Garmin 405 not measuring correctly?

Please feel free to guess and speculate and bet all you want. With some luck, you will be right some of the time.Any kind of error-correction (e.g. real-time filter) must be based on certain assumptions. If these assumptions do not hold true (which seems to be Brian's case on the track), then the result is typically not "almost certainly better".Polar uses a simple, standard technique real-time filter to smooth noisy speed data, yet this causes significant errors, when the speed changes are real, such as in fartleks and repetitions.Dynastream (footpod developers) also has some smart people doing research, and some complex curve fitting, but here you presume that these smart people are getting it wrong -- based on what, uninformed speculation? Dynastream researchers did show much better results, under more variety of conditions, with a calibrated acceleromoter than a stride rate calibrated pedometer.Just because simple, standard techniques exist, doesn't mean they were applied, or applied correctly. Maybe Garmin's marketing people got in the way of Garmin's smart research people. See, your idle speculation can go both ways. What makes Garmin immune from the pitfalls you speculate plagues everyone else?I've given you real information about Polar and Dynastream, and can give you lots more. Do you have any real information how Garmin connects the dots, or are you content to just go back and forth trading speculation?

Friends, just wondering, can you tell me how many angels can dance on the tip of the typical pin? Thanks is advance.

scotth wrote:

Friends, just wondering, can you tell me how many angels can dance on the tip of the typical pin? Thanks is advance.

If you use the Garmin 405, the answer 2 because it can't pick up the curves of the pin.

But,

If you use a Polar, the answer is 17.

:)

I would need more information before I can answer your question.

scotth wrote:

Friends, just wondering, can you tell me how many angels can dance on the tip of the typical pin? Thanks is advance.

Mr. Ray wrote:

I would need more information before I can answer your question.

Ah yes, I forgot, I'm thinking of the part of the pin you push on to stick it into stuff. Not the sharp, pointy end. I already know how many angels can dance on the pointy end.

Yes, I'm speculating what Garmin does to calculate distance, because contrary to what Brian emphatically states in his posts, none of us KNOWS what Garmin is doing.

Any kind of error-correction (e.g. real-time filter) must be based on certain assumptions. If these assumptions do not hold true (which seems to be Brian's case on the track), then the result is typically not "almost certainly better".

How about the assumption that people don't run down the road (or around curves) in a zig-zag path? I think you're pretty safe with that one 99% of the time. As far as Brian's results from the track, there are others that claim Garmins are nearly perfect on the track. Which only goes to prove the old adage that the plural of anecdote is not data.

Dynastream (footpod developers) also has some smart people doing research, and some complex curve fitting, but here you presume that these smart people are getting it wrong -- based on what, uninformed speculation? Dynastream researchers did show much better results, under more variety of conditions, with a calibrated acceleromoter than a stride rate calibrated pedometer.

My "uninformed speculation" comes from experience with other testing with accelerometers, where displacements computed from integrated accelerations are always considered suspect. If the folks at Dynastream have figured out a way to get around all the fundamental issues with trying to do this, I'd be interested in seeing it. Please post a description of what they did. I'd be especially interested in seeing how they "calibrated" the accelerometer.

Hey guys, just wondering,

for the forerunner 205, what does smoothing do?

For now, mine is on the \"more\" setting. Would putting it on the \"most\" setting make calculating my pace any more accurate?

Okay you all, I did a little experiment today and had some interesting results.

I taped two Garmin 205's and one Garmin 305 to a board and had them placed 3" apart. I placed the board on the dash of my John Deere Gator with an unobstructed view of the sky. The sky was partly cloudy.

I drove the gator around an 8K course at 8:30 mile pace to ensure numerous recordings going around the gentle turns of the course. I place a cone every 800 meters according to each individual device. I split every 800m so that i could be able to identify EXACTLY (according to the device) when I hit 1600 meters. If you just wait until the device shows 1600 you could be as much as 10 meters off and it would still show 1600. I repeated this process to see how the 6 measurements would compare. The course that I measured does not have a tree or building anywhere near it and I had 100% view of the southern sky.

My results: 6 completely different measurements that were as far off as 60 meters at the end of the 8K course. The 305 put me the furthest away from the finish line and two 205's were usually within 8-13 meters of eachother. The 305 was clearly getting different readings than the 205's.

Who knows which device is more accurate, but just thought I would share this with you.

5 x 8:30 = 42:30 x 6 = 255:00/60 = 4.25 hours

I'd give my left one to have 4.25 hours to kill for a letsrun experiment!

When my computer shows all the "dots" on the oval of the track, does that mean that I am getting accurate distances from my 305? (In other words, the satellite is not cutting the tangents of the curve.)

William J. Clinton wrote:

5 x 8:30 = 42:30 x 6 = 255:00/60 = 4.25 hours

I'd give my left one to have 4.25 hours to kill for a letsrun experiment!

Hey Billy,

I hope you are using plenty of time to study math and/or reading. :) Re-read my post. I said I used 3 Garmins at once and drove the course twice. It took about 1.5 hours. Lastly, it wasn't for an experiment, I coach XC and needed to check our course for distance. NOW TO REALLY BLOW YOUR MIND. I spent almost the entire day today measuring the 8K and the 5K courses with a 100 meter steel tape measure. Now that took a lot of time, but I am really glad I did it. All three Garmins were way off!

garmin doesnt count uphill and downhill.....like a hill A to B is longer than a straight path A to B

Just for fun wrote:

Hey Billy,

I hope you are using plenty of time to study math and/or reading. :) Re-read my post. I said I used 3 Garmins at once and drove the course twice. It took about 1.5 hours. Lastly, it wasn't for an experiment, I coach XC and needed to check our course for distance. NOW TO REALLY BLOW YOUR MIND. I spent almost the entire day today measuring the 8K and the 5K courses with a 100 meter steel tape measure. Now that took a lot of time, but I am really glad I did it. All three Garmins were way off!

"Twice" was never mentioned in your post; however, "6 different measurements" was mentioned twice, so I assumed (silly me!) you did six different measurements with 3 different instruments.

Reading comprehension, as well as clarity in writing, are what makes this website so much fun!

[quotJust for fun wrote:

My results: 6 completely different measurements...

two 205's were usually within 8-13 meters of each other.

You consider measurements within 8-13 meters on an 8000 meter course completely different?

William J. Clinton wrote:

Just for fun wrote:

Hey Billy,

I hope you are using plenty of time to study math and/or reading. :) Re-read my post. I said I used 3 Garmins at once and drove the course twice. It took about 1.5 hours. Lastly, it wasn't for an experiment, I coach XC and needed to check our course for distance. NOW TO REALLY BLOW YOUR MIND. I spent almost the entire day today measuring the 8K and the 5K courses with a 100 meter steel tape measure. Now that took a lot of time, but I am really glad I did it. All three Garmins were way off!

"Twice" was never mentioned in your post; however, "6 different measurements" was mentioned twice, so I assumed (silly me!) you did six different measurements with 3 different instruments.

Reading comprehension, as well as clarity in writing, are what makes this website so much fun!

I wrote:

"I taped two Garmin 205's and one Garmin 305 to a board and had them placed 3" apart. I placed the board on the dash of my John Deere Gator with an unobstructed view of the sky. The sky was partly cloudy.

I drove the gator around an 8K course at 8:30 mile pace to ensure numerous recordings going around the gentle turns of the course. I place a cone every 800 meters according to each individual device. I split every 800m so that i could be able to identify EXACTLY (according to the device) when I hit 1600 meters. If you just wait until the device shows 1600 you could be as much as 10 meters off and it would still show 1600. I repeated this process to see how the 6 measurements would compare. The course that I measured does not have a tree or building anywhere near it and I had 100% view of the southern sky."

Hmmm??? Two 205's & one 305 taped to a board and attached to a john deere gator. Drove the course and repeated the process.

Sounds pretty clear to me.

dukerdog wrote:

[quotJust for fun wrote:

My results: 6 completely different measurements...

two 205's were usually within 8-13 meters of each other.

You consider measurements within 8-13 meters on an 8000 meter course completely different?

By definition? Yes! They are different.

I anticipated that since they are basically the same product, getting information from the same satellites and were located in the same place in the same conditions, that they would be more accurate or at least more consistant. My tape measurements today really demonstrated how inaccurate they are.

dukerdog wrote:

[quotJust for fun wrote:

My results: 6 completely different measurements...

two 205's were usually within 8-13 meters of each other.

You consider measurements within 8-13 meters on an 8000 meter course completely different?

Just an FYI, the 205's were usually within 8-13 meters of eachother, but not in relation to the actual finish line. Like I said, results were as far as 60 meters off.

Just for fun wrote:

I repeated this process to see how the 6 measurements would compare.

Obviously it wasn't that clear.

You performed two different measurements (driving the Deere on the course) with three different devices, for a total of 6 different results.

You say to-may-toe, I say toe-mah-toe

William J. Clinton wrote:

Just for fun wrote:

I repeated this process to see how the 6 measurements would compare.

Obviously it wasn't that clear.

You performed two different measurements (driving the Deere on the course) with three different devices, for a total of 6 different results.

You say to-may-toe, I say toe-mah-toe

I guess it all depends on what the definition of "is" is. :)

I am very sorry it wasn't clear to you. I honestly felt I accurately decribed the process used. At least in my mind, 3 devices used twice would equate to 6 measurements. I will be more diligent in my future posts.

By the way, I can't believe you are actually supporting Obama.