GPS Distance errors
06-07-21 | 09:59 AM
#76
Senior Member
Joined: Sep 2007
Posts: 15,260
Likes: 1,759
From: Far beyond the pale horizon.
When you say “curvy path”, yes, I agree if there were lots of relatively “small” curves. But as I was measuring rides between 2km and 30+km with straights and curves, I would think that the impact of “curves” over those distances would be somewhat negligible when compared to the circumference of my tyre.
I think you are absolutely correct if I was riding “short” and “curvy” distances every time. But that is not the case.
I think I will start my third list next week and see what the outcome is. Maybe I could compare my third list measurements/distances with measurements on Google Maps or some other map to check if gentle* curves make any difference over “larger” distances.
* gentle road curves vs concentric or tight circles and curves.
I think you are absolutely correct if I was riding “short” and “curvy” distances every time. But that is not the case.
I think I will start my third list next week and see what the outcome is. Maybe I could compare my third list measurements/distances with measurements on Google Maps or some other map to check if gentle* curves make any difference over “larger” distances.
* gentle road curves vs concentric or tight circles and curves.
Last edited by njkayaker; 06-07-21 at 10:02 AM.
06-07-21 | 10:10 AM
#77
Senior Member
Joined: Jul 2015
Posts: 2,673
Likes: 1,387
From: reno, nv
Bikes: yes, i have one
Spelger, you correctly say “if the bike computer has the correct information”. IF!!!
I have always been a little suspicious about using “standard” conversion tables for translating wheel “size” to wheel circumference. (eg 700 x 23C = 25-622 = 2096)
My general knowledge tells me that standard wheel size will only equal distance travelled for any given tyre pressure or even tyre wear. A flatter, less inflated 700x23C will have a smaller circumference compared to a well/highly inflated 700x23C. Tyre wear and even manufacturing tolerances (exact amount of rubber all the way around a tyre) would affect circumference in a similar, albeit to a smaller extent.
Therefore, using 2096 in a BC assumes that your actual wheel on your bike rolls 2096mm for every revolution. Thus, using standard conversion table circumferences is assuming your BC is using the “correct” information. To me, that is not necessarily a good assumption and necessitated some sort of recording and comparison.
I have always been a little suspicious about using “standard” conversion tables for translating wheel “size” to wheel circumference. (eg 700 x 23C = 25-622 = 2096)
My general knowledge tells me that standard wheel size will only equal distance travelled for any given tyre pressure or even tyre wear. A flatter, less inflated 700x23C will have a smaller circumference compared to a well/highly inflated 700x23C. Tyre wear and even manufacturing tolerances (exact amount of rubber all the way around a tyre) would affect circumference in a similar, albeit to a smaller extent.
Therefore, using 2096 in a BC assumes that your actual wheel on your bike rolls 2096mm for every revolution. Thus, using standard conversion table circumferences is assuming your BC is using the “correct” information. To me, that is not necessarily a good assumption and necessitated some sort of recording and comparison.
06-07-21 | 10:59 AM
#78
Senior Member
Joined: Jul 2013
Posts: 1,286
Likes: 864
From: NJ, USA
Bikes: two blacks, a blue and a white.
Well, let's chime in with this angle...
The units being measured (miles, km, feet, whatever) are actually pretty arbitrary. Cubits? why not. The higher order question is "what are you doing with it and what precision is required?"
So this thread gets into precision, which I do find interesting. But IRL, I find that most situations rarely benefit from more than 4 significant digits, usually 3 will do and 2 is often good enough for decision making.
If I had ridden 36 miles, does 36.4 matter?
If I did a season of 3000 miles, would 30 matter? would 3?
I can find you both cases in finance, where the way we need to work sometimes says "need it to the penny," other times we round to the nearest million.
So, what I'm asking this thread is : "what level of precision is _useful_?" and when does a lack of precision become troublesome?
The units being measured (miles, km, feet, whatever) are actually pretty arbitrary. Cubits? why not. The higher order question is "what are you doing with it and what precision is required?"
So this thread gets into precision, which I do find interesting. But IRL, I find that most situations rarely benefit from more than 4 significant digits, usually 3 will do and 2 is often good enough for decision making.
If I had ridden 36 miles, does 36.4 matter?
If I did a season of 3000 miles, would 30 matter? would 3?
I can find you both cases in finance, where the way we need to work sometimes says "need it to the penny," other times we round to the nearest million.
So, what I'm asking this thread is : "what level of precision is _useful_?" and when does a lack of precision become troublesome?
06-07-21 | 12:31 PM
#79
Thread Killer
Joined: Aug 2008
Posts: 13,140
Likes: 2,162
From: Ann Arbor, MI
Bikes: 15 Kinesis Racelight 4S, 76 Motebecane Gran Jubilée, 17 Dedacciai Gladiatore2, 12 Breezer Venturi, 09 Dahon Mariner, 12 Mercier Nano, 95 DeKerf Team SL, 19 Tern Rally, 21 Breezer Doppler Cafe+, 19 T-Lab X3, 91 Serotta CII, 23 3T Strada
Well, let's chime in with this angle...
The units being measured (miles, km, feet, whatever) are actually pretty arbitrary. Cubits? why not. The higher order question is "what are you doing with it and what precision is required?"
So this thread gets into precision, which I do find interesting. But IRL, I find that most situations rarely benefit from more than 4 significant digits, usually 3 will do and 2 is often good enough for decision making.
If I had ridden 36 miles, does 36.4 matter?
If I did a season of 3000 miles, would 30 matter? would 3?
I can find you both cases in finance, where the way we need to work sometimes says "need it to the penny," other times we round to the nearest million.
So, what I'm asking this thread is : "what level of precision is _useful_?" and when does a lack of precision become troublesome?
The units being measured (miles, km, feet, whatever) are actually pretty arbitrary. Cubits? why not. The higher order question is "what are you doing with it and what precision is required?"
So this thread gets into precision, which I do find interesting. But IRL, I find that most situations rarely benefit from more than 4 significant digits, usually 3 will do and 2 is often good enough for decision making.
If I had ridden 36 miles, does 36.4 matter?
If I did a season of 3000 miles, would 30 matter? would 3?
I can find you both cases in finance, where the way we need to work sometimes says "need it to the penny," other times we round to the nearest million.
So, what I'm asking this thread is : "what level of precision is _useful_?" and when does a lack of precision become troublesome?
Only the GPS can record where one has traveled, and as such is a much richer data source than a $20 cyclometer. It’s hard for me to imagine the point of this “accuracy exercise” in 2021; GPS data can be analyzed more easily and in a more consequential manner than the simple data from the cyclometer.
06-07-21 | 01:25 PM
#80
Recreational Road Cyclist
Joined: Sep 2013
Posts: 612
Likes: 305
From: MetroWest, Mass.
Bikes: 1990 Peter Mooney road bike, 1996 Gary Fisher X-Caliber mountain bike
I admire your rigorous approach to tracking down the errors. But I am lazier (by a lot!) than you are.
I have a Cateye Astrale wired bike computer that is C&V-worthy, and I track my rides on a phone using Ride/w/GPS. When they did not agree, I lowered/raised the circumference number in the Cateye until they did agree within a tenth! Took a few rides, but not many, to get them synced.
Which one was correct to begin with? I don't know, but I think the GPS. (There are no cul-de-sacs on my routes.) Why must they agree? Because I like them to agree!
So now, they're either both right, or both wrong, but they agree.
All best to you in Oz.
Last edited by BCDrums; 06-07-21 at 03:14 PM.
06-07-21 | 04:40 PM
#81
Newbie
Joined: Apr 2021
Posts: 25
Likes: 11
From: Melbourne, Australia
Bikes: Cannondale CAAD8 105 2016, Trek DS 8.5 2015, ProAce (1986?)
wkc,
I admire your rigorous approach to tracking down the errors. But I am lazier (by a lot!) than you are.
I have a Cateye Astrale wired bike computer that is C&V-worthy, and I track my rides on a phone using Ride/w/GPS. When they did not agree, I lowered/raised the circumference number in the Cateye until they did agree within a tenth! Took a few rides, but not many, to get them synced.
Which one was correct to begin with? I don't know, but I think the GPS. (There are no cul-de-sacs on my routes.) Why must they agree? Because I like them to agree!
So now, they're either both right, or both wrong, but they agree.
All best to you in Oz.
I admire your rigorous approach to tracking down the errors. But I am lazier (by a lot!) than you are.
I have a Cateye Astrale wired bike computer that is C&V-worthy, and I track my rides on a phone using Ride/w/GPS. When they did not agree, I lowered/raised the circumference number in the Cateye until they did agree within a tenth! Took a few rides, but not many, to get them synced.
Which one was correct to begin with? I don't know, but I think the GPS. (There are no cul-de-sacs on my routes.) Why must they agree? Because I like them to agree!
So now, they're either both right, or both wrong, but they agree.
All best to you in Oz.
And other posters are spot on too re: degree of accuracy. That doesn’t matter so much to me. Consistency more so.
Thank you to all for your input. I’m happy to have found this forum. I love my rides, even if just short and sharp, not the massive ones I see on MMR and Strava.
06-07-21 | 04:45 PM
#82
Newbie
Joined: Apr 2021
Posts: 25
Likes: 11
From: Melbourne, Australia
Bikes: Cannondale CAAD8 105 2016, Trek DS 8.5 2015, ProAce (1986?)
You’re absolutely right on, although another high-order question which could be asked is which info is most important: distance traveled or distance and location?
Only the GPS can record where one has traveled, and as such is a much richer data source than a $20 cyclometer. It’s hard for me to imagine the point of this “accuracy exercise” in 2021; GPS data can be analyzed more easily and in a more consequential manner than the simple data from the cyclometer.
Only the GPS can record where one has traveled, and as such is a much richer data source than a $20 cyclometer. It’s hard for me to imagine the point of this “accuracy exercise” in 2021; GPS data can be analyzed more easily and in a more consequential manner than the simple data from the cyclometer.
06-07-21 | 05:24 PM
#83
Senior Member
Joined: May 2005
Posts: 1,197
Likes: 690
From: northWET washington
Just did a ride which was
Phone apps using gps:
GAIA: 15.25
Mission Control: 15.44
Specialized Display Unit: 15.4
I guess I could join the "lose sleep over it" group or just learn to live with the differences. Elevation tracks differently, too, but gps elevation stats are always a bit hit/miss. Mission Control is always higher.
Phone apps using gps:
GAIA: 15.25
Mission Control: 15.44
Specialized Display Unit: 15.4
I guess I could join the "lose sleep over it" group or just learn to live with the differences. Elevation tracks differently, too, but gps elevation stats are always a bit hit/miss. Mission Control is always higher.
06-07-21 | 09:05 PM
#84
Senior Member
Joined: May 2021
Posts: 3,135
Likes: 2,342
From: San Francisco
Bikes: addict, aethos, creo, vanmoof, sirrus, public ...
Just did a ride which was
Phone apps using gps:
GAIA: 15.25
Mission Control: 15.44
Specialized Display Unit: 15.4
I guess I could join the "lose sleep over it" group or just learn to live with the differences. Elevation tracks differently, too, but gps elevation stats are always a bit hit/miss. Mission Control is always higher.
Phone apps using gps:
GAIA: 15.25
Mission Control: 15.44
Specialized Display Unit: 15.4
I guess I could join the "lose sleep over it" group or just learn to live with the differences. Elevation tracks differently, too, but gps elevation stats are always a bit hit/miss. Mission Control is always higher.
RwGPS: 25.5 miles, mission control, 25.65.
RwGPS: 74.0 miles, mission control 74.02.
a few PSI less air in the first, most recent one explains the difference. inconsequential for sure.
06-07-21 | 09:11 PM
#85
Gruppetto Bob
Joined: Sep 2020
Posts: 11,465
Likes: 11,690
From: Seattle-ish
Bikes: Orbea Orca, Bianchi Infinito & Campione de Mundo
My GPS on a downhill today gave me a top speed of 58.2 MPH where it was more likely around 45. ‘Technology is a wonderous thing.... when it works.
__________________
“A watt saved is a watt earned” 🚴🏻♂️
Not a CAT
“A watt saved is a watt earned” 🚴🏻♂️
06-07-21 | 10:39 PM
#86
Newbie
Joined: Apr 2021
Posts: 25
Likes: 11
From: Melbourne, Australia
Bikes: Cannondale CAAD8 105 2016, Trek DS 8.5 2015, ProAce (1986?)
Let’s talk about phone GPS precision for a minute
I promise you that I did not edit this pic. That was my max speed at the very start of this ride.
06-08-21 | 04:25 AM
#87
Senior Member
Joined: Sep 2007
Posts: 15,260
Likes: 1,759
From: Far beyond the pale horizon.
The odd speed might be due to a difference between an initial inaccurate location and a later accurate location.
Last edited by njkayaker; 06-09-21 at 08:08 AM.
06-08-21 | 04:27 AM
#88
Senior Member
Joined: Sep 2007
Posts: 15,260
Likes: 1,759
From: Far beyond the pale horizon.
06-08-21 | 06:15 AM
#89
Thread Killer
Joined: Aug 2008
Posts: 13,140
Likes: 2,162
From: Ann Arbor, MI
Bikes: 15 Kinesis Racelight 4S, 76 Motebecane Gran Jubilée, 17 Dedacciai Gladiatore2, 12 Breezer Venturi, 09 Dahon Mariner, 12 Mercier Nano, 95 DeKerf Team SL, 19 Tern Rally, 21 Breezer Doppler Cafe+, 19 T-Lab X3, 91 Serotta CII, 23 3T Strada
https://ride.lezyne.com/collections/...macro-easy-gps
It’s fully capable of running Bluetooth sensors, too, so you can hook up a speed sensor for enhanced accuracy if you want it.
06-08-21 | 06:53 AM
#90
Thread Killer
Joined: Aug 2008
Posts: 13,140
Likes: 2,162
From: Ann Arbor, MI
Bikes: 15 Kinesis Racelight 4S, 76 Motebecane Gran Jubilée, 17 Dedacciai Gladiatore2, 12 Breezer Venturi, 09 Dahon Mariner, 12 Mercier Nano, 95 DeKerf Team SL, 19 Tern Rally, 21 Breezer Doppler Cafe+, 19 T-Lab X3, 91 Serotta CII, 23 3T Strada
Just did a ride which was
Phone apps using gps:
GAIA: 15.25
Mission Control: 15.44
Specialized Display Unit: 15.4
I guess I could join the "lose sleep over it" group or just learn to live with the differences. Elevation tracks differently, too, but gps elevation stats are always a bit hit/miss. Mission Control is always higher.
Phone apps using gps:
GAIA: 15.25
Mission Control: 15.44
Specialized Display Unit: 15.4
I guess I could join the "lose sleep over it" group or just learn to live with the differences. Elevation tracks differently, too, but gps elevation stats are always a bit hit/miss. Mission Control is always higher.
06-08-21 | 07:15 AM
#91
Senior Member
Joined: Jun 2015
Posts: 5,786
Likes: 1,800
From: North Central Wisconsin
My god...I disappear for a week and you guys are still babbling about GPS accuracy.
https://www.singletracks.com/mtb-gea...cyclocomputer/
https://www.singletracks.com/mtb-gea...cyclocomputer/
06-08-21 | 10:07 PM
#92
Gruppetto Bob
Joined: Sep 2020
Posts: 11,465
Likes: 11,690
From: Seattle-ish
Bikes: Orbea Orca, Bianchi Infinito & Campione de Mundo
Ayup. My other bikes have them. Strava consistently under reports my average speed by 0.5 MPH with no way to correct which is a bit frustrating after a good ride.
__________________
“A watt saved is a watt earned” 🚴🏻♂️
Not a CAT
“A watt saved is a watt earned” 🚴🏻♂️
06-09-21 | 04:00 AM
#93
Senior Member
Joined: Sep 2007
Posts: 15,260
Likes: 1,759
From: Far beyond the pale horizon.
The Garmins have an option to exclude speed below some limit from the average. This limit can be changed.
Strava is probably doing the same thing.
Given that a common use of Strava is comparing speed, it makes sense to use the same limit for everybody.
While it makes sense to exclude actual zero speed, any nonzero number is kind of arbitrary.
06-09-21 | 05:34 AM
#94
Newbie
Joined: Aug 2018
Posts: 48
Likes: 23
From: Georgia
Bikes: 79 Schwinn & 2017 Fuji Tread 2.0
Most GPS (phone and watch), update (talk to satellite), once every 5-6 seconds.
More accuracy would require a dedicated GPS device.
Another note on longer runs, you are handed off to different satellites every 20 minutes.
I work with surveying grade GPS and we have to set up and wait the 20 minutes to reconfirm our accuracy, +- 0.00'
More accuracy would require a dedicated GPS device.
Another note on longer runs, you are handed off to different satellites every 20 minutes.
I work with surveying grade GPS and we have to set up and wait the 20 minutes to reconfirm our accuracy, +- 0.00'
06-09-21 | 07:15 AM
#95
Senior Member
Joined: Jul 2007
Posts: 8,603
Likes: 3,532
From: South shore, L.I., NY
Bikes: Trek Emonda SL7, Cannondale Topstone, Miyata City Liner, Specialized Chisel, Specialized Epic Evo
06-09-21 | 08:10 AM
#96
Senior Member
Joined: Sep 2007
Posts: 15,260
Likes: 1,759
From: Far beyond the pale horizon.
My god...I disappear for a week and you guys are still babbling about GPS accuracy.
https://www.singletracks.com/mtb-gea...cyclocomputer/
https://www.singletracks.com/mtb-gea...cyclocomputer/
06-09-21 | 08:40 AM
#97
Senior Member
Joined: Sep 2007
Posts: 15,260
Likes: 1,759
From: Far beyond the pale horizon.
The accuracy of reasonably-priced handheld units (dedicated or not) doesn't seem to be much different. The chips might not be that different but the antennas might, in a few cases, be better for a dedicated device.
As far as I understand, the surveying grade GPSs don't really work while moving. It takes a while for any GPS device to "warm up".
Last edited by njkayaker; 06-09-21 at 08:48 AM.
06-09-21 | 10:31 AM
#98
Senior Member
Joined: Jun 2015
Posts: 5,786
Likes: 1,800
From: North Central Wisconsin