Garmin 405 not measuring correctly?

"If you walk the course with the GPS on your wrist, your arm swing can throw the measurement off my several meters over an 8k or 10k course."

I don't logically see why this would be. If the GPS takes a figurative picture of your position every second or so and then draws and measures the line between fixed, paused points, arm swing won't mean anything over the course of some time because it is as likely to take the picture on the backswing asn the front swing, thus cancelling it out. If the GPS could actually read continuous motion, I could see why, but otherwise I don't get it. I am willing ot be convinced otehrwise, though.

Isn't that Jones Counter what they used to measure the Men's Marathon Trials Course? I saw i video on it and they took other factors such as temperature expanding the tape. That's a neat device.

It doesn't throw off measurement. That's simply a common misconception of people who do understand how GPS devices work

The arm swing can equate to about 6 ft or nearly 2m. To prove my point, go walk an 8K course with the GPS on the wrist, then go drive the course in a gator or something at a very slow pace and see the difference. I am not going to tell you the difference is mind bogling, but there is a difference.

I do see why speed matters because when you are covering less ground between fixed readings, you are less likely to get wild tangents or over-runs of a given point. Still don't see why armswing matters. My GPS is very, very good on certified courses.

Several of the running paths I run here have markers every .1 miles that were measured using a high quality field GPS unit and my GPS is almost dead with them every time on the linear paths and has discrepency in the same spots every time on the loop paths. My GPS also tend to measure the same course almost the exact same distance every run - within a hundredth or two. If armwsing mattered, wouldn't I get more divergent readings on different runs of the same course?

First, a running arm swing is not nearly as pronounced as a walking arm swing. When running, the wrist rarely crosses the bodies mid-line. When walking, the arm wrist almost always crosses the mid-line.

Again, I didn't say it was a big difference, but there is a difference.

The GPS works by measuring the distance between points. It matters not a whit whether your arm is forward in a swing or backward.

If the GPS logs at back swing and then at front swing, there will be approximately a 6' differential from the actual location of the bodies mid-line. That right there is almost a 2m error. It can simulate the old saying, "three steps forward, on step back" except the step back is really an armswing.

To better illustrate this, draw a line on paper and pick 12 random points on the line. Then measure to each point and based on the flip of a coin, measure 1/4" back or 1/4" forward and see if it measurement equals the actual straight line measurement of that line.

It is not an error at all. The GPS records where the unit is at one instant. Then it records where it was at another instant. IF you have it set for that instant pace reading, yes you might be a little bit off. But as has been mentioned previously, that setting is useless anyway. For any reasonable split like a mile, it will be right on what it was if you'd mounted it on your torso.

You correct that the reading is not an error as it is logging exactly where the device is at that moment, but if the objective is to get an accurate tangent measurement of a course, then those recordings will throw off the actual distance measured. Forget about pace settings. Set the device to record every second (the fastest possible) and walk, run and drive the course. Repeat that 3 times and come back and tell us that there is no difference.

Note: To be able to better see the differences down to meters, you should hit the split button at steel tape measured markers. Otherwise the total will fall between 0-10 meters without showing on the Garmin.

I still don't get it. You have to forget about the fact that the arm is in motion and just reckon with the fixed points that it freezes. If you reckon the middle of the swing axis to be the "actual" point you want to measure each snapshot will be either a bit ahead or a bit behind that point. The tracks that measure short will be cancelled out by the ones that measure long. I still might be wrong, but I am drawing it out on paper and don't see how. If there is some inaccuracy, I don't see how it would be very much - certainly no more than normal inaccuracy.

Brian,

If your armswings are consitant and always one forward then one back, then the difference will be minimal. BUT, do you walk at exactly the same pace with each and every step/arm swing? I seriously doubt it.

GPS-er wrote:

The GPS works by measuring the distance between points. It matters not a whit whether your arm is forward in a swing or backward.

OK, somehow the way you worded it miade it clear for me. How far you travel will be exactly the same as how far the GPS traveled if you measure by freezing fixed points and drawing a line to connect them and armswing certainly doesn't matter. You don't even need to think about forward and backward swings cancelling each other out. Imagine if I carry a ball across a 50' room by going 5 feet and take a reading, then carry it backwards 2 feet and then forward another 3 feet and take a reading and I repeat this process until I am across the room. The backtracking doesn't matter at all - the forward distance the ball traveled will be the same as the length of the room so long as the only readings taken were ones that didn't include the backtracking from the previously frozen point.

"...so long as the only readings taken were ones that didn't include the backtracking from the previously frozen point."

True. But the GPS unit adds up all the readings (+ or -) to give you a total distance. The GPS doesn't know that your hand is behind you.

Dogman wrote:

Brian,

If your armswings are consitant and always one forward then one back, then the difference will be minimal. BUT, do you walk at exactly the same pace with each and every step/arm swing? I seriously doubt it.

Of course he does. For all practical purposes, on a flat course, everyone does. The miniscule variations matter not at all. Unless there is some SYSTEMIC difference (e.g. he slows down dramatically every time his right arm is back and speeds up dramatically every time his right arm is forward) the differences cancel one another out.

If a course is twisting and turning then the measurement will be off but that's inherent in the process of measuring point to point and has nothing to do with armswing.

Measure a course with a Garmin on your wrist and then another with it on your torso and the differences are not going to be anymore tha the inherent margin of error built into the device based on the course and the number and strength of satellite signals.

GPS-er said,

"Of course he does. For all practical purposes, on a flat course, everyone does. The miniscule variations matter not at all. Unless there is some SYSTEMIC difference (e.g. he slows down dramatically every time his right arm is back and speeds up dramatically every time his right arm is forward) the differences cancel one another out."

You are kidding right? You are trying tell me that every second when the GPS takes reading, that your wrist is in the exact same location? No way! No one is that consistant.

Dogman wrote:

GPS-er said,

"Of course he does. For all practical purposes, on a flat course, everyone does. The miniscule variations matter not at all. Unless there is some SYSTEMIC difference (e.g. he slows down dramatically every time his right arm is back and speeds up dramatically every time his right arm is forward) the differences cancel one another out."

You are kidding right? You are trying tell me that every second when the GPS takes reading, that your wrist is in the exact same location? No way! No one is that consistant.

It doesn't have to be. I am more sure of my position now tha I have thought it out. Read my though experiment about measuring the length of a room with fixed points while purposefully retrograding the measuring object. So long as the next reading is ahead of the previous reading, it doesn't matter what happened to the GPS between readings.

No I am trying to tell you that the arms swing backward and forward pretty much the same throughout the run and that it does not matter whether it is forward or backward when the reading is taken because the differences cancel one another out.

Dogman wrote:

If the GPS logs at back swing and then at front swing, there will be approximately a 6' differential from the actual location of the bodies mid-line.

Incidentally, this is incorrect. Next time you go for a run, see how far back your wrist goes compared to your torso on a backswing. The wrist generally comes back to the hip which is right online with the torso.

Then see how far forward it goes on a forward swing. The wrist generally comes up to the chest, a couple of inches from the torso.

The difference is one of inches. Far, far less than the margin of error of the GPS readings. You will never see one of the current Garmin GPS units accuracy fall much below 20 feet for any length of time. That is why the GPS pace reading are far more accurate over the course of a mile than over the course