...of a few feet
(to finish that last sentence)
...of a few feet
(to finish that last sentence)
Tell you what. To demonstrat the impact of swing:
Go to your local park and attach the GPS the seat on the swing set. For fun, go ahead and sit on it yourself. Start the GPS unit and begin swinging and continue for 15 minutes. Then stop, stop the GPS, and get off the swing. Linearly, you have gone nowhere, but I bet the GPS says you have traveled a significant distance.
So based on your GPS what is the distance of the course? You say the swing doesn't matter?
If you stand still and swing your arms back and forth, a GPS device on your wrist will say you have traveled some distance when obviously you have not. The distance increments it calculates don't "cancel out" based on their direction.
That's not the reason arm swing doesn't matter. The reason arm swing doesn't matter, or matters very little, is because there is no part of your body, including your wrist, that moves backward relative to the ground when you run. If any part of your body ever moves backward relative to the ground, you are a horribly inefficient runner.
Now take that swing set and put it on the back of a truck. Drive the truck exactly 10 miles. Now take the truck back and drive it again those same 10 miles, only this time keeping the GPS stationary on it.
The difference in readings will be less than the margin of error of the GPS, (based on linearity of the course, number of satellites, etc)
You know why? Because for every reading where you were swinging BACK, there will be another reading where you were swinging forward.
I can't explain it any better than that.
GPS-er, Do you not read?
This is what I said earlier,
"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."
My input on all this was in regard to walking a course. Running it is even more inaccurate due to the speed at which you are traveling and the increase in tangent error potential.
Dogman wrote:
GPS-er, Do you not read?
This is what I said earlier,
"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."
And I said there was no difference in the final reading on the GPS. Because there is none.
My input on all this was in regard to walking a course. Running it is even more inaccurate due to the speed at which you are traveling and the increase in tangent error potential.
Now you are describing the problems inherent in measuring the distance of a twisting course with point to point, segmented readings. This is a problem with the measurement process itself, not affected at all by the swing of an arm.
GPS-er wrote:
Now take that swing set and put it on the back of a truck. Drive the truck exactly 10 miles. Now take the truck back and drive it again those same 10 miles, only this time keeping the GPS stationary on it.
The difference in readings will be less than the margin of error of the GPS, (based on linearity of the course, number of satellites, etc)
You know why? Because for every reading where you were swinging BACK, there will be another reading where you were swinging forward.
I can't explain it any better than that.
That all depends on speed. If that truck is traveling .5 mph and the swing is traveling 10mph the over all distance will be off.
You know why? Because of the swing!
I can't explain it any better than that.
Dogman wrote:
That all depends on speed. If that truck is traveling .5 mph and the swing is traveling 10mph the over all distance will be off.
You know why? Because of the swing!
I can't explain it any better than that.
Wrong. Just plain wrong.
Look, you have clearly demonstrated that you don't understand the process. I am done wasting my time trying to help you understand. If you manage to perform this or any of you other silly tests and actually get a difference beyond the margin of error of the device, you come back and let us know, ok?
Until then you are hopeless.
I can actually do this experiment. My wife has a 405 also and I will afix one to my torso somehow and the other on my writs and see if they differ.
I will assert again, though, that you don't even need ot think about forward and backward swings cancelling each other out so long as the GPS has progressed forward between readings from its previous point. It makes no difference what the GPS does between readings. I am ashamed it took me this long to figure it out for sure and spent so much time thinking that the swing mattered in any way that even required cancelling out. Your swingset in a track example also proves it - so long as no reading is taken while the swing is retrograde compared to its last reading, the measurement of the line between readings will equal the line the truck travelled (well, we would have to discount the arc of the swing, which would matter since the line would then have slope).
Where is Kevin Beck (Kemibe) at? He is a math genius. he can settle this. Doesn't David Honea post on here? I think he is anothe genius runner who could settle this.
This is a true statement:
Brian said,
"...so long as no reading is taken while the swing is retrograde compared to its last reading, the measurement of the line between readings will equal the line the truck travelled (well, we would have to discount the arc of the swing, which would matter since the line would then have slope)."
The problem is that a reading does take place while the swing is back. Yes. You can set up an experiment where the swing is in perfect timing with the forward movement AND with the timing of the readings. BUT that is not realistic to expect in a real life spantaneous scenario.
To prove my point, just put the GPS on your wrist, start the device and begin swinging your arms as if walking. Now, walk very very slow to some landmark maybe 50m away. When you get there, stop your GPS. Then, steel tape the actual distance walked and compare it to the GPS. I guarantee there will be difference.
Please report back.
"The problem is that a reading does take place while the swing is back"
Not relative to its position AT THE LAST READING. That is all that is important.
Revisit my previous example. Let's say I am walking a dog down a 100m hallway. Every 5 seconds and I say "heel" and my dog comes right beside me, either just a little ahead or a little behing my feet and sits and I make a mark on the floor, but between those 5 seconds, I let him run all over the place. The dog obviously traveled more distance than me, but when I draw lines between the dots and add them up, it will produce the same distance that I walked so long as my total motion progressed between measurements.
Here is some data:
walked 25'(steel tape) at 2.1 KPH swinging my arms at a normal walking speed rate.
Results:
Garmin = 39' 5" or 12m
Actual distance traveled = 25' or 7.62m
A difference of 4.38m which actually exceed the expected error of 3m by the device.
This is just a sampling of 25 feet. Imagine how far off it could be over 5 or 10 miles.
Dogman wrote:
To prove my point, just put the GPS on your wrist, start the device and begin swinging your arms as if walking. Now, walk very very slow to some landmark maybe 50m away. When you get there, stop your GPS. Then, steel tape the actual distance walked and compare it to the GPS. I guarantee there will be difference.
Oh Christ. So now you are trying to create a test which intentionally skews the results?
Who does this and uses a Garmin for it? Really, if I swing it on the end of a 10 foot rope and walk at the pace of a 90 year old man, I'm sure I create a very interesting track as well. But that's not how it is used. In any normal use, every reading of the Garmin is going to be forward of the previous one.
Try a more realistic test. One the mimics how the device is used. Walk normally. Swing your arms normally. And try to pick a course that is significantly longer than the accuracy than the device itself. 50 meters isn't much more than twice the accuracy of a 405 at its most accurate.
You will probably still get a difference. It will be less than the margin of error of the GPS - which is to say, statistically, zero. You'll need to pick a much longer walk to get a meaningful reading.
Now try RUNNING the course. You know, the way that people actually use the Garmin?
Dogman wrote:
Here is some data:
walked 25'(steel tape) at 2.1 KPH swinging my arms at a normal walking speed rate.
Results:
Garmin = 39' 5" or 12m
Actual distance traveled = 25' or 7.62m
A difference of 4.38m which actually exceed the expected error of 3m by the device.
This is just a sampling of 25 feet. Imagine how far off it could be over 5 or 10 miles.
I just did this in the street in front of my house so that I could give you real data.
Again, the key is the speed that you are walking and the speed of the swing. Just like in my swing on the truck example. I walked very very slowly, but swung my arms at a normal rate. The result was a major difference in distance.
"Now try RUNNING the course. You know, the way that people actually use the Garmin?"
That was not the point of this thread. It was about accurate measurements. That is what I was addressing.
You said the swing didn't matter and would not affect results. I said it did matter and would affect the results and then I proved it.
End of story.
You started off with this:
"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"
And then proceeded to a test in which you walked only 50 meters with an intentionally slow speed and an intentionally exaggerated arm swing.
You proved you could intentionally skew the readout on a Garmin over 50 meters allright. Congratulations! No doubt that will be of tremendous use to the rest of us here.
"A difference of 4.38m which actually exceed the expected error of 3m by the device."
This might be the problem. The error of the dvice doesn't mean error when measuring a distance (what distance would it mean?). It means that the GPS will be within 3' of where you on the earth at any given single reading.
The inaccuracy of the GPS evens out over longer distances and is exacerbated on short distances. I have seen this when doing fartleks, as I reported in an earlier post. I did a workout of 15 x 3 min hard/90 easy last week and it always overestimated my pace on the recoveries (no way was I recovering at 6:50 pace). If you see my explanation of what they mean by 3' accuracy you would see why. Because the inaccuracy of any given point is a higher percentage of the total distance for short distances. You can see this yourself with your GPS if you take miles splits for a whole run and then take a very short split at the end of maybe 20 seconds and see how much that one little section differs from the pace of the rest of the miles splits of your run.
this stuff is crazy.
i use my garmin for mile repeats on the track, and it consistently reports 1.00-1.02 miles. a usatf certified half marathon came out at 13.13, and a usatf certified 20k came out at 20.03.
1-2% error (at most) from a machine that i can wear on my wrist, and given the fact that the government does not really allow for any greater precision than that, seems like something i can live with.
if you need better than that, don't buy one. and take something to relax yourself a bit.
GPS-er wrote:
You started off with this:
"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"
And then proceeded to a test in which you walked only 50 meters with an intentionally slow speed and an intentionally exaggerated arm swing.
You proved you could intentionally skew the readout on a Garmin over 50 meters allright. Congratulations! No doubt that will be of tremendous use to the rest of us here.
I did not intentionally skew the readout. I did however demonstrate the validity of my original statement of which I noticed you did not dissagree with now.
You went on to insult me by saying that I didn't understand how GPS works when in fact I understand exactly how GPS works.
My whole point was that everyone runs, walks and swings their arms at different paces and ratios and that as a result, a measurement could be off by a considerable degree. I still stand by my statements.
Dogman wrote:
I did not intentionally skew the readout. I did however demonstrate the validity of my original statement of which I noticed you did not dissagree with now.
Look again. I still disagree with it. Walk normally and your arm will never be behind where it was previously. Certainly not for a runner. You objection proves only that someone who sets out to skew the results can do so.
Can we talk about how crazy hard this Olympic marathon course is?
2024 College Track & Field Open Coaching Positions Discussion
I Was An MIT Educated Neurosurgeon Now I'm Unemployed And Alone In The Mountains How Did I Get Here?
BREAKING: Athing Mu running 800m in Gainesville on Friday at Holloway Pro Classic