BMW K1600 Forum banner
  • Hey Everyone! Vote for the Site Favourite BOTM winner for the year of 2021 HERE!
21 - 40 of 54 Posts

·
Registered
Joined
·
91 Posts
Anyone ever checked either odo or speedo against the check zones on some interstate freeways? Not sure how many states have them. We have a few such sections in the West. Gives you 5 miles across which to check accuracy (of either speedo or odo, depending whether you time it or just compare to your trip meter).
 

·
Premium Member
Joined
·
1,244 Posts
The ODO's are not accurate either. After a long trip, my ODO and GPS distances are off by about 3% from each other in the same directions as the speedometer is off.
IDK, they are corrected in cars. Not sure why bikes would be any different.

 

·
Registered
Joined
·
587 Posts
My odometer and GPS do not match. On my most recent long distance trip, my GPS said I went much further than my bike did. I have not matched the speed and odometer against something like mile markers etc. on this bike However, I did match my last motorcycle with mile markers over a significant distance and found that both the odometer and the bike were off 2.6%. My mile mark count confirmed that my GPS was correct on that motorcycle. That was a Garmin GPS, similar to my BMW, so I tend to believe the Garmin is correct. I'll count it out the next long ride I take.

Sent from my Pixel 3a using Tapatalk
 

·
Premium Member
Joined
·
1,244 Posts
I'm not surprised they don't match. Which one is more accurate is debatable. The GPS does calculations every second or so. So it is segmented. The odometer is contiguous. Altitude variation also plays a role (not even sure if it's part of the distance calc) as does how good of a signal and how many satellites the GPS is locked onto. Lots of places for error.
 

·
Registered
Joined
·
587 Posts
I'm not surprised they don't match. Which one is more accurate is debatable. The GPS does calculations every second or so. So it is segmented. The odometer is contiguous. Altitude variation also plays a role (not even sure if it's part of the distance calc) as does how good of a signal and how many satellites the GPS is locked onto. Lots of places for error.
On my 2009 Harley Davidson Ultra Classic, I watched the mile markers for really long time on a couple different trips. The GPS was dead accurate. The odometer / speedometer was not. I had that ultra classic for nearly 90,000 mi, so, I did that test numerous times over many long trips. I've not done that on the BMW yet, but the results are very similar to what I found on that ultra classic.

Sent from my Pixel 3a using Tapatalk
 

·
Premium Member
Joined
·
1,244 Posts
On my 2009 Harley Davidson Ultra Classic, I watched the mile markers for really long time on a couple different trips. The GPS was dead accurate. The odometer / speedometer was not. I had that ultra classic for nearly 90,000 mi, so, I did that test numerous times over many long trips. I've not done that on the BMW yet, but the results are very similar to what I found on that ultra classic.

Sent from my Pixel 3a using Tapatalk
Not surprised the gps is spot on going down a straight, level road in the open. It would do best there. My Harleys always read high too. The odometer and speedo where the same so I'm not surprised they tracked each other. I worked on HDs all the time and would recal speedos all the time.
 

·
Registered
Joined
·
593 Posts
None of my vehicle speedos ever matched the GPS. I run the GTL central display in digital speedo mode because I can't read the dial accurately or fast enough to make changes when needed. I usually set the cruise by the Garmin and ride manually by the bike speedo. I wish I had checked the top speed on my Garmin before sending it off. I read 123 on the bike after passing a few vehicles in MT. Wonder what the true GPS speed was.
 

·
Registered
Joined
·
556 Posts
Maximum speed on my 2000 K1200 LT via the Garmin StreetPilot 2820 was 327 mph.

Even the GPS can be wrong from time to time.
 
  • Like
  • Haha
Reactions: KL51212 and coyotek

·
Registered
Joined
·
24 Posts
When I got my 2013 GTL last year, my speedo and GPS readouts differed by 2-3 MPH depending on my speed. I wanted to know which unit was more accurate so I used a certified LIDAR unit to determine my accurate speed.

During the test I was on a flat, dry, unobstructed, non-grooved surface. Cruise control was set at 60 MPH based on the speedo. Tires were stock in size and had about 3k miles on them. LIDAR had an unobstructed LOS (line of sight) with <5% angle of deviation. LIDAR operator stood in the same spot for both runs.

During the first run the speedometer registered 59-60 MPH, GPS 62 MPH, and LIDAR 61.8 MPH
We did a second run with the GPS registering 62, LIDAR 61.9 MPH with the speedo going between 59-60 MPH.

A motorcycle is a small LIDAR/RADAR acquisition target when compared to a vehicle or large truck. The bigger the target the more accurate the LIDAR result so even though I was small target my unsurprising conclusion is the GPS is as accurate as I was going to get. Your bike setup and equipment may result in minor different results.
 

·
Registered
K1600 GT SE 2011
Joined
·
21 Posts
As a guess, I would say GPS was good for (instantaneous more or less) speed, but would be out over a distance for distance measurement. GPS doesn't really know if you are riding round the inside / outside or "correct line" through corners etc - only the wheels / gearbox revolving knows that. Speed wise its only working out what time it has taken to get from one specific point to another, on a short timescale.
 

·
Premium Member
Joined
·
421 Posts
GPS doesn't really know if you are riding round the inside / outside or "correct line" through corners etc
Interesting point. I've had the GPS think I was on an adjacent road on occasion too. When I pass through massive construction re-routing, it gets all crazy, like when you're detoured onto the opposite side of the highway, for repaving. Especially, if it's not a simple divider and the opposite side is a reasonable distance away.
 

·
Registered
K1600 GT SE 2011
Joined
·
21 Posts
Domestic (none military) GPS isn't amazingly accurate, well coded SatNav software predicts that you are staying on the same road unless your route takes you off it. Thats why at intersections sometimes the SatNav takes a few seconds to work out you have taken a slip road. Friend of mine who used to be in the Police told me about the pains they had when sticking GPS trackers onto suspect vehicles - they would always have two (or more) following vehicles and they would always split at junctions just incase the GPS was delayed. Our caravan sat in the back yard has a GPS tracker, amazing what it gets up to if I look at its historical trace...
 

·
Scammer Hammer
Joined
·
3,795 Posts
As a guess, I would say GPS was good for (instantaneous more or less) speed, but would be out over a distance for distance measurement. GPS doesn't really know if you are riding round the inside / outside or "correct line" through corners etc - only the wheels / gearbox revolving knows that. Speed wise its only working out what time it has taken to get from one specific point to another, on a short timescale.
While you might think so GPS enabled speedometers/odometers have been a big thing in the hot rod and classic car market for the last few years. I have never heard a report of them being inaccurate.
 

·
Premium Member
Joined
·
1,244 Posts
As a guess, I would say GPS was good for (instantaneous more or less) speed, but would be out over a distance for distance measurement. GPS doesn't really know if you are riding round the inside / outside or "correct line" through corners etc - only the wheels / gearbox revolving knows that. Speed wise its only working out what time it has taken to get from one specific point to another, on a short timescale.
Pretty much what I already said. The GPS knows you went from sample point A to B. It assumes a straight line. Also if you're on a hilly road, it won't fully account for the added travel distance. There is no longer reduced accuracy for civilian gps. That went away years ago.
 

·
Registered
K1600 GT SE 2011
Joined
·
21 Posts
Cars are a lot wider than 'bikes, so the error related to path taken will be smaller. But I bet the GPS odometer isn't as accurate as knowing how many revolutions a fixed circumference has travelled.
 

·
Registered
K1600 GT SE 2011
Joined
·
21 Posts
There is no longer reduced accuracy for civilian gps. That went away years ago.
I hope nobody fires a cruise missile at my caravan then, could take out the local church! :)
 

·
Premium Member
Joined
·
2,675 Posts
I'm not surprised they don't match. Which one is more accurate is debatable. The GPS does calculations every second or so. So it is segmented. The odometer is contiguous. Altitude variation also plays a role (not even sure if it's part of the distance calc) as does how good of a signal and how many satellites the GPS is locked onto. Lots of places for error.
Are you sure about that or are you assuming that based on the fact that the log files only have sample points at certain times. Just because the logs don't capture more frequently doesn't mean the device is only calculating that infrequently.

Also, even at 1 second intervals any error relating to curves is going to be negligible over time. You're not going around tight curves so fast that you're going to cut the corners enough to matter....certainly not at the level to introduce a 2-3% error.
 

·
Premium Member
Joined
·
1,244 Posts
Are you sure about that or are you assuming that based on the fact that the log files only have sample points at certain times. Just because the logs don't capture more frequently doesn't mean the device is only calculating that infrequently.

Also, even at 1 second intervals any error relating to curves is going to be negligible over time. You're not going around tight curves so fast that you're going to cut the corners enough to matter....certainly not at the level to introduce a 2-3% error.
While there are faster updating modules available, these type of consumer units are updating at 1Hz.

I'll wait for you to do the math on the curves and hills. No rush, I'm not holding my breath. ;)
 

·
Premium Member
Joined
·
2,675 Posts
While there are faster updating modules available, these type of consumer units are updating at 1Hz.

I'll wait for you to do the math on the curves and hills. No rush, I'm not holding my breath. ;)
Funny you should ask. I just watched (yesterday) a very interesting video that was about 20 minutes long on the calculation of Pi and how it was done prior to Newton and how Newton changed the whole game and got to far more precision with less work than the prior methods. (Yes, I enjoy watching those kinds of things.)

The errors in calculating the distances in navigating a curve are essentially the same as the old method (over 400 years old) of approximating Pi. Taking a very rudimentary approximation of 12 segments on a full circle, the error between the approximation and real Pi is 1.137%. So, for a 90 degree curve, that amounts to 3 segments. If you take the curve at a speed that takes you 3 seconds from entrance to exit, your error for that curve due to the straight line segment approximations would be no more than 1.137%.

If you divide the full circle into 24 segments (6 segments in a 90 degree turn), the error drops to 0.274%. So, if you take 6 seconds to enter and exit, the distance error caused by sampling every 1 second would be no more than 0.274%.

Once you put in a few straights or long sweepers, the error in distances due to sampling approximations is essentially gone.....maybe 0.01% or less (a guess). The change in the circumference in your tire due to wear from beginning to end of life is on the order of 2% according to some sources. That doesn't even mention the variations caused by different brands of tires (Michelin Road 5, e.g.) or tire pressures. So, your mechanical speedo is potentially going to be off by a lot more than the GPS just due to the physical variations that exist.

The video:

 

·
Registered
Joined
·
90 Posts
Both my 14 and 20 GTLs were off by 4km/h. As the tire wears it does up to 5 km/h…I would believe the GPS…
 
21 - 40 of 54 Posts
Top