Just completed a short but lumpy ride on the Dairi pak challenge this weekend. Only 22 miles but the published route advised 5220 ft of altitude gain, which for me anyway was fairly stiff in such.a short distance.

Long story short, Strava came up with 3220 ft of altitude gain.

Conclusions, I am much less “nails” than I thought I was (which was not very “nails” at all) or one or other piece of kit (Garmin or I-phone/Strava) is rubbish. 2000 ft out is a poor margin of error

Thoughts please…..

Well your garmin if a 500 or better has a barometric altimeter

Yur phone just relys on the mapped data overlay.

What valuedid the garmin give ?

I’d say the published info is out.

3220ft is pretty hilly but much more realistic than 5220ft in 22 miles.

Rough rule of thumb I use is 1000ft altitude gain per 10 miles which generally works well unless you’re specifically seeking out every hill or you live in East Anglia.

5220 ft. Not my Garmin but now I look at it it was actually Garmin Fit (iPhone), 1.4.0.0. Whatever that is.

Given that both are on the phone maybe they’re as bad as each other ?

FB conversation after Ronde Van Calderdale (about 3000m of climbing so big enough to highlight differences) was that Garmins seemed to give higher (but consistent with each other) elevation gains than other devices, be they other brands of dedicated GPS, or smart phones running apps.

if you have auto pause turned on and at times you go very slow lets say when going up hill that can affect your total miles recorded on the garmin

Phone’s generally over-estimate especially in hilly areas where sometimes the overlay gets confused if you’re on a ridge or climbing it. I ran Strava on my phone and my Garmin 800 simultaneously for a while and the phone was always 10%-15% over. Although the Garmin wasn’t perfect either (recorded about 3% diff over the same route).

With auto pause on, my GPS seems to be about 10% less k’s as well as my phone. It does auto pause on very steep climbs tho which is a bit depressing when you’re blowing it up out of you ar£s as Andy W says.

I once had all this properly explained by Andy @ Tracklogs but forgot most of it!

3500 feet of climbing in 22 miles? That’s not hilly 😉

Friend and I were discussing this yesterday. Both 800s we had were different… They’re obviously affected by pressure so as I understand it the weather can alter the readings. I assumed that mapped data should be more accurate as it’s correctly surveyed data. But this too depends on how your device is tracking the land and its subsequent conversion in strava.

GPS is not as good at determining height* as it is at determining position around the Earth’s surface. I don’t know if the Garmin in question ONLY uses it’s altimeter, but this may a factor. Having said that, that there is a fairly hefty difference in those figures which this may not entirely explain!

*GPS uses triangulation, and while the sats are spaced out well, the “Height” of the triangle they are mapping changes by VERY small margins when you climb a hill or two.

It does auto pause on very steep climbs tho which is a bit depressing when you’re blowing it up out of you ar£s

Why not just change the speed at which auto pause kicks in to suit your riding?

I get a significant differance between run keeper and strava over the same route, in the same phone recording. Not sure why but Strava is often 40% less altitude.

Strava does the same with me. Always 20-30% less altitude than my Satmap has measured.

There’s a really old thread about this somewhere. I put an identical Garmin file in to (iirc) Strava, Everytrail and Endomondo to see what they do. In short, (iirc)
– Everytrail just added up all the small + an – gains/losses in elevation (so the total height of a flat road is not zero, due to GPS/Alt errors).
– Endomondo did some kind of smoothing (so it makes you total climb/descent look less than what might be published)
– Strava conjured up extra height gain/loss that was not even in the file (no clue why or how), but I think in general it does much the same as Everytrail with little smoothing. Nuked my Strava a/c so I can’t easily double check this.

I’d normally ignore published altitude gain for a trip, because you have no idea how it was recorded. iPhone? GPS+Barometric? Smoothed out data? All the main dips/valleys and summit altitudes read directly from an OS map?

It also goes a bit asymmetric if you have lots of data points going uphill (slowly), and fewer going downhill (faster), especially with lots of switchbacks, unless you’re lucky and teh GPS records a datapoint right at each apex.

Did Helvellyn a few weeks ago….

Bryton Rider 50 said 5500ft

When uploaded from the Bryton device to Strava on computer it said 5000ft

When viewing my Strava iPhone app it’s 4600ft

??????

Looking at the Strava records of some guys in my club that did the May Flyer in Surrey on Sunday I see vast differences – one using an 800 got 1496m ascent whilst another using an 810 got 2659m. I reckon the course looks closer to what the 800 said.

Hmm, yesterday’s fell run over Mam Tor and Kinder:

Route as planned/mapped on Memory-Map – c830m ascent/descent
Garmin (Forerunner 610) – c750m
Strava – c 650m

This morning my legs certainly feel like it was the MM figure…

I’ve noticed different results between 2 GPS apps on the same phone running at the same time.

This was between maverick & strava.
Keep meaning to fire up a third or fourth & see if there is any agreement.

I did a 24 mile ride on the Quantocks last weekend. At the end of the ride my phone had recorded over 5000 feet of climbing but when I saved the ride and uploaded it to strava it changed down to 3500 feet which seems more likely.

I just assumed that it corrects the phones data to the actual elevation when you upload it? Seems to do it with the mileage too. I’ve only just started using strava as sports tracker pro seems to take ages to pick up a GPS signal these days.

Seems that Strava underestimates. Which I’d rather have in all honesty. I hav no wish at all to book onto a challenge that a GPS system says has an amount of climbing I can do, and then die a nasty death up some ridiculously lumpy course.

I’d rather be pleasantly surprised.

Strava sometimes takes a while to calculate height accumulated. My last ride came up at 1500m climbed initially and then went up to 2500m when I checked back later. Noticed it a few times.

Its weird, did the Fred Whitton on Sunday, the website says it should be 112 miles and 13000 feet, my Bryton got 110 miles and 11000 feet, my Dad’s Garmin 50 got 109 miles and 9000 feet. No idea why the differences are so much.

Sounds like nothing much has the altitude thing sorted ! Mobile altimeter stuff has been around for ages. You’d think they’d have of sorted by now.