Polar S625X

Groper said:

<< Can your pace
really differ by a minute per mile every 5 seconds or so? >>

If you have AD[Hi!]D it can.
I spend a lot of time striding
& prancing, just relieve the monotony
* boredome.

_______
Blog, or dog? Who knows. But if you see my lost pup, please ping me!
<A
HREF="http://journals.aol.com/virginiaz/DreamingofLeonardo">http://journal
s.aol.com/virginiaz/DreamingofLeonardo</A>

I know some you can calibrate it. I would suggest finding a track or a
distance you ABSOLUTELY know is measured off correctly and calibrate it.

"Jan Grobbelaar" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

I finally decided to buy this running computer instead of a GPS
device, as I mostly run in wooded areas where I had found GPS devices
to be very ineffective. I had to work through the instructions in the
manual to figure out how to use it, which took me about 45 frustrating
minutes. I bought this primarily for the "true" speed, distance and
pace measurements it would provide. After my first two runs with the
device, I have the following observations/questions:

1. The distances measured out and back are different (as downloaded
from the watch to my PC). My two data points are:

Run 1; Out: 5.0 miles, Back: 4.9 miles
Run 2; Out: 1.7 miles, Back: 1.6 miles

The distances out and back are physically identical. Why is it
measuring a shorter distance back? Does anyone have any ideas?

2. The pace measurement is very "erratic" in that it could vary wildly
while running at a "steady" pace. I set the pace limit at 08:40 per
mile with a tolerance of 5 seconds, which proved to be pretty useless
as the pace would be all over the place ranging from 08:15 to 09:15
while I am running at a steady pace, as far as I know. I usually
paced by looking at time and distance and considering rate of
breathing, so I think I can run at quite a steady pace. Can your pace
really differ by a minute per mile every 5 seconds or so?

I don't have the Polar version, I have one of the older Nike models, but
I believe the technology is same. I gave up testing its accuracy because
it gave very repeatable results. I tested it on grass, gravel and
asphalt with very close if not identical results. My "gold standard" was
the ups and downs of a parking garage I use to hill training. Always
within one one hundredths of each other.

Is that the best accuracy it gives, tenths of a mile? If it had given
you closer results it might have be just cutting corners.

Does this unit require that you run a standard distance to set it for
maximum accuracy? Did you set the offset and retest?

Is one direction of your trails more uphill than the other, or is there
a cant to the trail, similar to the crown of a road?

Do you tend to run one direction faster than the other way (this really
shouldn't make that much of a difference if you set it for your stride
and pace.

I always found that for some reason I would have to start running for
several yards before starting the watch or linking and thus accuracy
would suffer.

Only way to tell for sure if there is a problem with the unit would be a
nice flat mile or longer and run it both ways, and or take it to a track
and run several laps and see how repeatably accurate it reads.

Maybe someone else can make suggestions for the pace issue, I don't use
it so I can't help there. You might try a goggle for sdm or fitsense.

Jan Grobbelaar said:

I finally decided to buy this running computer instead of a GPS
device, as I mostly run in wooded areas where I had found GPS devices
to be very ineffective. I had to work through the instructions in the
manual to figure out how to use it, which took me about 45 frustrating
minutes. I bought this primarily for the "true" speed, distance and
pace measurements it would provide. After my first two runs with the
device, I have the following observations/questions:

1. The distances measured out and back are different (as downloaded
from the watch to my PC). My two data points are:

Run 1; Out: 5.0 miles, Back: 4.9 miles
Run 2; Out: 1.7 miles, Back: 1.6 miles

The distances out and back are physically identical. Why is it
measuring a shorter distance back? Does anyone have any ideas?

2. The pace measurement is very "erratic" in that it could vary wildly
while running at a "steady" pace. I set the pace limit at 08:40 per
mile with a tolerance of 5 seconds, which proved to be pretty useless
as the pace would be all over the place ranging from 08:15 to 09:15
while I am running at a steady pace, as far as I know. I usually
paced by looking at time and distance and considering rate of
breathing, so I think I can run at quite a steady pace. Can your pace
really differ by a minute per mile every 5 seconds or so?

In article said:

1. The distances measured out and back are different

Quoted message said:

2. The pace measurement is very "erratic" in that it could vary wildly
while running at a "steady" pace. I set the pace limit at 08:40 per
mile with a tolerance of 5 seconds, which proved to be pretty useless

I'm afraid your expectations are way out of line compared with the
capabilities of current technology. Notice how no claims are made as
to pace accuracy, instead they make meaningless claims like 99%
accurate? Now you know why. I rarely run without my GPS, but
I really only take notice of distance and average pace over a lap or
whole run.

Paul

"Anonymous" <[email hidden]> wrote in message news:<[email hidden]>...

Quoted message said:

I know some you can calibrate it. I would suggest finding a track or a
distance you ABSOLUTELY know is measured off correctly and calibrate it.

I don't think calibration is the issue, as the device is measuring
distance and not velocity i.e. it should not be direction dependent -
5 miles should be 5 miles in any direction..

[email hidden] (Bumper) wrote in message news:<[email hidden]>...

Quoted message said:

I don't have the Polar version, I have one of the older Nike models, but
I believe the technology is same. I gave up testing its accuracy because
it gave very repeatable results. I tested it on grass, gravel and
asphalt with very close if not identical results. My "gold standard" was
the ups and downs of a parking garage I use to hill training. Always
within one one hundredths of each other.

Is that the best accuracy it gives, tenths of a mile? If it had given
you closer results it might have be just cutting corners.

I'll have to check the accuracy claims.

Quoted message said:

Does this unit require that you run a standard distance to set it for
maximum accuracy? Did you set the offset and retest?

You can calibrate the distance measurement, but his would not resolve
measuring the same distance in different directions - it should be the
same distance regardless of direction.

Quoted message said:

Is one direction of your trails more uphill than the other, or is there
a cant to the trail, similar to the crown of a road?

Out is more downhill then slightly up, Back is slightly down then
uphill all the way. There is no camber on the trail - flat - ex
railroad bed.

Quoted message said:

Do you tend to run one direction faster than the other way (this really
shouldn't make that much of a difference if you set it for your stride
and pace.

Usually back is 3 minutes slower, but not in these cases.

Quoted message said:

I always found that for some reason I would have to start running for
several yards before starting the watch or linking and thus accuracy
would suffer.

I think I will try a running start next time instead of a standing
start to see if that makes any difference.

Quoted message said:

Only way to tell for sure if there is a problem with the unit would be a
nice flat mile or longer and run it both ways, and or take it to a track
and run several laps and see how repeatably accurate it reads.

Maybe someone else can make suggestions for the pace issue, I don't use
it so I can't help there. You might try a goggle for sdm or fitsense.

Thanks

[email hidden] wrote in message news:<[email hidden]>...

Quoted message said:

In article said:

1. The distances measured out and back are different

Quoted message said:

2. The pace measurement is very "erratic" in that it could vary wildly
while running at a "steady" pace. I set the pace limit at 08:40 per
mile with a tolerance of 5 seconds, which proved to be pretty useless

I'm afraid your expectations are way out of line compared with the
capabilities of current technology. Notice how no claims are made as
to pace accuracy, instead they make meaningless claims like 99%
accurate? Now you know why. I rarely run without my GPS, but
I really only take notice of distance and average pace over a lap or
whole run.

Paul

This new Suunto model has distance/pace capability. I wonder how much
the foot sensor will add to the already high price of this model?
http://www.suunto.com/minisites/t6/index_flash.html

Actually not correct, the following is from the Dynastream web site

http://www.dynastream.com/home/

SpeedMax is the first implementation of Dynastream's patented
technology, customized for the recreational running market. Its
multi-dimensional motion tracking ability delivers a highly accurate and
continuous method of measuring distance and velocity traveled for both
runners and walkers. SpeedMax's functionality allows it to generate
analytical information about the stride such as stride length and foot
lift. Since that initial application, the technology has been
implemented in numerous related applications building on the same core
technology.

If I remember it correctly it measures the deflection of a heated bubble
and translates it into speed and distance, thus it is acceleration
(velocity) if the bubble as your foot takes off and plants.

They invented the technology that is used in the Polar, Nike and
Fitsense systems. There are a couple of pdf's about it on their site.

Jan Grobbelaar said:

I don't think calibration is the issue, as the device is measuring
distance and not velocity i.e. it should not be direction dependent -
5 miles should be 5 miles in any direction..

Again I cannot speak with detailed knowledge of the Polar unit, but I
know that the original Nike units were very sensitive to overpronation
and movement of the pod on the shoe.

Another factor of the accuracy was pace. If you calibrated at a 9:00
pace and then slowed down to a 13 or 14:00 pace or sped up to a 5:00
pace the accuracy suffered. Nike reccomended that you calibrate and use
the unit with one pair of shoes. The newer Nike units, like the
Fitsense, allow for an easier changing of the offset, but you still need
to calibrate for each pair of shoes.

I know that all of this doesn't sound like the ads, but once you have
set everything up and tested and retested the repeatabilty of the system
is quite good, typically 98% or better, so no matter how much you might
be off, it's still going to be more accurate than guestimating.

Jan Grobbelaar said:

[email hidden] (Bumper) wrote in message news:

<[email hidden]>...

Quoted message said:

Quoted message said:

I don't have the Polar version, I have one of the older Nike models, but
I believe the technology is same. I gave up testing its accuracy because
it gave very repeatable results. I tested it on grass, gravel and
asphalt with very close if not identical results. My "gold standard" was
the ups and downs of a parking garage I use to hill training. Always
within one one hundredths of each other.

Is that the best accuracy it gives, tenths of a mile? If it had given
you closer results it might have be just cutting corners.

I'll have to check the accuracy claims.

Quoted message said:

Does this unit require that you run a standard distance to set it for
maximum accuracy? Did you set the offset and retest?

You can calibrate the distance measurement, but his would not resolve
measuring the same distance in different directions - it should be the
same distance regardless of direction.

Quoted message said:

Is one direction of your trails more uphill than the other, or is there
a cant to the trail, similar to the crown of a road?

Out is more downhill then slightly up, Back is slightly down then
uphill all the way. There is no camber on the trail - flat - ex
railroad bed.

Quoted message said:

Do you tend to run one direction faster than the other way (this really
shouldn't make that much of a difference if you set it for your stride
and pace.

Usually back is 3 minutes slower, but not in these cases.

Quoted message said:

I always found that for some reason I would have to start running for
several yards before starting the watch or linking and thus accuracy
would suffer.

I think I will try a running start next time instead of a standing
start to see if that makes any difference.

Quoted message said:

Only way to tell for sure if there is a problem with the unit would be a
nice flat mile or longer and run it both ways, and or take it to a track
and run several laps and see how repeatably accurate it reads.

Maybe someone else can make suggestions for the pace issue, I don't use
it so I can't help there. You might try a goggle for sdm or fitsense.

Thanks

In article said:

Another factor of the accuracy was pace. If you calibrated at a 9:00
pace and then slowed down to a 13 or 14:00 pace or sped up to a 5:00
pace the accuracy suffered.

Don't you find it a little ironic that a device that measures pace
is less accurate if you vary your pace?

Paul

<< Again I cannot speak with detailed knowledge of the Polar unit, but I
know that the original Nike units were very sensitive >>

true.

_______
Blog, or dog? Who knows. But if you see my lost pup, please ping me!
<A
HREF="http://journals.aol.com/virginiaz/DreamingofLeonardo">http://journal
s.aol.com/virginiaz/DreamingofLeonardo</A>

<< If I remember it correctly it measures the deflection of a heated bubble and
translates it into speed and distance, thus it is acceleration (velocity) if
the bubble as your foot takes off and plants. >>

!WARNING!

~winged ~ mice ~

#* be# * #wary *#
Baaattttttttttttttttssssssssssssssss ~~

~ W ~ W ~ W ~ W ~ W ~ W ~ W ~
~ W ~ W ~ W ~ W ~ W ~ W ~ W ~

_______
Blog, or dog? Who knows. But if you see my lost pup, please ping me!
<A
HREF="http://journals.aol.com/virginiaz/DreamingofLeonardo">http://journal
s.aol.com/virginiaz/DreamingofLeonardo</A>

Jan Grobbelaar said:

I finally decided to buy this running computer instead of a GPS
device, as I mostly run in wooded areas where I had found GPS devices
to be very ineffective.

I'm assuming this is trail.

Regarding your GPS, are you carrying that by hand, on belt, or on top a
back pack. FWIW, I've had really crummy readings when I use my GPS by
hand, but fairly reliable readings when mounting it on my camelbak.
(I've never checked satellite configs to see if they happen to be good
on days with cb and bad on hand-carry days, since that could be an issue
also.) I'm just interested in total distance and elevation, not in
instantaneous pace readings since I train by effort. I also think that
expecting great precision from these devices on trails may be pushing
expectations a little.

Quoted message said:

1. The distances measured out and back are different (as downloaded
from the watch to my PC). My two data points are:

Run 1; Out: 5.0 miles, Back: 4.9 miles
Run 2; Out: 1.7 miles, Back: 1.6 miles

The distances out and back are physically identical. Why is it
measuring a shorter distance back? Does anyone have any ideas?

You may have run a slightly different path out/back. Or there may have
been enough variability in the way you ran it to account for the 2-6%
difference - esp. if on trail. I'd be more than happy with that level of
accuracy / repeatability for trail runs, but your needs may differ.

Quoted message said:

2. The pace measurement is very "erratic" in that it could vary wildly
while running at a "steady" pace. I set the pace limit at 08:40 per
mile with a tolerance of 5 seconds, which proved to be pretty useless
as the pace would be all over the place ranging from 08:15 to 09:15
while I am running at a steady pace, as far as I know. I usually
paced by looking at time and distance and considering rate of
breathing, so I think I can run at quite a steady pace. Can your pace
really differ by a minute per mile every 5 seconds or so?

Probably measurement error, same as the distance. I'm not sure if any of
the devices have had reliable instantaneous pace estimates, but I can't
remember all the reviews over last couple years. Depending on the type
of trail you're on, I can easily see 1 min/mile variation. If you're
running by effort (breathing) then you're pace on hills would be likely
to vary.

In case anyone is wondering my 625 came last week - about the time of
the first snow that looks like it's going to stay, but it may melt with
this warm weather again. The annoying snow - makes things slippery and
icy, but not enough to snowshoe on other than in front of my house the
day it fell. So far I'm finding the IR interface a little persnippity,
but that might be some quirk of my system or operator.

I used it (uncalibrated) for a short walk on Sunday and for Tues night
run in the snow. Total recorded distance came out to about 3.4 miles on
a route I've been calling 3 miles (bike measurements). The calibration
is just a multplication factor, so it's just a matter of seeing if it
measures things consistently, rather than accurately, for me at the
moment. It looks like it might have as much as 10% error in distance
measurements - BUT, this was in sloppy snow on trail, so I was
pleasantly surprised. I haven't had a chance to line up it's distances
against my bike distances by various landmarks (partly since I wasn't
consistent in my run - passed by some landmarks in the grayness of the
day). I'm going to calibrate, when I can (maybe not til spring depending
on snow), on the gravel edge of a flat section-line road (1 mile for
sure), then see how that plays out on other routes.

As for the foot pod, it stays in place well, but it would be more
comfortable for me if I had thicker shoe tongues (Trespasses are fairly
thin) and longer feet (I think when I start doing hills it may be a
little restrictive, but it might feel totally differently when not in snow).

I think the altitude sensor may work reasonably well, but I've only used
it on essentially flat terrain (which means it the trace was pretty
flat, but it did seem to detect a few 10- to 20-ft hills).

The software is a big improvement over the PCCoach of the 410, although
it doesn't look like it has the power of a database in terms of being
able to search for things, but I haven't played with it too much yet.
Too many ROL things at the moment.

Dot

--
"So many people get stuck in the routine of life that their dreams waste
away. This is about living the dream." - Cave Dog

I did a run today (6 miles) with a running start and the unit measured
exactly the same distance out and back. I guess that the algorithms
they use to integrate acceleration to obtain velocity (and hence
speed) may assume that the initial and final velocities are equal.

Anyway, as an overpronator I think there is merit in doing the
calibration, but I still am skeptical about how that would resolve
measuring different distances over the same distance in different
directions. I wish I could lay may hands on an exact description of
the technology and algorithms used...

[email hidden] (Bumper) wrote in message news:<[email hidden]>...

Quoted message said:

Again I cannot speak with detailed knowledge of the Polar unit, but I
know that the original Nike units were very sensitive to overpronation
and movement of the pod on the shoe.

Another factor of the accuracy was pace. If you calibrated at a 9:00
pace and then slowed down to a 13 or 14:00 pace or sped up to a 5:00
pace the accuracy suffered. Nike reccomended that you calibrate and use
the unit with one pair of shoes. The newer Nike units, like the
Fitsense, allow for an easier changing of the offset, but you still need
to calibrate for each pair of shoes.

I know that all of this doesn't sound like the ads, but once you have
set everything up and tested and retested the repeatabilty of the system
is quite good, typically 98% or better, so no matter how much you might
be off, it's still going to be more accurate than guestimating.

Jan Grobbelaar said:

[email hidden] (Bumper) wrote in message news:

<[email hidden]>...

Quoted message said:

Quoted message said:

I don't have the Polar version, I have one of the older Nike models, but
I believe the technology is same. I gave up testing its accuracy because
it gave very repeatable results. I tested it on grass, gravel and
asphalt with very close if not identical results. My "gold standard" was
the ups and downs of a parking garage I use to hill training. Always
within one one hundredths of each other.

Is that the best accuracy it gives, tenths of a mile? If it had given
you closer results it might have be just cutting corners.

I'll have to check the accuracy claims.

Quoted message said:

Does this unit require that you run a standard distance to set it for
maximum accuracy? Did you set the offset and retest?

You can calibrate the distance measurement, but his would not resolve
measuring the same distance in different directions - it should be the
same distance regardless of direction.

Quoted message said:

Is one direction of your trails more uphill than the other, or is there
a cant to the trail, similar to the crown of a road?

Out is more downhill then slightly up, Back is slightly down then
uphill all the way. There is no camber on the trail - flat - ex
railroad bed.

Quoted message said:

Do you tend to run one direction faster than the other way (this really
shouldn't make that much of a difference if you set it for your stride
and pace.

Usually back is 3 minutes slower, but not in these cases.

Quoted message said:

I always found that for some reason I would have to start running for
several yards before starting the watch or linking and thus accuracy
would suffer.

I think I will try a running start next time instead of a standing
start to see if that makes any difference.

Quoted message said:

Only way to tell for sure if there is a problem with the unit would be a
nice flat mile or longer and run it both ways, and or take it to a track
and run several laps and see how repeatably accurate it reads.

Maybe someone else can make suggestions for the pace issue, I don't use
it so I can't help there. You might try a goggle for sdm or fitsense.

Thanks

Did you read the white paper on the dynastream site? It gives a hint of
how it's done.

2. SpeedMax technology utilizes accelerometers housed inside a foot pod,
to measure the acceleration of the pod more than 1000 times per second.
Making use of kinematics, the SpeedMax technology accurately determines
the position of the pod in space based on the measured acceleration.
The foundation of the SpeedMax technology is that the foot continually
accelerates and decelerates with each stride. With the pod attached to
the foot, the major phases of gait can be extracted from the sensor
signals using advanced DSP (digital signal processing) techniques (as
shown in Figure 1). SpeedMax measures the accelerations throughout each
stride, computes foot angles, and then determines the resultant
acceleration in the direction of travel. These data are integrated for
each and every stride to provide the running speed and distance in real
time to the user.

3. What makes SpeedMax technology particularly unique is its user
independence, accuracy and ease of use. Right out of the box, the
accuracy of the distance computation using the SpeedMax algorithm during
running is typically 97%.
As mentioned, one of the inherent advantages of SpeedMax technology is
it actually tracks and recreates the motion of the foot through its
stride. By recreating each and every stride, it identifies the length of
each stride to determine the distance a runner travels in the forward
direction.
SpeedMax technology measures accelerations in the forward and vertical
directions. In general, the foot pod will never be perfectly aligned to
the direction of the running motion. As a result, lateral rolling
motions of the foot during running may result in small and repeatable
measurement inaccuracies for some users. Each runner has a different
stride. If a user pronates or supinates significantly, or wears
orthotics, he may want to calibrate the foot pod in order to get the
best possible distance data. SpeedMax accuracy after calibration is
approximately

A good place to get more info is to look up the patents listed for the
various units. I know fitsense says they have several additional patents
in addition the the dynastream original ones.

Jan Grobbelaar said:

I did a run today (6 miles) with a running start and the unit measured
exactly the same distance out and back. I guess that the algorithms
they use to integrate acceleration to obtain velocity (and hence
speed) may assume that the initial and final velocities are equal.

Anyway, as an overpronator I think there is merit in doing the
calibration, but I still am skeptical about how that would resolve
measuring different distances over the same distance in different
directions. I wish I could lay may hands on an exact description of
the technology and algorithms used...

[email hidden] (Bumper) wrote in message news:

<[email hidden]>...

Quoted message said:

Quoted message said:

Again I cannot speak with detailed knowledge of the Polar unit, but I
know that the original Nike units were very sensitive to overpronation
and movement of the pod on the shoe.

Another factor of the accuracy was pace. If you calibrated at a 9:00
pace and then slowed down to a 13 or 14:00 pace or sped up to a 5:00
pace the accuracy suffered. Nike reccomended that you calibrate and use
the unit with one pair of shoes. The newer Nike units, like the
Fitsense, allow for an easier changing of the offset, but you still need
to calibrate for each pair of shoes.

I know that all of this doesn't sound like the ads, but once you have
set everything up and tested and retested the repeatabilty of the system
is quite good, typically 98% or better, so no matter how much you might
be off, it's still going to be more accurate than guestimating.

Jan Grobbelaar said:

[email hidden] (Bumper) wrote in message news:

<[email hidden]>...

Quoted message said:

> I don't have the Polar version, I have one of the older Nike models, but
> I believe the technology is same. I gave up testing its accuracy because
> it gave very repeatable results. I tested it on grass, gravel and
> asphalt with very close if not identical results. My "gold standard" was
> the ups and downs of a parking garage I use to hill training. Always
> within one one hundredths of each other.
>
> Is that the best accuracy it gives, tenths of a mile? If it had given
> you closer results it might have be just cutting corners.
>

I'll have to check the accuracy claims.

> Does this unit require that you run a standard distance to set it for
> maximum accuracy? Did you set the offset and retest?

You can calibrate the distance measurement, but his would not resolve
measuring the same distance in different directions - it should be the
same distance regardless of direction.

> Is one direction of your trails more uphill than the other, or is there
> a cant to the trail, similar to the crown of a road?
>

Out is more downhill then slightly up, Back is slightly down then
uphill all the way. There is no camber on the trail - flat - ex
railroad bed.

> Do you tend to run one direction faster than the other way (this really
> shouldn't make that much of a difference if you set it for your stride
> and pace.
>

Usually back is 3 minutes slower, but not in these cases.

> I always found that for some reason I would have to start running for
> several yards before starting the watch or linking and thus accuracy
> would suffer.

I think I will try a running start next time instead of a standing
start to see if that makes any difference.

> Only way to tell for sure if there is a problem with the unit would be a
> nice flat mile or longer and run it both ways, and or take it to a track
> and run several laps and see how repeatably accurate it reads.
>
> Maybe someone else can make suggestions for the pace issue, I don't use
> it so I can't help there. You might try a goggle for sdm or fitsense.
>
Thanks

What BS! Will you STFU and just run fer christs sakes...

[email hidden] (Bumper) wrote in message news:<[email hidden]>...

Quoted message said:

Did you read the white paper on the dynastream site? It gives a hint of
how it's done.

2. SpeedMax technology utilizes accelerometers housed inside a foot pod,
to measure the acceleration of the pod more than 1000 times per second.
Making use of kinematics, the SpeedMax technology accurately determines
the position of the pod in space based on the measured acceleration.
The foundation of the SpeedMax technology is that the foot continually
accelerates and decelerates with each stride. With the pod attached to
the foot, the major phases of gait can be extracted from the sensor
signals using advanced DSP (digital signal processing) techniques (as
shown in Figure 1). SpeedMax measures the accelerations throughout each
stride, computes foot angles, and then determines the resultant
acceleration in the direction of travel. These data are integrated for
each and every stride to provide the running speed and distance in real
time to the user.

3. What makes SpeedMax technology particularly unique is its user
independence, accuracy and ease of use. Right out of the box, the
accuracy of the distance computation using the SpeedMax algorithm during
running is typically 97%.
As mentioned, one of the inherent advantages of SpeedMax technology is
it actually tracks and recreates the motion of the foot through its
stride. By recreating each and every stride, it identifies the length of
each stride to determine the distance a runner travels in the forward
direction.
SpeedMax technology measures accelerations in the forward and vertical
directions. In general, the foot pod will never be perfectly aligned to
the direction of the running motion. As a result, lateral rolling
motions of the foot during running may result in small and repeatable
measurement inaccuracies for some users. Each runner has a different
stride. If a user pronates or supinates significantly, or wears
orthotics, he may want to calibrate the foot pod in order to get the
best possible distance data. SpeedMax accuracy after calibration is
approximately

A good place to get more info is to look up the patents listed for the
various units. I know fitsense says they have several additional patents
in addition the the dynastream original ones.

Jan Grobbelaar said:

I did a run today (6 miles) with a running start and the unit measured
exactly the same distance out and back. I guess that the algorithms
they use to integrate acceleration to obtain velocity (and hence
speed) may assume that the initial and final velocities are equal.

Anyway, as an overpronator I think there is merit in doing the
calibration, but I still am skeptical about how that would resolve
measuring different distances over the same distance in different
directions. I wish I could lay may hands on an exact description of
the technology and algorithms used...

[email hidden] (Bumper) wrote in message news:

<[email hidden]>...

Quoted message said:

Quoted message said:

Again I cannot speak with detailed knowledge of the Polar unit, but I
know that the original Nike units were very sensitive to overpronation
and movement of the pod on the shoe.

Another factor of the accuracy was pace. If you calibrated at a 9:00
pace and then slowed down to a 13 or 14:00 pace or sped up to a 5:00
pace the accuracy suffered. Nike reccomended that you calibrate and use
the unit with one pair of shoes. The newer Nike units, like the
Fitsense, allow for an easier changing of the offset, but you still need
to calibrate for each pair of shoes.

I know that all of this doesn't sound like the ads, but once you have
set everything up and tested and retested the repeatabilty of the system
is quite good, typically 98% or better, so no matter how much you might
be off, it's still going to be more accurate than guestimating.

Jan Grobbelaar <[email hidden]> wrote:

> [email hidden] (Bumper) wrote in message news:

<[email hidden]>...

Quoted message said:

Quoted message said:

> > I don't have the Polar version, I have one of the older Nike models, but
> > I believe the technology is same. I gave up testing its accuracy because
> > it gave very repeatable results. I tested it on grass, gravel and
> > asphalt with very close if not identical results. My "gold standard" was
> > the ups and downs of a parking garage I use to hill training. Always
> > within one one hundredths of each other.
> >
> > Is that the best accuracy it gives, tenths of a mile? If it had given
> > you closer results it might have be just cutting corners.
> >
>
> I'll have to check the accuracy claims.
>
> > Does this unit require that you run a standard distance to set it for
> > maximum accuracy? Did you set the offset and retest?
>
> You can calibrate the distance measurement, but his would not resolve
> measuring the same distance in different directions - it should be the
> same distance regardless of direction.
>
> > Is one direction of your trails more uphill than the other, or is there
> > a cant to the trail, similar to the crown of a road?
> >
>
> Out is more downhill then slightly up, Back is slightly down then
> uphill all the way. There is no camber on the trail - flat - ex
> railroad bed.
>
> > Do you tend to run one direction faster than the other way (this really
> > shouldn't make that much of a difference if you set it for your stride
> > and pace.
> >
>
> Usually back is 3 minutes slower, but not in these cases.
>
> > I always found that for some reason I would have to start running for
> > several yards before starting the watch or linking and thus accuracy
> > would suffer.
>
> I think I will try a running start next time instead of a standing
> start to see if that makes any difference.
>
> > Only way to tell for sure if there is a problem with the unit would be a
> > nice flat mile or longer and run it both ways, and or take it to a track
> > and run several laps and see how repeatably accurate it reads.
> >
> > Maybe someone else can make suggestions for the pace issue, I don't use
> > it so I can't help there. You might try a goggle for sdm or fitsense.
> >
> Thanks

<<
What BS! Will you STFU and just run fer christs sakes... >>

geezz.

_______
Blog, or dog? Who knows. But if you see my lost pup, please ping me!
<A
HREF="http://journals.aol.com/virginiaz/DreamingofLeonardo">http://journal
s.aol.com/virginiaz/DreamingofLeonardo</A>

In article said:

I'll have to check the accuracy claims.

Take a look at these:
http://health.groups.yahoo.com/group/S625x/message/51
http://health.groups.yahoo.com/group/S625x/message/47
and the images here:
http://health.groups.yahoo.com/group/S625x/files/

You might have to join the yahoo group to view the images.
http://health.groups.yahoo.com/group/S625x/

Paul

Dot <dot.h@#duh?att.net> wrote in message news:<[email hidden]>...

Quoted message said:

Jan Grobbelaar said:

I finally decided to buy this running computer instead of a GPS
device, as I mostly run in wooded areas where I had found GPS devices
to be very ineffective.

I'm assuming this is trail.

Correct.

Quoted message said:

Regarding your GPS, are you carrying that by hand, on belt, or on top a
back pack. FWIW, I've had really crummy readings when I use my GPS by
hand, but fairly reliable readings when mounting it on my camelbak.
(I've never checked satellite configs to see if they happen to be good
on days with cb and bad on hand-carry days, since that could be an issue
also.) I'm just interested in total distance and elevation, not in
instantaneous pace readings since I train by effort. I also think that
expecting great precision from these devices on trails may be pushing
expectations a little.

My experience is based on using a GPS mounted on my bike. The foilage
is just too thick in places for it to stayed locked onto the
sattelites. In winter it works a bit better given less foilage. I
figured if a "normal" GPS does not work well, the smaller devices that
you strap on your arm would be even less effective.

Quoted message said:

Quoted message said:

1. The distances measured out and back are different (as downloaded
from the watch to my PC). My two data points are:

Run 1; Out: 5.0 miles, Back: 4.9 miles
Run 2; Out: 1.7 miles, Back: 1.6 miles

The distances out and back are physically identical. Why is it
measuring a shorter distance back? Does anyone have any ideas?

You may have run a slightly different path out/back. Or there may have
been enough variability in the way you ran it to account for the 2-6%
difference - esp. if on trail. I'd be more than happy with that level of
accuracy / repeatability for trail runs, but your needs may differ.

Now that you quantify it like that, the accuracy is actually not bad
at all. I have since achieved better results (100%).

Quoted message said:

Quoted message said:

2. The pace measurement is very "erratic" in that it could vary wildly
while running at a "steady" pace. I set the pace limit at 08:40 per
mile with a tolerance of 5 seconds, which proved to be pretty useless
as the pace would be all over the place ranging from 08:15 to 09:15
while I am running at a steady pace, as far as I know. I usually
paced by looking at time and distance and considering rate of
breathing, so I think I can run at quite a steady pace. Can your pace
really differ by a minute per mile every 5 seconds or so?

Probably measurement error, same as the distance. I'm not sure if any of
the devices have had reliable instantaneous pace estimates, but I can't
remember all the reviews over last couple years. Depending on the type
of trail you're on, I can easily see 1 min/mile variation. If you're
running by effort (breathing) then you're pace on hills would be likely
to vary.

I think the problem is that there is a lag between the displayed pace
and the actual pace, which takes some practise getting used to. When
the pace limit alarm started beeping when the pace was too fast, I
tended to run slower and slower while watching the pace for a change
and would then overcompensate and run too slow with the result that
the pace limit alarm would start beeping again because the pace was
too slow. So you end up in these oscillations between running too
fast and too slow. The trick is too anticipate the lag in the
displayed pace and change your pace very slowly if required. I
actually achieved much better results now that I have gotten more used
to it. One should also not set the tolerance too low - 5 seconds is
too tight - 10 to 15 seconds work better.

Quoted message said:

In case anyone is wondering my 625 came last week - about the time of
the first snow that looks like it's going to stay, but it may melt with
this warm weather again. The annoying snow - makes things slippery and
icy, but not enough to snowshoe on other than in front of my house the
day it fell. So far I'm finding the IR interface a little persnippity,
but that might be some quirk of my system or operator.

The IRda interface works great. I am very impressed with the 625, and
the software does everything and more for me that I used to do in
Excel myself. I am very, very happy with this product.

My solution to the weather problem was to get a decent treadmill.

Quoted message said:

I used it (uncalibrated) for a short walk on Sunday and for Tues night
run in the snow. Total recorded distance came out to about 3.4 miles on
a route I've been calling 3 miles (bike measurements). The calibration
is just a multplication factor, so it's just a matter of seeing if it
measures things consistently, rather than accurately, for me at the
moment. It looks like it might have as much as 10% error in distance
measurements - BUT, this was in sloppy snow on trail, so I was
pleasantly surprised. I haven't had a chance to line up it's distances
against my bike distances by various landmarks (partly since I wasn't
consistent in my run - passed by some landmarks in the grayness of the
day). I'm going to calibrate, when I can (maybe not til spring depending
on snow), on the gravel edge of a flat section-line road (1 mile for
sure), then see how that plays out on other routes.

I haven't calibrated yet, as the distance measurements are very close
to the actual, known distances that I measured with my bike
(calibrated) as well as on topograpical mapping software using GPS
waypoints.

Quoted message said:

As for the foot pod, it stays in place well, but it would be more
comfortable for me if I had thicker shoe tongues (Trespasses are fairly
thin) and longer feet (I think when I start doing hills it may be a
little restrictive, but it might feel totally differently when not in snow).

I think the altitude sensor may work reasonably well, but I've only used
it on essentially flat terrain (which means it the trace was pretty
flat, but it did seem to detect a few 10- to 20-ft hills).

The software is a big improvement over the PCCoach of the 410, although
it doesn't look like it has the power of a database in terms of being
able to search for things, but I haven't played with it too much yet.
Too many ROL things at the moment.

Dot