Author Topic: Digital EPO  (Read 9171 times)

Samuel D

Re: Digital EPO
« Reply #25 on: 19 June, 2017, 05:03:29 pm »
Yeah, that image I can see.

Cudzoziemiec

  • Ride adventurously and stop for a brew.
Re: Digital EPO
« Reply #26 on: 19 June, 2017, 05:59:17 pm »
Yes, I can see that one too. Still can't see it in the original post though.
Riding a concrete path through the nebulous and chaotic future.

Re: Digital EPO
« Reply #27 on: 19 June, 2017, 07:38:42 pm »
My personal favourite segment.

(I'd like to point out I changed my brake pads half way up on my 'PR')

Kim

  • Timelord
    • Fediverse
Re: Digital EPO
« Reply #28 on: 19 June, 2017, 07:43:01 pm »
If it needs new brakes on the way up, I think I'll be giving that one a miss...

rogerzilla

  • When n+1 gets out of hand
Re: Digital EPO
« Reply #29 on: 19 June, 2017, 07:58:29 pm »
The record down Snap Hill (something that takes no real athletic prowess, just a bit of a sprint over the top and then tucking everything in) is about 75mph, if I've correctly worked it out from the times.  I'm doubtful; Snap Hill's not really long enough to get that sort of speed up.  Must look up the times for Pork Hill into Tavistock - that one was supposed to be good for 72mph back in the 1990s.  It was about a mile and a half long, had a good surface, was wide enough to stay off the brakes and the only problem was catching cars.  I hit 56mph on tri-bars (bike was still accelerating) before I had to brake behind a Land Rover  :(
Hard work sometimes pays off in the end, but laziness ALWAYS pays off NOW.

Re: Digital EPO
« Reply #30 on: 10 July, 2017, 11:28:25 am »
Digital EPO is flawed, you can see the peaks on the speed trace when analysing the segment to know if someone has doctored their ride.

That used to be true, the old version would just remove a trackpoint every so often and then shuffle up the timestamps accordingly, so an original track moving at a constant pace would be interpreted as "20kph, 20kph, 20kph, 40kph, 20kph, 20kph, 20kph, 20kph, 20kph, 40kph, 20kph, ..." and easy to spot.

The revised implementation does interpolation between the points to make the speed increase look consistent, that way it can still correlate to the peaks/troughs in the power/cadence/HR rates in the data, but it requires frequent data points (ideally one per second) otherwise the interpolated locations may be off the road (and easy to spot that way).

At one point in time (before they fixed Digital EPO) I was going to write a version that did the above (smoother modifications) but included a way of watermarking the resulting data in such a way that it was detectable by companies like Strava at some point in the future, i.e. change the least significant digit of the lat/lon values to be a checksum against the other digits. I'd wait for a while so lots of people were using it and then hand over the details to Strava/GC/etc so they could flag such rides en masse.
 
A device that logged cryptographically signed trackpoints would be nice.

Thankfully all of these sites seem to have been taken down. Not sure if Strava had a hand in it but its a good thing in any case.

Re: Digital EPO
« Reply #31 on: 14 July, 2017, 02:00:55 pm »
Thankfully all of these sites seem to have been taken down. Not sure if Strava had a hand in it but its a good thing in any case.

If they've all been taken down then maybe I'll resurrect my idea, cram the page with advertising and profit.
"Yes please" said Squirrel "biscuits are our favourite things."

Re: Digital EPO
« Reply #32 on: 18 July, 2017, 10:45:10 am »
Haha! Do it! Theres money to be made!

Phil W

Re: Digital EPO
« Reply #33 on: 18 July, 2017, 11:30:21 am »
Digital EPO is flawed, you can see the peaks on the speed trace when analysing the segment to know if someone has doctored their ride.

That used to be true, the old version would just remove a trackpoint every so often and then shuffle up the timestamps accordingly, so an original track moving at a constant pace would be interpreted as "20kph, 20kph, 20kph, 40kph, 20kph, 20kph, 20kph, 20kph, 20kph, 40kph, 20kph, ..." and easy to spot.

The revised implementation does interpolation between the points to make the speed increase look consistent, that way it can still correlate to the peaks/troughs in the power/cadence/HR rates in the data, but it requires frequent data points (ideally one per second) otherwise the interpolated locations may be off the road (and easy to spot that way).

At one point in time (before they fixed Digital EPO) I was going to write a version that did the above (smoother modifications) but included a way of watermarking the resulting data in such a way that it was detectable by companies like Strava at some point in the future, i.e. change the least significant digit of the lat/lon values to be a checksum against the other digits. I'd wait for a while so lots of people were using it and then hand over the details to Strava/GC/etc so they could flag such rides en masse.
 
A device that logged cryptographically signed trackpoints would be nice.

Thankfully all of these sites seem to have been taken down. Not sure if Strava had a hand in it but its a good thing in any case.

Surely it'd be easier to just shift the timestamps and keep the trackpoints exactly where they are if you want to increase the speed?

Re: Digital EPO
« Reply #34 on: 18 July, 2017, 11:34:23 am »
I upload some rides to Strava: criteria being whether I had the Etrex mounted and whether  I could be arsed.  They're manual uploads after the fact, and I have no idea where segments begin or end so no temptation to flail the aging legs.

Re: Digital EPO
« Reply #35 on: 18 July, 2017, 11:35:50 am »
Surely it'd be easier to just shift the timestamps and keep the trackpoints exactly where they are if you want to increase the speed?

If you want to maintain 1 point per second recording then that won't work as you don't have sub-second timing granularity.

If the original file wasn't 1 point per second then it's relatively easy (especially to make it look ok enough on Strava) but modifying a file so that it passes closer scrutiny is quite a bit harder.
"Yes please" said Squirrel "biscuits are our favourite things."

Phil W

Re: Digital EPO
« Reply #36 on: 18 July, 2017, 04:41:26 pm »
Surely it'd be easier to just shift the timestamps and keep the trackpoints exactly where they are if you want to increase the speed?

If you want to maintain 1 point per second recording then that won't work as you don't have sub-second timing granularity.

If the original file wasn't 1 point per second then it's relatively easy (especially to make it look ok enough on Strava) but modifying a file so that it passes closer scrutiny is quite a bit harder.


Erm my Etrex doesn't do 1 point per second recording and neither do many other GPS.  So that's a non point  ;D  Modification is piss easy, you just adjust the timestamps to change the speed by an uplift of say 10% or something. No one is going to be able to tell, least of all an automated tool as there's no pattern to find. As long as you're not trying to change your data to be Mark Cavendise sprint type level. In which case your mates might twig.

frankly frankie

  • I kid you not
    • Fuchsiaphile
Re: Digital EPO
« Reply #37 on: 19 July, 2017, 09:01:06 am »
The older Etrex C models do have the option to record once per second - haven't checked if this is also true of the newer models [edit: yes my Etrex 30 has this option too].  Not that I'm commending it - ruddy daft if you ask me.
when you're dead you're done, so let the good times roll

Kim

  • Timelord
    • Fediverse
Re: Digital EPO
« Reply #38 on: 19 July, 2017, 12:46:31 pm »
Point per second has its uses (it's probably the least-worst option if you're doing laps of a short track), but most of them aren't relevant to what we normally record rides for.  Garmins generally do a good job of working out when to record a track point on their own.

Re: Digital EPO
« Reply #39 on: 20 July, 2017, 05:18:41 pm »
Anyone using a power meter is recording at 1 point per second, otherwise you're throwing away an awful lot of data.

Also the cycling specific Edge series of Garmins default (I think) to 1 point per second recording if certain accessories are detected (HR strap, cadence, power meter) unless otherwise configured.

I'd also guess that the Garmin Edge is the most likely make/line of GPSes for Strava uploads from cyclists (excluding phone apps).

A further guess it that the largest contingent of uploads to the original Digital EPO sites were originally recorded on Garmin Edge devices. It's not about cycle tourists using Etrexes.

Modification is piss easy, you just adjust the timestamps to change the speed by an uplift of say 10% or something.

As I said, coming up with a basic but vaguely convincing fake (e.g. trackpoints lat/lon/timestamp/elevation every 15 seconds) would be quite easy. Coming up with something that passes closer scrutiny is a lot harder, i.e. something that contains power, cadence and HR data at 1 second intervals.
"Yes please" said Squirrel "biscuits are our favourite things."