Garmin Edge 500/800 Gradient Accuracy

martincashman
martincashman Posts: 116
edited September 2011 in Road buying advice
Anyone know or have experience with the gradient feature on the Edge 500 and 800. I currently have a Sigma ROX9.0 and gradient seems fairly accurate (gives same gradient in same places over time) also seems to respond quickly to gradient change.

I like the gradient feature as I know how hard I can push for each step up to 10%, so I can easily find my comfort zone on an unfamiliar hill by seeing what gradient I'm on.
MTB HardTail: GT Aggressor XC2 '09
Road Summer(s): Kuota Kharma '10
Road Winter(w): Carrera Virtuoso '10
Full Suspension: Trek Fuel Ex 8 '11

http://app.strava.com/athletes/130161

Comments

  • I've an edge 500 and the gradient seems pretty good on the road, much less reliable off road though...
  • amaferanga
    amaferanga Posts: 6,789
    Gradient via GPS is pretty inaccurate in general. Never heard of anyone using gradient to pace themselves up a hill - doesn't really matter how steep it is surely, its the effort you can sustain over different periods of time that counts.
    More problems but still living....
  • rolf_f
    rolf_f Posts: 16,015
    amaferanga wrote:
    Gradient via GPS is pretty inaccurate in general. Never heard of anyone using gradient to pace themselves up a hill - doesn't really matter how steep it is surely, its the effort you can sustain over different periods of time that counts.

    +1. GPS has errors both in the horizontal and vertical planes over short distances. Combine the two and you could be significantly out. A normal computer should be a bit better as the horizontal distance will be consistent but you'll still get errors in the vertical.

    I suppose the other point is that the gradient information is for what you've just cycled, not what you are currently cycling. So unless the gradient is constant, you aren't getting the current info whatever sort of computer you have!
    Faster than a tent.......
  • Altitude is by barometric pressure not GPS difference

    Hence if you look at your altitude on the screen you will often see your house has moved a few hundred feet up or down depending on the weather.

    It is an estimate of what is going on.
  • Well that's the thing, is the Edge using barometric or GPS?

    As for why gradient is important to me, I don't know but have just become accustomed to using it as an indicator on long climbs. If it's say a 5km climb at 4 to 5% then I know I might keep up for 2 to 3 km at 27 to 28kph but I'll explode long before the top, whereas if the group is going 25 or 26kph I know it will be a hard pace but I won't explode before the top. Obviously I use other indicators too and road surface and wind direction play their part but I have always found gradient useful to.
    MTB HardTail: GT Aggressor XC2 '09
    Road Summer(s): Kuota Kharma '10
    Road Winter(w): Carrera Virtuoso '10
    Full Suspension: Trek Fuel Ex 8 '11

    http://app.strava.com/athletes/130161
  • rolf_f
    rolf_f Posts: 16,015
    Hence if you look at your altitude on the screen you will often see your house has moved a few hundred feet up or down depending on the weather.

    That sort of error isn't really relevant in this case as the air pressure isn't likely to vary much over the course of a climb unless you are a very, very slow climber. :lol:

    What is important is relative values - not absolutes.

    The recorded value shouldn't be too bad in a relative sense but still potentially variable enough to upset the calc a little. That said, if you climb a long hill that you reckon has a fairly constant gradient, you can easily assess the accuracy of the GPS unit.
    Faster than a tent.......
  • I've found my 705 a bit hit & miss, on grade and particularly elevation.

    Can't see the 500 or 800 are going to be much better but I'm open to correction. It would be nice if they were and might actually persuade me to change to the 800.
    There's no such thing as too old.
  • Rolf F wrote:
    Hence if you look at your altitude on the screen you will often see your house has moved a few hundred feet up or down depending on the weather.

    That sort of error isn't really relevant in this case as the air pressure isn't likely to vary much over the course of a climb unless you are a very, very slow climber. :lol:

    You'd think that, wouldn't you?

    In actual fact on my Edge 800 the gradient/barometric altimeter is terrible. I don't think my unit is faulty - everything else works perfectly. Get ready for wild downhills when you're climbing and vice versa. If you want an 800 for the gradient function, you will be disappointed.

    I think a mini built in spirit level would be far more useful than what I've currently got.

    ETA: elevation seems to work well which is strange.
    "That's it! You people have stood in my way long enough. I'm going to clown college! " - Homer
  • On a hill that that ive ridden maybe 30 or so times on both my Edge 800 and 305 the gradient has been anything from 9 to 13%, and the gradient readings going downhill seem to always be less than the uphill readings on any given hill, maybe the Edge has more time to calculate a more accurate grade at the slower 5-10 mph speeds going up a steep than the 30-40 mph or more going downhill.
  • sungod
    sungod Posts: 17,382
    gps gradient is just barking, especially on the side of steep things where the hill/mountain/trees/etc. block some of the satellites

    barometer-based gradient can be more accurate, but unless you are going at constant speed on a constant slope, the combination of barometer lag and gps error can result in equally odd results
    my bike - faster than god's and twice as shiny
  • Am I right in thinking that the calcuation of gradient uses both barometric pressure and GPS? (I have an Edge 500). I pretty sure that the altitude change is by barometric pressure and the speed and distance over the ground by GPS (I haven't fitted my speed/cadence sensor yet). Using some basic trigonometry, the angle of the hypotenuse, ie gradient, can be calculated
    The accuracy of the result can be a bit hit and miss as my 500 struggles with GPS signals where there is significant tree cover over local roads.