Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie
Hi all! We have been experiencing an issue on site where threads have been missing the latest postings. The platform host Vanilla are working on this issue. A workaround that has been used by some is to navigate back from 1 to 10+ pages to re-sync the thread and this will then show the latest posts. Thanks, Mike.
Hi there,
There is an issue with role permissions that is being worked on at the moment.
If you are having trouble with access or permissions on regional forums please post here to get access: https://www.boards.ie/discussion/2058365403/you-do-not-have-permission-for-that#latest

GPS accuracy. MAKE THIS A STICKY!

  • 23-04-2014 1:13pm
    #1
    TRR
    Closed Accounts Posts: 2,642 ✭✭✭


    So we've all heard the arguments regards GPS accuracy. How one persons GPS measures a course long while another measures it short etc etc blah blah, high buildings, bridges, tree cover blah blah blah

    Well at last someone has done a pretty decent study of GPS accuracy. Can we just make this a sticky and point all future GPS queries towards it! It's not perfect but it is significantly better than most peoples sample of 1.

    http://fellrnr.com/wiki/GPS_Accuracy


Comments

  • #2
    Cabaal
    Moderators, Technology & Internet Moderators, Regional South East Moderators Posts: 28,510 Mod ✭✭✭✭Cabaal


    Honestly didn't think this topic comes up all that much,

    Sure from time to time it pops up but its hardly the most commonly asked question on the forum?

    Glad to see my Garmin Forerunner 310XT isn't awful :)


  • #3
    adrian522
    Moderators, Computer Games Moderators, Recreation & Hobbies Moderators, Sports Moderators Posts: 16,139 Mod ✭✭✭✭adrian522


    Guess I won't be upgrading from the 610 to the 620 then.


  • #4
    TRR
    Closed Accounts Posts: 2,642 ✭✭✭TRR


    Cabaal wrote: »
    Honestly didn't think this topic comes up all that much,

    Comes up a lot of the threads in the events subforum. I was just being a smart arse regards the sticky aspect.


  • #5
    emerald007
    Registered Users, Registered Users 2 Posts: 667 ✭✭✭emerald007


    THE MEASUREMENT OF ROAD RACE COURSES outlines how a course should be measured.


  • #6
    Krusty_Clown
    Registered Users, Registered Users 2 Posts: 19,534 ✭✭✭✭Krusty_Clown


    Good investigation, but he seems to have overlooked (or just not mentioned) one variable - the satellites. He should really be testing all of the devices simultaneously over his established test route, as the satellite cover (visibility), will be different at different times (based on the satellites' orbital paths) which will impact accuracy. If your sample is large enough this will get normalized, but I'd imagine you'd need pretty massive sample sizes.


  • Advertisement
  • #7
    rom
    Registered Users, Registered Users 2 Posts: 5,142 ✭✭✭rom


    I can verify that the 620 is a bucket of .... Garmin 305 was the best watch ever.

    iPhone 4s . Now its only as good as the program you use. If you use a program that has a low sample time to save satellite loads then not much good.


  • #8
    TRR
    Closed Accounts Posts: 2,642 ✭✭✭TRR


    Krusty_Clown wrote: »
    Good investigation, but he seems to have overlooked (or just not mentioned) one variable - the satellites. He should really be testing all of the devices simultaneously over his established test route, as the satellite cover (visibility), will be different at different times (based on the satellites' orbital paths) which will impact accuracy. If your sample is large enough this will get normalized, but I'd imagine you'd need pretty massive sample sizes.

    he does mention about testing them on different days (point 5) and how this has an impact so I guess that does address it somewhat!


  • #9
    ger664
    Registered Users, Registered Users 2 Posts: 2,402 ✭✭✭ger664


    I agree it should be a sticky. Still people wont read it and still claim a courses inaccuracy because it doesn't line up on there overpriced inaccurate watch.
    Had a muppet at a race claim a course was long recently because they just missed out on a sub 40 for 10K. He was giving the RD a bit of an earful which was unfair as the course was bang on from a measurement prospective.

    People using them to train seem to lose sight of the fact the 10 Miles on a garmin is less then 10 miles and 8 minute/mile pace is probably nearer to 8:05 pace.


  • #10
    Krusty_Clown
    Registered Users, Registered Users 2 Posts: 19,534 ✭✭✭✭Krusty_Clown


    TRR wrote: »
    he does mention about testing them on different days (point 5) and how this has an impact so I guess that does address it somewhat!
    He seems to put it down to the watch having a bad day though and makes no connection with satellite locations/connectivity. Here's a site that lets you track the orbital paths. Not relevant to the conversation, just interesting! Navstar 67 will be flying overhead shortly.


  • #11
    TRR
    Closed Accounts Posts: 2,642 ✭✭✭TRR


    Krusty_Clown wrote: »
    He seems to put it down to the watch having a bad day though and makes no connection with satellite locations/connectivity.

    He does have analysis for the Garmin 310xt and the Garmin 620 taken "at the same time" (section 6)
    The same for Section 5 (310 and 910 watches). He states that he has "accumulated a lot of data to do the statistical analysis work". So whilst he cannot test EVERY device at the same time (not enough arms or hands... or enough people to run in the exact same place at the exact same pace with a single device) he's done an incredible job.

    He cannot control the weather or the satellites. It's not feasible to be out in the dark with the them so there is a finite window for testing. He's also plotted normal distributions (kernel densities) for most of the watches. So this shows that there is enough data to allow for normalisation.

    Look at the count field in Section 3 (first table) - he's taken around 1000 measurements for each device and each measurement has 1000 data points in it. There is a serious analysis!


  • Advertisement
  • #12
    Peterx
    Registered Users, Registered Users 2 Posts: 2,118 ✭✭✭Peterx


    Krusty_Clown wrote: »
    Here's a site that lets you track the orbital paths. Not relevant to the conversation, just interesting! Navstar 67 will be flying overhead shortly.

    From that website I see in Ireland are at roughly -4 degrees declination. I've been using -6 degrees for going from map to compass. Close enough for me :)


  • #13
    spurscormac
    Registered Users, Registered Users 2 Posts: 1,523 ✭✭✭spurscormac


    Would be interesting to see dcrainmaker try this out with his rolling pin of watches over the same course multiple times.
    Broader spread of devices and each recording the same route at the same time - how you would synch them all to start at the same time, I don't know though ;-)


  • #14
    Krusty_Clown
    Registered Users, Registered Users 2 Posts: 19,534 ✭✭✭✭Krusty_Clown


    TRR wrote: »
    Look at the count field in Section 3 (first table) - he's taken around 1000 measurements for each device and each measurement has 1000 data points in it. There is a serious analysis!
    The average Garmin watch will product around 460 data points per mile. So when he refers to 14,000 data points, isn't that around 30 miles of data? He does mention 3,500 miles of data though. Not trying to pull holes (I find nothing wrong with his analysis), just interested. Haven't done statistics for many, many years, so relying on your input!


  • #15
    viperlogic
    Registered Users, Registered Users 2 Posts: 1,304 ✭✭✭viperlogic


    Its hit and miss with the new Garmin range (FR220, FR620, Fenix2), some days my FR620 is bang on other days its off, and when I say off I mean a part of the track is the right shape but offset by x meters but distance has always been correct. Currently have FR620, with 2.6 FW and 3.1 GPS, for comparison I used to have a FR410. But overall, it has a lot more good days than bad days.

    Few friends were having issues with their FR220 and FR620 but after a FW update to 2.6, GPS update to 3.1, and a hard reset they are happy enough with the watch now from a GPS accuracy point of view. There are other issues, but I wont get into them as not GPS related.

    The watch gets a GPS signal usually in under 10 seconds but I always wait a few minutes for the elevation to stop jumping around, perfect time for some static and dynamic stretches! Pity the FR620 doesn't have a GPS accuracy field like the FR410, therefore I use elevation field. I'm living up a hill and when it gets a satellite lock within 5-10 seconds and says elevation is -14, you know its not accurate! The quick satellite fix is just a sales/marketing gimmick, sure my FR410 got a lock within a few seconds also but took a few minutes before it locked in on a few satellites and accuracy was within a few meters before the watch prompted satellite/GPS lock got (or what ever the message was!) and beeped.


  • #16
    HelenAnne
    Closed Accounts Posts: 2,035 ✭✭✭HelenAnne


    TRR wrote: »
    So we've all heard the arguments regards GPS accuracy. How one persons GPS measures a course long while another measures it short etc etc blah blah, high buildings, bridges, tree cover blah blah blah

    Well at last someone has done a pretty decent study of GPS accuracy. Can we just make this a sticky and point all future GPS queries towards it! It's not perfect but it is significantly better than most peoples sample of 1.

    http://fellrnr.com/wiki/GPS_Accuracy

    I still think my idea of adding some kind of word-recognition software to the events forum, so that whenever the words 'My Garmin measured it ...' are detected the comment will be deleted, is the best way to go :-)


  • #17
    morceli
    Registered Users Posts: 364 ✭✭morceli


    I think all GPS watches should be banned from races just like headphones, its a form of cheating. If you can't pace yourself right without the aid of a 3'rd party you shouldn't be running.


  • #18
    adrian522
    Moderators, Computer Games Moderators, Recreation & Hobbies Moderators, Sports Moderators Posts: 16,139 Mod ✭✭✭✭adrian522


    morceli wrote: »
    I think all GPS watches should be banned from races just like headphones, its a form of cheating. If you can't pace yourself right without the aid of a 3'rd party you shouldn't be running.

    Oh dear, should we ban runners as well? If you can't run barefoot you shouldn't be running? What about gels, water stations, mile markers? Let's ban running altogether while we are at it.


  • #19
    morceli
    Registered Users Posts: 364 ✭✭morceli


    adrian522 wrote: »
    Oh dear, should we ban runners as well? If you can't run barefoot you shouldn't be running? What about gels, water stations, mile markers? Let's ban running altogether while we are at it.
    Nope runners are ok, i'd just ban barefoot runners actually. People should just race and not pace, was at a 10 mile race on monday and the amount of time people spent looking at their arm was just silly.


  • #20
    rom
    Registered Users, Registered Users 2 Posts: 5,142 ✭✭✭rom


    morceli wrote: »
    I think all GPS watches should be banned from races just like headphones, its a form of cheating. If you can't pace yourself right without the aid of a 3'rd party you shouldn't be running.

    A GPS watch results in a person running slower than they can. They are simply a training aid. If you try running a race without a watch you have a higher sense of arousal (not that kind :) ). Even better if you are racing people that you know that they are normally a bit faster than you. Get ahead and stay and run scared. I know I under achieve with a watch on my hand. Ran great in all XC I did without a watch.


  • #21
    fellrnr
    Registered Users Posts: 1 ✭fellrnr


    Krusty_Clown wrote: »
    The average Garmin watch will product around 460 data points per mile. So when he refers to 14,000 data points, isn't that around 30 miles of data? He does mention 3,500 miles of data though. Not trying to pull holes (I find nothing wrong with his analysis), just interested. Haven't done statistics for many, many years, so relying on your input!

    Excellent question; I'd not thought of it from that perspective. Each count is the measurement of a quarter mile segment, so each count represents far more GPS point recordings, though that would vary by device.

    I'll add a note that hopefully clarifies:

    "The count field is how many measurements I have for that combination of condition and device, with each measurement being a quarter mile distance"


  • Advertisement
  • #22
    opus
    Registered Users, Registered Users 2 Posts: 5,027 ✭✭✭opus


    rom wrote: »
    I can verify that the 620 is a bucket of .... Garmin 305 was the best watch ever.

    Yup would certainly agree with that, Garmin replaced my original 620 but the new one doesn't seem much better. Some days works well, other days goes nuts on the same route.

    Currently waiting for my replacement HRM as well as there was some defect in the original batch that ate the battery after ~3 months.


  • #23
    Krusty_Clown
    Registered Users, Registered Users 2 Posts: 19,534 ✭✭✭✭Krusty_Clown


    fellrnr wrote: »
    Excellent question; I'd not thought of it from that perspective. Each count is the measurement of a quarter mile segment, so each count represents far more GPS point recordings, though that would vary by device.
    Good to have you on board, fellrnr and thanks for the very complete analysis (and other useful materials on your site). Any idea why the 620 performed so poorly? Have you used the watch for general use outside of your test route and did it perform equally as bad? It seems to use the same GPS chipset as the 220, so I'd imagine that the 220 must be plagued by the same issues.


  • #24
    sconhome
    Registered Users, Registered Users 2 Posts: 5,468 ✭✭✭sconhome


    As the resident Polar fan I'm curious as to why the only Polar watch tested was the RC3 GPS. Possibly only considering inbuilt GPS?


  • #25
    crisco10
    Registered Users, Registered Users 2 Posts: 6,221 ✭✭✭crisco10


    rom wrote: »
    A GPS watch results in a person running slower than they can. They are simply a training aid. If you try running a race without a watch you have a higher sense of arousal (not that kind :) ). Even better if you are racing people that you know that they are normally a bit faster than you. Get ahead and stay and run scared. I know I under achieve with a watch on my hand. Ran great in all XC I did without a watch.

    Slightly OT but..
    Would echo that....have had a few cracking races when I gave my Garmin to the otherhalf and just used a €10 casio to check for my final time.

    I think for shorter distances the Garmin "bullys" me into pacing heavily in the early half of a race and I never get that time back. Where as without, I just run as fast as I feel I can from the start. I think, for my next race I'll wear the garmin but with black tape on the screen; that way I still have my splits/gpx file but can't look at it mid race!


  • #26
    Krusty_Clown
    Registered Users, Registered Users 2 Posts: 19,534 ✭✭✭✭Krusty_Clown


    AKW wrote: »
    As the resident Polar fan I'm curious as to why the only Polar watch tested was the RC3 GPS. Possibly only considering inbuilt GPS?
    Possobly because that's the only one he had access to! Maybe you should ship him a different Polar? ;)


  • #27
    snailsong
    Registered Users, Registered Users 2 Posts: 1,140 ✭✭✭snailsong


    rom wrote: »
    I can verify that the 620 is a bucket of .... Garmin 305 was the best watch

    I sent my 305 back to Garmin as the battery life was very poor after 4 years heavy use.

    Last week received a brand new 305 in the post. Happy out. Bargain at €73.


  • #28
    morceli
    Registered Users Posts: 364 ✭✭morceli


    snailsong wrote: »
    I sent my 305 back to Garmin as the battery life was very poor after 4 years heavy use.

    Last week received a brand new 305 in the post. Happy out. Bargain at €73.
    Thats not a bad deal at all. Might do the same myself in a few months.


Advertisement