How fast do you really go while carving?

[BlueB]

Just downloaded... Wants email, paswords, account created...

Not for me, thanks.

[dlokshin]

Instead of looking for an iPhone 4 visit that lasted all day, take a look at this one:

http://www.alpinereplay.com/stats/u?uId=152&vId=1353

Pretty typical: 10am to 2pm (you can check by clicking on the first bar and then the last bar. This was on a Samsung Infuse which I know has worse battery life than an iPhone 4 (because I had one, and my gf has the iPhone 4).

As a note: 1. We ask you to switch off WiFi when you start logging. Really no reason to have it going while on the mountain 2. Android uses accelerometers which iPhone does not, so the bat should last even longer.

BlueB,

You need to create an account because AlpineReplay is more about just tracking your own day on the mountain. You can compare against others using the leaderboard, race in 3D, and enter into challenges. Not really possible unless we ask you to create an account.

[BlueB]

You need to create an account because AlpineReplay is more about just tracking your own day on the mountain. You can compare against others using the leaderboard, race in 3D, and enter into challenges. Not really possible unless we ask you to create an account.

I don't need all of that. Less unnecessary networking, the better.

[softbootsurfer]

are the G's... :D and if it does not Snow...

[Jack M]

Thanks for dropping by David! Wish I had an iPhone! :(

Got an example of a map that shows somebody carving round sinusoidal turns?

BlueB, I think we can assume that the "social networking" side of this app is what allows it to be free. I agree it would be nice if there was a pay version without that.

[Corey]

David, do you need a data plan/3G on the hill to use this?

I use an iPhone but often have zero reception at resorts near me, and when I travel in the US I won't pay the ludicrous international data rates.

[eajracing]

I go plaid

after i got my mr. coffee i think i saw that on the radar.

[dlokshin]

Hey Jack,

How about this: http://cl.ly/452s052h1D3y1L3x1z3s. The dude was a former US hopeful, so the carves are a bit tighter than your average skier.

Corey,

You don't need to have a data connection to log your data. Once you get back to a data connection (either 3G or WiFi), send the data over and our servers will process it.

Keep the questions coming!

[Jack M]

Hmm, not quite, but it looks like it might work for us. Basically I want to know if it can track a racer through a PGS course:

<iframe width="420" height="315" src="http://www.youtube.com/embed/cWGOIw0jaYA" frameborder="0" allowfullscreen></iframe>

[Dan]

Hey Jack,

How about this: http://cl.ly/452s052h1D3y1L3x1z3s. The dude was a former US hopeful, so the carves are a bit tighter than your average skier.

Hey David, welcome to Bomber -- it's very cool of you to hop in here. I feel that sometimes the way we use the word "carve" here doesn't match the way that the rest of the world uses it. For example, this video is a good example of "BOL-carving", <a href="

">

</a>.

As you can see, the turn radius is often fairly small. Google says that at 30 mph you travel about 45 feet in one second, so at the minimum sample rate of 1 hz that you mentioned, it seems to me that you could actually be missing quite a bit of the distance traveled. That is, if we take the 1 hz sampling rate for argument's sake, the interpolated line drawn through sampling points would miss the "apex" of many/most turns, and yield slower times than actual.

It seems to me that a sampling rate around 10 Hz would address this pretty well. Does my description of the problem make sense, and is it possible to get sampling rates that high?

[dlokshin]

Hey Dan,

Thanks for the kind words. Glad I found you guys, it's an awesome community.

You're right, the curves won't be crisp enough for the accuracy that you guys are probably looking for. There should still be some curvature to the curve we record. There are some smoothing algos that we use depending on the pattern of points, etc. I'd love for someone to actually take this on a race and ping me so that I can take a closer look at the data.

As a matter of point (or theory) we'd love to eventually build hardware that synchs seamlessly with AR with a 5Hz GPS chip (10Hz is prohibitively expensive).

Let me know what you guys think once you hit the slopes!

[*Ace*]

[quote name=

John Gilmour]... Ruthies- my favorite steep carving run..70+ mph on that carving laid over in soft boots. My gloves would melt away every 3 days ...

Found this quote, had to share...

[philw]

In terms of speed, it's pretty accurate. For most entries, probably to a half mile per hour. A majority of modern chip sets use Doppler to figure out speed, and not a back calculation from two points. For those that don't use Doppler, they heavily rely on something called a Kalman filter which works very well on straight continuous motion, but not as well on dynamical motion like skiing.

... my co-founder was the CTO of Magellan Navigation for more than a decade, so we know a lot about sensors and GPS. Can't wait to see you all on AlpineReplay!

Good point about using doppler - you don't have to integrate the position to find the speed. But I'm still trying to understand the wildly over-optimistic speeds I've seen devices report (not the ones reported here: I did not witness those, have not used the Android product discussed here. I'm just curious.).

Here and here are some data on the Doppler approach. Perhaps there is something more recent somewhere? The basic physics should not have changed. There are a few obvious parameters in there which I think you'll need to be careful with. Their tests are for open water sailing, so there are no canyon effects, and that they specifically mention the "window" over which you're averaging, and issues with oscillations of the receiver. Other sources suggest sat visibility and receiver motion are particularly problematic.

Android's android.location.Location.getSpeed() is presumably what application developers are using. As the sampling rate and other parameters are clearly important in the error, even if confidence levels were given (I have not found any), then those must be dependent upon the application (see above links). What window is the speed averaged over? What accuracy is actually being claimed? (I have read the thread, but accuracy needs to be expressed as a percentage of the value, or a confidence level, it's not an absolute value)

[mattj]

I have been playing around with recording my tracks using my phone (Droid Bionic) and then putting them on Google Earth.

Seeing my track in Google Earth was kinda neat but I really wanted to see what my speed was during the track.

It took a while to sort out how to get what I wanted and I now have a working process.

My process is:

a) Find software that records a track without draining the battery too badly (see below);

a) Select the finest/most accurate GPS logging option that the software supports - usually 1s increment;

b) Record the track; and

c) Export it as a GPX file for further processing.

Software:

Alpine Replay: Haven't really played with uploading tracks to the website.

Endomondo Sports Tracker Pro: Didn't do what I wanted.

GPS Logger: Works fine.

Open GPS Tracker: Works very well

Sports Tracker: Works OK

Smart Tracker: Works very well

My Tracks: Works very well

Processing:

This was the fun part. Google Earth can display GPX and KML files but it only shows the tracks, not the speed.

After a fair amount of digging and playing, I thought I would have to write my own KML builder using python/shell script as ESRI ArcMap etc wouldn't do it.

A better Google search uncovered "GPS Visualizer" which reads KML/GPX/CSV files and builds KML's that can be coloured by speed. This worked really well. It also has a heap of other options for colouring the track - Altitude, Time, Climb Rate etc.

http://www.gpsvisualizer.com/map_input?form=googleearth

Result:

This screen shot from Google earth shows some tracks from Loveland yesterday (7 Dec 11) and the estimated speeds.

I'm not sure how accurate the "Estimated Speed" is. My Donek 184 GS is pretty quick and I'm getting better on it but I'm pretty sure that I wasn't going 67.3 mph at a few points on my track......

[TLN]

mattj, you forgot about 1 more program.

Sports track live.

It do everything you want. You can export your KMZ,GPX and watch it on google maps.

Here's my fastest track: http://www.sportstracklive.com/track/detail/kovyrshin/Snowboarding/ridder/427438 It shows 107.4kph that's about 67mph. That was done on soft Never summer board with a straight down run.

Cannnot comment if the speed is true. But it definately shows correct, i mean once i run faster it shows more speed. But my opinion, that you got to correct all the sppeds of a long amount of a time because of a mountains. IMHO GPS system and speed measuring far from excellent in a mountaing or once you losing altitude. For example 10am runs x0.7, 12pm runs x0.8 and so on, because i don't think such a speeds are real.

[Futahaguro]

I use "Maverick" for my Motorola Droid 2.2. It can use downloaded maps, not cached, that you can create from a few programs. I also upload to www.gobreadcrumbs.com when I am done so I can see max speed and share my tracks. Here is one, http://www.gobreadcrumbs.com/user/futahaguro/snowboarding/afton-alps-mn/afton-alps-mn-392011 , and the corresponding screen shot below. It says I went 35 MPH on the last run which sounds right because I do not go that fast and I am in Minnesota where there is not much of a hill to gain speed.</IMG>

[OhD]

Found this quote, had to share...

It's kind of surprising the first time you feel your hand getting warm from the friction on a packed snow surface. I've noticed it on Ruthies myself. Not at 70mph (unless I was experiencing some kind of time dilation and was really going much faster than it seemed). I'm not sure it's plausible that plastic gloves might actually melt - nylon has a much higher melt point than P-tex after all, but on the other hand the unit pressure on the glove might be locally higher than obtains on the base. One can certainly shred nylon and leather on snow!

Does speed influence one's perception of risk more than risk influences one's perception of speed? Or vice versa?

Two more research projects - now let's go see about some funding....

[Puddy Tat]

Two more research projects - now let's go see about some funding....

Mmmmmm.. government funded Kesslers with boiler plates.

[kieran]

for those who want mapped tracks with speed and elevation charts, http://runkeeper.com is quite good. you can import GPX files to it and it'll do the stuff to make it pretty.

this is a route i did on the bike. i logged it using mytracks on android, and uploaded the gpx afterward.

http://runkeeper.com/user/somnambulant_sasquatch/activity/57270366