry detail forced me to haul my laptop along the continental divide route in order to GPS it all. (There were other solutions, of course, but the point is that the GPS unit itself should have been able to store all the data).

Scott

Maybe 10-15 years more and we'll have what you describe. There are similar problems in the GPS world (where things are far behind the times). One example is in tracklog capacity--units with over 50MB of memory are still limited to a 10,000 point tracklog. This sorry detail forced me to haul my laptop along the continental divide route in order to GPS it all. (There were other solutions, of course, but the point is that the GPS unit itself should have been able to store all the data).

Scott

One of the nice features of the Magellan Meridian series is that you can save your tracklog to the SD memory card. On multi-day trips I've just saved it as "Day1" Day2" etc... I can do the same with waypoints, I can load multiple sets of waypoints using files saved on the memory card. I use this for navigation aids (plot out the route on the PC before I go, download the waypoints then save them to a file). I am a bit jealous of the more compact units out there now though, the meridian is pretty bulky.

One of the nice features of the Magellan Meridian series is that you can save your tracklog to the SD memory card. On multi-day trips I've just saved it as "Day1" Day2" etc... I can do the same with waypoints, I can load multiple sets of waypoints using files saved on the memory card. I use this for navigation aids (plot out the route on the PC before I go, download the waypoints then save them to a file). I am a bit jealous of the more compact units out there now though, the meridian is pretty bulky.

Yup. I probably should have used a Meridian for the task. I have a sportrak map, though, and am not very satisifed with the tracklog it records (Garmin's 'auto' algorithm produces better tracks IMO). Plus I wasn't sure I could get the data into GPX format once it was saved to the SD card.

The SD card is a nice solution, but it'd be even better to not have to save out (or load in) data. Tracklog memory is not so big that it couldn't fit in the space for maps.

On your multi-day trips were you touring (if so, where), or day riding?