User Tag List

Results 1 to 15 of 15

Thread: Garmin accuracy - grade on climbs

  1. #1
    Join Date
    Nov 2010
    Location
    My own private Idaho
    Posts
    1,025
    Post Thanks / Like
    Mentioned
    1 Post(s)
    Tagged
    0 Thread(s)

    Default Garmin accuracy - grade on climbs

    Anyone know how a Garmin head unit determines grade on a climb (in my case a 520S)? How accurate is it? Is it dependent on the unit being mounted perfectly parallel to the ground?
    I know that there's a delay of a few seconds in the reading change on rolling terrain. However, sometimes even on steady climbs (or descents), I see readings that don't concur with what I see visually or feel on the bike (effort, gearing, etc.). Not a big deal, but I've had this random bit of curiosity a few times on rides. Thanks.
    Lou D'Amelio
    Bucks County PA

  2. #2
    Join Date
    Sep 2014
    Location
    Stow, MA
    Posts
    4,167
    Post Thanks / Like
    Mentioned
    9 Post(s)
    Tagged
    0 Thread(s)

    Default Re: Garmin accuracy - grade on climbs

    My 820 deeply undercounts elevation gained compared to what RideWithGPS says it should be... I never look at the grade so I can't help there...
    Guy Washburn

    Photography > www.guywashburn.com
    "I think that the whole Republican Party has been put in a position, with a few notable exceptions, of being mildly intimidated by the sitting President." Joe Biden, President Elect.
    “A lawsuit without provable facts showing a statutory or constitutional violation is just a tweet with a filing fee,” said Justin Levitt, a professor at Loyola Law School in Los Angeles.
    “Instructions for living a life: Pay attention. Be astonished. Tell about it.”
    – Mary Oliver

  3. #3
    Join Date
    Jun 2013
    Posts
    106
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Re: Garmin accuracy - grade on climbs

    I've always understood that especially with consumer grade GPS, the horizontal accuracy is much better than the vertical. Plus, I suppose you'd notice vertical difference more than horizontal as well.
    My name is David Moeny

  4. #4
    Join Date
    Jun 2008
    Posts
    13,912
    Post Thanks / Like
    Mentioned
    13 Post(s)
    Tagged
    0 Thread(s)

    Default Re: Garmin accuracy - grade on climbs

    I don’t really trust the vertical data. Saturday and Sunday I did a pair of identical 25 mile rides. Exactly the same loop each day. The climbing data was hundreds of feet different on less than 1000’ of vertical total elevation change.
    La Cheeserie!

  5. #5
    Join Date
    Mar 2010
    Location
    NJ
    Posts
    2,116
    Post Thanks / Like
    Mentioned
    8 Post(s)
    Tagged
    0 Thread(s)

    Default Re: Garmin accuracy - grade on climbs

    It has nothing to do with the unit being parallel to the ground, it doesnt work like a bubble level.

    see this explanation, it's basically a rolling trigonometry calculation.:

    https://galebernhardt.com/blogs/news...-percent-grade

    On a recent mountain bike ride, a couple of us with Garmin devices began to wonder how percent grade is determined by the device. Through a local contact, I was able to go directly to the design source, Jared Bancroft, PhD – Sensors & Algorithms Team Leader.

    Here is what Jared told me:

    Grade is responsive to the slope of the trail or road. It relies on the distance as determined from GPS and the barometric altimeter sensor. Basically, it’s a ratio of the change in altitude over distance traveled. The longer the distance, the more averaged the grade is. We try to use the smallest distance, but if the distance is too small it looks really noisy and the grade bounces all over the place. Also, anytime GPS isn’t available or sufficiently accurate enough (e.g. heavy tree coverage in the mountains) grade also suffers. That said, if you have a speed sensor, percent grade comes in more accurately since we can use the speed sensors distance rather than the GPS based distance that may or may not be correct.

    Grade typically responds within several seconds. It won’t react to large rocks, wheelies, or popping up your front tire. Its more trying to emulate the slope of the terrain you take. We have to filter the results so that delays it a bit.

    Accuracy of percent grade is more based on GPS reception quality than time on the hill. Small dips get averaged out but longer hills are usually pretty good.

    Grade should be consistent on rides. If you take the same path, it should be the same (both by definition and seen on the device). That said, the distance grade is calculated over, the smoothing filter and the GPS quality all play a role in the final number you see on your device.

    Now you know!

  6. #6
    Join Date
    Aug 2010
    Location
    W Mass
    Posts
    11,808
    Post Thanks / Like
    Mentioned
    8 Post(s)
    Tagged
    1 Thread(s)

    Default Re: Garmin accuracy - grade on climbs

    I just looked at eight rides in the woods with the dog, just under 60 miles.

    My vertical was 5900 feet, the dog's was 8100. Almost 40% more. On four of the rides, the dog's vertical was double mine. What's interesting is that the dog's mileage is actually 3% lower. She does cut off parts of the trail and follows contour lines, but I'm skeptical of that difference.

    I use Strava on my phone, the dog wears a Garmin wristwatch on her collar.

    You can actually see a burp on some rides where there's a step change in elevation (but not in position). Other than the fact that the dog regularly goes swimming, there's not much difference in our routes. We're seldom, if ever, not in visual contact. She's a shepherd and thinks this is her job.

    I think it's likely a precision vs. accuracy thing. With friends on a 60-mile ride with 6000 feet of climbing, our verticals will be within a few hundred feet. On a six-mile ride with the dog, we can be a few hundred feet off. So I think asking the GPS to accurately read the grade of climb while you're on it is asking too much of the instrument.
    Last edited by thollandpe; 09-21-2020 at 10:36 AM.
    Trod Harland, Physical Educator

    Not everything that is faced can be changed, but nothing can be changed until it is faced. — James Baldwin

  7. #7
    Join Date
    Mar 2010
    Location
    NJ
    Posts
    2,116
    Post Thanks / Like
    Mentioned
    8 Post(s)
    Tagged
    0 Thread(s)

    Default Re: Garmin accuracy - grade on climbs

    Quote Originally Posted by thollandpe View Post
    I use Strava on my phone, the dog wears a Garmin wristwatch on her collar.
    but a major difference to consider here is that some garmin's have built in barometric altimeters, where as some are relying on gps alone to determine elevation.

  8. #8
    Join Date
    Jun 2012
    Location
    Carrollton TX
    Posts
    545
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Re: Garmin accuracy - grade on climbs

    Ha ! Garmin altitude. Meaningless.

    I carry a Garmin 500 in my pocket (so it cannot distract me while riding).
    On a ride last week, I gained 400 feet of altitude while stopped at a gas station getting some calories.
    Mark Walberg
    Building bike frames for fun since 1973.

  9. #9
    Join Date
    Aug 2010
    Location
    W Mass
    Posts
    11,808
    Post Thanks / Like
    Mentioned
    8 Post(s)
    Tagged
    1 Thread(s)

    Default Re: Garmin accuracy - grade on climbs

    Quote Originally Posted by AngryScientist View Post
    but a major difference to consider here is that some garmin's have built in barometric altimeters, where as some are relying on gps alone to determine elevation.
    Interesting. The Fenix 2's altimeter is barometric. So maybe my dog's habit of full immersion in any body of water more than 4 inches deep may be why the altitude readings are skewed?
    Trod Harland, Physical Educator

    Not everything that is faced can be changed, but nothing can be changed until it is faced. — James Baldwin

  10. #10
    Join Date
    Apr 2011
    Location
    New York City
    Posts
    17,957
    Post Thanks / Like
    Mentioned
    18 Post(s)
    Tagged
    0 Thread(s)

    Default Re: Garmin accuracy - grade on climbs

    Quote Originally Posted by ldamelio View Post
    Anyone know how a Garmin head unit determines grade on a climb (in my case a 520S)? How accurate is it? Is it dependent on the unit being mounted perfectly parallel to the ground?
    I know that there's a delay of a few seconds in the reading change on rolling terrain. However, sometimes even on steady climbs (or descents), I see readings that don't concur with what I see visually or feel on the bike (effort, gearing, etc.). Not a big deal, but I've had this random bit of curiosity a few times on rides. Thanks.
    Someone (here or elsewhere) told me that the grade/ascent information is interpreted by the service used. So if you record with Garmin and upload that data to Garmin Connect, Strava and Ride With GPS, you will get different numbers for each service based on each site's interpretive dance with the numbers. I think Strava has a recalibration feature that gets Strava to do some kind of matchup with your route and known elevation data from maps, but otherwise my impression is that altimeter data reporting on devices and ride data sites is merely a relative figure and not absolutely accurate. I know my wife's computer and mine - both Wahoo Bolts - are often hundreds of feet off from each other after a ride.
    Jorn Ake
    poet

    Flickr
    Books

  11. #11
    Join Date
    Aug 2012
    Location
    Washington DC metropolitan area
    Posts
    19
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Re: Garmin accuracy - grade on climbs

    Glad its not just me (or my two Garmin 520 Plus devices). I have had three of these units (exchanged one). All three have taken quite a long time to begin registering for a climb - longer than a few seconds. Also, there is a delay for a descent, though it doesn't seem as long. The delay seems most pronounced when I first start a climb, as opposed to apparent increases in grade during a climb. It would make sense that if the Garmin is showing a 0 percent gain while I am clearly climbing, the net gain will be off. I ride with an Apple watch and the net elevation gain between the Garmin and the watch are considerably different, though I think the watch is the least accurate. (I think/hope that the watch is a software glitch that will be cleared up with the latest update).

  12. #12
    Join Date
    Sep 2013
    Location
    Hillsdale, NY
    Posts
    2,275
    Post Thanks / Like
    Mentioned
    3 Post(s)
    Tagged
    0 Thread(s)

    Default Re: Garmin accuracy - grade on climbs

    Quote Originally Posted by thollandpe View Post

    I use Strava on my phone, the dog wears a Garmin wristwatch on her collar.
    Is your dog on Strava?

    SPP
    My name is Peter Miller.

  13. #13
    Join Date
    Apr 2015
    Location
    Mont Tremblant, Quebec & UES, NYC
    Posts
    960
    Post Thanks / Like
    Mentioned
    5 Post(s)
    Tagged
    0 Thread(s)

    Default Re: Garmin accuracy - grade on climbs

    And another thing that may have more to do with consistency than accuracy although it can’t hurt accuracy that I learned when taking one of my Garmins from a place I live where I start at sea level to one where I start well above sea level...there is a way on some of the units in settings to put your elevation into it. If you stay in one place for your rides then you just have to enter it once. I figure if nothing else it will give the machine a reference point to adjust off of while I ride.
    « If I knew what I was doing, I’d be doing it right now »

    -Jon Mandel

  14. #14
    Join Date
    Jan 2013
    Posts
    1,134
    Post Thanks / Like
    Mentioned
    1 Post(s)
    Tagged
    0 Thread(s)

    Default Re: Garmin accuracy - grade on climbs

    Quote Originally Posted by SlowPokePete View Post
    Is your dog on Strava?

    SPP
    a couple weeks back I was joking with my GF that we should invest in a social media fitness platform for pets. Call it something like ROMP. Instead of kudos and KOMs we'd have good boy points and balls. The aim of the game is to collect all the balls and turn them in for treats and toys.

  15. #15
    Join Date
    Jan 2013
    Location
    Philadelphia
    Posts
    120
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Re: Garmin accuracy - grade on climbs

    I like ROMP idea..

    anyway, I always wondered why Garmin "prefers" barometric altimeter when you have GPS, shouldn't that give you all the data you need for true altitude? Maybe I'm missing something, but I assume these data should be part of the GPS anyway like a topo map. I don't really worry about it though, assume it's not for betting purposes, just for entertainment

Similar Threads

  1. @brofaxx shredding Eldridge grade.
    By admin in forum #Random Social Media Fetch
    Replies: 0
    Last Post: 08-11-2019, 05:20 AM
  2. When a torque wrench starts to lose accuracy...
    By Chik in forum VSalon Pro-Bicycle Tech
    Replies: 18
    Last Post: 05-18-2019, 04:27 AM
  3. frame design/build accuracy
    By bryand in forum General Discussion
    Replies: 32
    Last Post: 06-25-2012, 08:46 AM
  4. Mt. Tam -- Old Railroad Grade
    By Watoni in forum General Discussion
    Replies: 23
    Last Post: 03-13-2011, 07:50 PM
  5. Ergo lever up grade
    By Martin T in forum General Discussion
    Replies: 0
    Last Post: 03-02-2010, 11:05 PM

Bookmarks

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •