Discussion
Have a look at veloviewer and that TT segment. Once you get through the initial reading your rides etc there is an alternative leaderboard you can get that shows more data. The distances people ride for a segment can vary wildly as Strava uses the nearest sample point to start and finish and it may explain it.
raceshape.com also good for analysing a segment if it's puzzling you
raceshape.com also good for analysing a segment if it's puzzling you
upsidedownmark said:
2) Timing - the fixes delivered by the API aren't timestamped; the device is doing other stuff in the background, thus there will be some delay in the delivery of the GPS fix to strava. If that's reasonably constant, it's not an issue, but if that is variable (and it is), the strava app doesn't know - it can only take the time of delivery as the time of fix. Over short segments those sort of issues could lead to big differences: Say on entry to a short segment, the phone is 'busy' - it takes a while for the fix to get through.. then it frees up and catches up, taking a few seconds off the run..
Maybe not from the GPS API, but the iPhone itself has accurate enough timing hardware anyway. I really don't think the app could be programmed in the way you suggest, or it would be massively inaccurate every time the network data signal was poor. Or are you talking about scheduling between apps/within the kernel? In which case I would be very interested to know quite how big the latency there is? It would be very surprising to me if that was anything noticeable.
More likely that the inaccuracies come from the GPS hardware itself, which is only accurate to approx 5m long/lat and 1.5x that in alt.
Edited by budgie smuggler on Thursday 6th June 10:16
Dizeee said:
Did a 10 mile official TT tonight, timed et al.
......
It seems, Strava has failed here.
How did he start it? Because Strava goes off way points, as it were, and TTs are started at a line, if (while stood statoinary at the line waiting for his time to start) Strava picks him up as having crossed the start of the segment then that's it - his time's started.......
It seems, Strava has failed here.
TonyHetherington said:
How did he start it? Because Strava goes off way points, as it were, and TTs are started at a line, if (while stood statoinary at the line waiting for his time to start) Strava picks him up as having crossed the start of the segment then that's it - his time's started.
That could explain it, he has since said that when he started his Garmin off before he was moving, it had him as moving for some reason, so maybe thats where the issue lies.Dizeee said:
TonyHetherington said:
How did he start it? Because Strava goes off way points, as it were, and TTs are started at a line, if (while stood statoinary at the line waiting for his time to start) Strava picks him up as having crossed the start of the segment then that's it - his time's started.
That could explain it, he has since said that when he started his Garmin off before he was moving, it had him as moving for some reason, so maybe thats where the issue lies.
That could do it then. Does he have a wheel speed sensor, or just uses GPS ?
A stationary Garmin can think you're 5m that way or 5m that way as it checks satellites etc, and because it has no previous direction to calculate from it does appear a little random. So while waiting it's likely his Garmin momentarily throught he'd gone through the beginning of the segment - especially if the segment created is exactly on the start line.
Might be more sensible making the segment start 100m after the beginning of the TT.
A stationary Garmin can think you're 5m that way or 5m that way as it checks satellites etc, and because it has no previous direction to calculate from it does appear a little random. So while waiting it's likely his Garmin momentarily throught he'd gone through the beginning of the segment - especially if the segment created is exactly on the start line.
Might be more sensible making the segment start 100m after the beginning of the TT.
Dizeee said:
Well, seems Strava ain't faultless.
Did a 10 mile official TT tonight, timed et al.
Mate of mine started 1 minute behind me. We both did the 10 mile course no issues.
As I am finishing my mate is not too far behind, and as I turn arounf having finished I see him end, and as he finished around 30 seconds after me, yet started a minute later, I tell him he is looking at a time around 30 seconds quicker.
This is confirmed when the results sheet appears, I finish 24:49 and he finishes around 30 seconds earlier.
We then both upload to Strava.
The results are very different. Strava has me at my correct time - 24:49 (which I was impressed with the accuracy as I started it as I started and ended when I ended). It however has him at 25:41, and a few places down.
Both are today's date, both are fresh Garmin data and it even has us riding together as you would expect. I have spent all night trying to work it out, and can't.
It picked up the TT segment so he has not shortcutted it. He was timed by a human as was I, and I know he was quicker as I saw him finish just after I did despite setting off later.
It seems, Strava has failed here.
All this is why (currently) GPS-based devices are rough tools, not exact measurements. Every running race I do I hear people complaining that their Garmin/Phone/etc says the course was long/short. To get a race permit (and presumably for official TTs too) the course has to be measured by a certified official. Yes, occasionally mistakes happen - I thought I'd run a 10k pb on a hilly road course a few years ago, it turned out they'd put the finish in the wrong place! However, the vast majority of the time courses are correctly measured and setup.Did a 10 mile official TT tonight, timed et al.
Mate of mine started 1 minute behind me. We both did the 10 mile course no issues.
As I am finishing my mate is not too far behind, and as I turn arounf having finished I see him end, and as he finished around 30 seconds after me, yet started a minute later, I tell him he is looking at a time around 30 seconds quicker.
This is confirmed when the results sheet appears, I finish 24:49 and he finishes around 30 seconds earlier.
We then both upload to Strava.
The results are very different. Strava has me at my correct time - 24:49 (which I was impressed with the accuracy as I started it as I started and ended when I ended). It however has him at 25:41, and a few places down.
Both are today's date, both are fresh Garmin data and it even has us riding together as you would expect. I have spent all night trying to work it out, and can't.
It picked up the TT segment so he has not shortcutted it. He was timed by a human as was I, and I know he was quicker as I saw him finish just after I did despite setting off later.
It seems, Strava has failed here.
So, treat Strava as an additional tool but it's the official timekeepers time that matters and if you're just out for a ride (rather than racing) then do Strava/GPS glitches really matter at all?
TKF said:
Aside from the Strava issues a sub 25 for a 10TT on a standard road bike is impressive.
While I don't want to piss on Dizee's chips, I don't think the segment he quoted was a full 10. More like 9.2 
Good on you for going and doing a TT though, would like to give it a go myself.
TKF said:
Aside from the Strava issues a sub 25 for a 10TT on a standard road bike is impressive.
Thanks, however my actual time was 25:40 for the course, the segment stops just short of the end so cuts a bit off. Weirdly most of those that do this particular TT do it on a standard roadie, only around 5 aero TT bikes out of 25.ChrisMCoupe said:
While I don't want to piss on Dizee's chips, I don't think the segment he quoted was a full 10. More like 9.2
Good on you for going and doing a TT though, would like to give it a go myself.
Thanks, yes thats right the segment cuts off a bit at the end. Its the 3td time I have done it now, did it on the Rosa initially at 26:13, then next time took the Bianchi and did 26:31, then yesterday I tried to make more use of the drops and managed 25:40 which I am thrilled with - Garmin say's I averaged 22.3mph. Good on you for going and doing a TT though, would like to give it a go myself.
Think I will need to stop smoking completely to get any faster
THIS is the correct trace for the charlotteville 10, so not sure what you were doing as your thing looks totally different? http://app.strava.com/segments/1606948
IF the segment is created properly then there is no issue, all the TT's I have done the actual time has been within a couple of seconds of the reported one, and the distance is listed as exactly ten miles or whatever.
Regardless of that, around 25 is certainly decent on a roadbike. Not sure of that course, as obviously courses make a big difference, that one doesn't look quick as its not on a dual carriageway or anything like that.
IF the segment is created properly then there is no issue, all the TT's I have done the actual time has been within a couple of seconds of the reported one, and the distance is listed as exactly ten miles or whatever.
Regardless of that, around 25 is certainly decent on a roadbike. Not sure of that course, as obviously courses make a big difference, that one doesn't look quick as its not on a dual carriageway or anything like that.
Traffic suck, no corners to negotiate mostly, and any downhill obviously you can go at full pace as no need to worry about stopping really. Yes, many of them have 'grippy' surfaces, however.
The one i set my PB on (19.13) is an odd one, first two miles are total s
te, potholes left right and center, after the first turn its smooth, then you go down an absoloute ski slope (48 mph down there in tt bars) and then there is another turn to make up for the ski slope. TT'ing is an odd world, where people will do anything and race on any course if it means a PB. That particular course however (f11/10) is quite safe, but some of them are on fairly busy roads.
The one i set my PB on (19.13) is an odd one, first two miles are total s
te, potholes left right and center, after the first turn its smooth, then you go down an absoloute ski slope (48 mph down there in tt bars) and then there is another turn to make up for the ski slope. TT'ing is an odd world, where people will do anything and race on any course if it means a PB. That particular course however (f11/10) is quite safe, but some of them are on fairly busy roads.budgie smuggler said:
Maybe not from the GPS API, but the iPhone itself has accurate enough timing hardware anyway. I really don't think the app could be programmed in the way you suggest, or it would be massively inaccurate every time the network data signal was poor.
Or are you talking about scheduling between apps/within the kernel? In which case I would be very interested to know quite how big the latency there is? It would be very surprising to me if that was anything noticeable.
More likely that the inaccuracies come from the GPS hardware itself, which is only accurate to approx 5m long/lat and 1.5x that in alt.
Network data / signal doesn't come into it - I'm talking about the difference between the time the GPS hardware determines a fix, and the time that fix gets delivered to the strava app within the phone. On an iPhone, that fix only gets timestamped when it gets to the app.Or are you talking about scheduling between apps/within the kernel? In which case I would be very interested to know quite how big the latency there is? It would be very surprising to me if that was anything noticeable.
More likely that the inaccuracies come from the GPS hardware itself, which is only accurate to approx 5m long/lat and 1.5x that in alt.
Edited by budgie smuggler on Thursday 6th June 10:16
The fixes are then uploaded to strava which calculates the segments etc by looking at where you were and when; if the when is off then you're looking at some decent skew in the results: for example if the fix is 1/2 a sec late at 30mph, it's some 7m late. If it was a consistent 1/2 sec lag, that wouldn't matter, but when the lag varies then you get spikes. This is why stuff like harry's laptimer works far better with external GPS sources that communicate via their own API and provide fixes that are timestamped at the source (thus it doesn't matter how long the delivery takes)
Bear in mind that the iphone is far from a real-time env as far as apps are concerned: it is task switching / running apps at a pretty low priority. For general computing use I'd agree, for iphones, not so much
WRT GPS accuracy, you have a point, but: GPS error is pretty systematic. Therefore the accuracy of the fix might not be that great, but the consistency is much better. Consistent inaccuracy won't hurt, inconsistent accuracy will. GPS will generally be 'off' by a steady amount in any given location. At a simple level this is how DGPS/augmentation systems work - there is a GPS receiver at an accurately known location that compares it's calculated pos to the known pos and transmits a correction factor.
Also neatly demonstrated by my logging my commute on my iphone: I'm damn sure I didn't hit 100kph on a fixie - no such glitch on the garmin
okgo said:
THIS is the correct trace for the charlotteville 10, so not sure what you were doing as your thing looks totally different? http://app.strava.com/segments/1606948
The course was changed for 6 weeks due to roadworks on it.okgo said:
Regardless of that, around 25 is certainly decent on a roadbike. Not sure of that course, as obviously courses make a big difference, that one doesn't look quick as its not on a dual carriageway or anything like that.
Its an irritating course because you have to negotiate 9 very sharp left turn T junctions - this is because you have to do 3 laps round a circuit with 3 seperate junctions. I would like to think I could do it quicker if it was just a flowing ten miles, I am probably not the best cornerer and sometimes you have to slow a bitr extra for traffic at the junctions.S10GTA said:
Any body else having issues processing their ride today?
Yep! Just tried uploading direct from my Garmin, some sort of processing error, uploaded fine to garmin connect but even exporting the TCX or GPX and uploading hasn't worked...Too many people out enjoying the sunshine?
Gassing Station | Pedal Powered | Top of Page | What's New | My Stuff