iGolf/GPS Accuracy Update

golf-forums.net
Promoting golf discussion.

Main

Date: 11 Feb 2007 00:20:30
From: Dave Lee
Subject: iGolf/GPS Accuracy Update

Earlier I had posted a relatively negative review of GPS accuracy (using the
iGolf device) here

http://tinyurl.com/yop4a8

I've done some additional work with my iGolf device which has given me a
different (and much more positive) perspective on its usage.

I had a conversation with the iGolf folks and got some interesting info
regarding the best way to use the device. I had been using "STABILIZER ON"
and trying to remain at the measured point as long as possible. It turns out
that this does not give the best results.

The right way to use the device is "STABILIZER OFF". Just walk up to the
point to be measured and wait for a screen update (maybe a second or so).
The next reading is your best reading. Particularly with "STABILIZER OFF"
the device will sometimes just "wander off" one yard at a time for a dozen
yards or so. I had found this very disconcerting so had immediately stopped
using this mode. It seems that the best way to use the device is to 'let
this happen' and ignore the wandering. This is very counter-intuitive but
ultimately convenient. It means you get your reading quickly and if you are
unsure just back off five paces and re-approach the spot. This actually
seems to work (even though I do not understand why).

So I did an experiment to measure "accuracy" of the measurement process
only. Before I had measured the combination of mapping error and measurement
error. I put accuracy in quotes because all I was truly measuring was
repeatability (like before). But it is my opinion that for this device a
measurement of accuracy and repeatability are pretty much the same.

I mapped four points that are 100 to 200 yards from my house (surrounding my
house). Then I measured these four distances 70 times over a 5 day period
(280 measurements). Here is what I found.

91% of the measurements were within 3 yards of "correct" (where "correct" is
the average of all the measurements for that point)

97% within 4 yards

99% within 5 yards

This is very different from my earlier results for two reasons.

1) My earlier results including measurement error AND mapping error.

2) My measurement process was very different. In looking at my old data I
would guess that the measurement error window (removing mapping error) using
the old process was roughly double what I am seeing now.

I still seem to have an unresolved mapping error (seems to be larger than
measurement error). But I can now see how this technology can actually be
useful on the course and I may well end up being a user.

dave

Date: 12 Feb 2007 00:25:14
From: Tom K
Subject: Re: iGolf/GPS Accuracy Update

That "wander off" that you mention doesn't seem to make any sense from a GPS
perspective, unless it's simply not a very good unit.

A GPS calculates it's location based on the location of the satellites in
orbit, using triangulation. The more satellites in a good pattern, the more
accurate the reading (my Garmin actually give you it's level of accuracy
based on the number is satellites in view and the patter of them). There
should be no "wandering" off of the location since SA (Selective
Availability) was turned off during the Clinton years. With a good view of
the sky (trees do hinder the signal), you should be able to get an accuracy
of about 10 ft. It shouldn't get less accurate over time. Once the GPS
gets it's fix at the beginning of the round, it knows where it's at all the
time, until either the view of the sky is lost (due to trees), or it is
turned off.

There should be no wander unless there are a lot of trees around, and the
view of satellites is compromised.

--Tom

"Dave Lee" <DaveLeeNC@ix.netcom.RemovE.com > wrote in message
news:ijtzh.23236$yx6.4657@newsread2.news.pas.earthlink.net...
> Earlier I had posted a relatively negative review of GPS accuracy (using
> the
> iGolf device) here
>
> http://tinyurl.com/yop4a8
>
> I've done some additional work with my iGolf device which has given me a
> different (and much more positive) perspective on its usage.
>
> I had a conversation with the iGolf folks and got some interesting info
> regarding the best way to use the device. I had been using "STABILIZER ON"
> and trying to remain at the measured point as long as possible. It turns
> out
> that this does not give the best results.
>
> The right way to use the device is "STABILIZER OFF". Just walk up to the
> point to be measured and wait for a screen update (maybe a second or so).
> The next reading is your best reading. Particularly with "STABILIZER OFF"
> the device will sometimes just "wander off" one yard at a time for a dozen
> yards or so. I had found this very disconcerting so had immediately
> stopped
> using this mode. It seems that the best way to use the device is to 'let
> this happen' and ignore the wandering. This is very counter-intuitive but
> ultimately convenient. It means you get your reading quickly and if you
> are
> unsure just back off five paces and re-approach the spot. This actually
> seems to work (even though I do not understand why).
>
> So I did an experiment to measure "accuracy" of the measurement process
> only. Before I had measured the combination of mapping error and
> measurement
> error. I put accuracy in quotes because all I was truly measuring was
> repeatability (like before). But it is my opinion that for this device a
> measurement of accuracy and repeatability are pretty much the same.
>
> I mapped four points that are 100 to 200 yards from my house (surrounding
> my
> house). Then I measured these four distances 70 times over a 5 day period
> (280 measurements). Here is what I found.
>
> 91% of the measurements were within 3 yards of "correct" (where "correct"
> is
> the average of all the measurements for that point)
>
> 97% within 4 yards
>
> 99% within 5 yards
>
> This is very different from my earlier results for two reasons.
>
> 1) My earlier results including measurement error AND mapping error.
>
> 2) My measurement process was very different. In looking at my old data I
> would guess that the measurement error window (removing mapping error)
> using
> the old process was roughly double what I am seeing now.
>
> I still seem to have an unresolved mapping error (seems to be larger than
> measurement error). But I can now see how this technology can actually be
> useful on the course and I may well end up being a user.
>
> dave
>
>

Date: 12 Feb 2007 12:05:52
From: Dave Lee
Subject: Re: iGolf/GPS Accuracy Update

Date: 16 Feb 2007 20:07:45
From: Tom K
Subject: Re: iGolf/GPS Accuracy Update

"Dave Lee" <DaveLeeNC@ix.netcom.RemovE.com > wrote in message
news:AKYzh.262$_73.112@newsread2.news.pas.earthlink.net...
>
>
> Tom, you seem to understand how these things work and you used an
> interesting phrase "once the GPS gets its fix". This device has what seems
> like a 'warm-up time' and it had been my understanding that it was
> gatherering epherimis errors (orbital error data) put out by the
> satellites
> periodically. Is this what you were referring to or is there something
> else
> going on here?
>
> Thanks.
>
> dave
>

A GPS takes a few minutes at start up to get a fix on the satellites. Start
up time depends on a few things. How much time has elapsed since the last
start up, how far you are from the last start up, and the speed at which
you're going at start up. At least that's what I see with my Garmin.

When I take a flight from the US to Europe or Puerto Rico, it generally
takes longer for the unit to calculate the new configuration of satellite
locations. With my Garmin, you can actually look and see the satellite
locations. You can see the types of patterns that work and the ones
difficult to get a fix on. Since it triangulates, 3 satellites in a row
don't help a whole lot. It's best to have a good spread of satellites,
evenly spread across the sky.

When I don't use it for a few days it also takes longer to get a fix on the
location. And sometimes in a moving car, moving train, or moving plane it
also seems to take longer.

But turn it on a few minutes, or an hour after using it and it will get a
fix in seconds.

What also happens, is if you lose your signal, it has to go through the
re-acquiring process again. This can happen on a course, if you go under
deep tree cover.

But if you're out in open areas, it maintains it's fix. It knows exactly
where it is (+/- the 10 ft. accuracy). There shouldn't be any errors beyond
the 10 ft. accuracy. Not unless you lost the signal for some reason such as
trees or buildings obscuring enough of the sky. But if you parked your
bag/cart under a huge, dense tree, that could be enough to lose the signal.

Now in the old days (pre Clinton), they had a thing called SA that
intentionally did more scrambling. It was designed to keep an enemy missile
from using US GPS navigation to seek a target. But Clinton signed something
to have SA turned off.

--Tom

Date: 17 Feb 2007 01:33:17
From: Dave Lee
Subject: Re: iGolf/GPS Accuracy Update

"Tom K" <tkanitra@optonline.net > wrote in message
news:AzsBh.44$lv5.30@newsfe09.lga...
>
> "Dave Lee" <DaveLeeNC@ix.netcom.RemovE.com> wrote in message
> news:AKYzh.262$_73.112@newsread2.news.pas.earthlink.net...
> >
> >
> > Tom, you seem to understand how these things work and you used an
> > interesting phrase "once the GPS gets its fix". This device has what
seems
> > like a 'warm-up time' and it had been my understanding that it was
> > gatherering epherimis errors (orbital error data) put out by the
> > satellites
> > periodically. Is this what you were referring to or is there something
> > else
> > going on here?
> >
> > Thanks.
> >
> > dave
> >
>
> A GPS takes a few minutes at start up to get a fix on the satellites.
Start
> up time depends on a few things. How much time has elapsed since the last
> start up, how far you are from the last start up, and the speed at which
> you're going at start up. At least that's what I see with my Garmin.
>
> When I take a flight from the US to Europe or Puerto Rico, it generally
> takes longer for the unit to calculate the new configuration of satellite
> locations. With my Garmin, you can actually look and see the satellite
> locations. You can see the types of patterns that work and the ones
> difficult to get a fix on. Since it triangulates, 3 satellites in a row
> don't help a whole lot. It's best to have a good spread of satellites,
> evenly spread across the sky.
>
> When I don't use it for a few days it also takes longer to get a fix on
the
> location. And sometimes in a moving car, moving train, or moving plane it
> also seems to take longer.
>
> But turn it on a few minutes, or an hour after using it and it will get a
> fix in seconds.
>
> What also happens, is if you lose your signal, it has to go through the
> re-acquiring process again. This can happen on a course, if you go under
> deep tree cover.
>
> But if you're out in open areas, it maintains it's fix. It knows exactly
> where it is (+/- the 10 ft. accuracy). There shouldn't be any errors
beyond
> the 10 ft. accuracy. Not unless you lost the signal for some reason such
as
> trees or buildings obscuring enough of the sky. But if you parked your
> bag/cart under a huge, dense tree, that could be enough to lose the
signal.
>
> Now in the old days (pre Clinton), they had a thing called SA that
> intentionally did more scrambling. It was designed to keep an enemy
missile
> from using US GPS navigation to seek a target. But Clinton signed
something
> to have SA turned off.
>
> --Tom
>

Thanks, Tom. That is pretty much my understanding and experience with the
iGolf device. For me a short perios of losing a signal results in basically
instantaneous recovery (once you are back in the open). My sense of things
is that the device is not as accurate for the next (as a guess) 30 seconds
or so. I could be wrong about this. What is your experience with the Garmin
when you (very) temporarily lose a signal?

dave

Date: 18 Feb 2007 00:34:07
From: Tom K
Subject: Re: iGolf/GPS Accuracy Update

"Dave Lee" <DaveLeeNC@ix.netcom.RemovE.com > wrote in message
news:xXsBh.2950$tD2.701@newsread1.news.pas.earthlink.net...
> >>
>
> Thanks, Tom. That is pretty much my understanding and experience with the
> iGolf device. For me a short perios of losing a signal results in
> basically
> instantaneous recovery (once you are back in the open). My sense of things
> is that the device is not as accurate for the next (as a guess) 30 seconds
> or so. I could be wrong about this. What is your experience with the
> Garmin
> when you (very) temporarily lose a signal?
>
> dave
>

What you may be noticing is the difference in accuracy between a fix with
3-4 satellites and 8-10. With 3-4 you may only get 20-30 ft. accuracy.
With 8-10 you get the 10 ft. accuracy.

When you re-acquire after losing a signal, your unit may get a fix based on
just the first 3-4 satellites (it doesn't wait until all are acquired). So
the accuracy would be 20-30 ft. As more satellites are acquired, the
accuracy will be increased.

Does the unit have a screen where you can view the location of satellites in
the sky, and which ones are actually acquired? The Garmin does, but I don't
know if the golf units do.

--Tom

Date: 18 Feb 2007 12:04:43
From: Dave Lee
Subject: Re: iGolf/GPS Accuracy Update

"Tom K" <tkanitra@optonline.net > wrote in message
news:jzRBh.9011$1y5.7954@newsfe12.lga...
>
> "Dave Lee" <DaveLeeNC@ix.netcom.RemovE.com> wrote in message
> news:xXsBh.2950$tD2.701@newsread1.news.pas.earthlink.net...
> > >>
> >
> > Thanks, Tom. That is pretty much my understanding and experience with
the
> > iGolf device. For me a short perios of losing a signal results in
> > basically
> > instantaneous recovery (once you are back in the open). My sense of
things
> > is that the device is not as accurate for the next (as a guess) 30
seconds
> > or so. I could be wrong about this. What is your experience with the
> > Garmin
> > when you (very) temporarily lose a signal?
> >
> > dave
> >
>
> What you may be noticing is the difference in accuracy between a fix with
> 3-4 satellites and 8-10. With 3-4 you may only get 20-30 ft. accuracy.
> With 8-10 you get the 10 ft. accuracy.
>
> When you re-acquire after losing a signal, your unit may get a fix based
on
> just the first 3-4 satellites (it doesn't wait until all are acquired).
So
> the accuracy would be 20-30 ft. As more satellites are acquired, the
> accuracy will be increased.
>
> Does the unit have a screen where you can view the location of satellites
in
> the sky, and which ones are actually acquired? The Garmin does, but I
don't
> know if the golf units do.
>
> --Tom
>

Yes - the unit does have a screen with a "map" of the satellites and their
strength and "status" (acquired ot not acquired). Less than five satellites
is unusual in my experience, even shortly after a brief loss of signal. In
fact when I do my experiments I automatically "rule out" any measurement
take with less than 5, although in some conditions they do seem to come and
go much more quickly than you'd expect.

dave

Date: 12 Feb 2007 07:46:19
From: sfb
Subject: Re: iGolf/GPS Accuracy Update

Two things in play.

Wide Area Augmentation System (WAAS) is a system of ground stations that are
used to calibrate the satellites so that you get the 3 to 5 feet accuracy.

If the device sits in one place for a few minuets the locations varies
within WAAS as satellites move out and into range.

I had the Sky Caddy on the kitchen counter playing a hole across town about
5000 yards away. Over an hour of so, the distances varied within plus or
minus 5 yards which is a miniscule error unless you are playing a typical
RSG pitching wedge from my kitchen.

"Dave Lee" <DaveLeeNC@ix.netcom.RemovE.com > wrote in message
news:AKYzh.262$_73.112@newsread2.news.pas.earthlink.net...
>
> "Tom K" <tkanitra@optonline.net> wrote in message
> news:_SSzh.76$0O1.69@newsfe12.lga...
>> That "wander off" that you mention doesn't seem to make any sense from a
> GPS
>> perspective, unless it's simply not a very good unit.
>>
>> A GPS calculates it's location based on the location of the satellites in
>> orbit, using triangulation. The more satellites in a good pattern, the
> more
>> accurate the reading (my Garmin actually give you it's level of accuracy
>> based on the number is satellites in view and the patter of them). There
>> should be no "wandering" off of the location since SA (Selective
>> Availability) was turned off during the Clinton years. With a good view
> of
>> the sky (trees do hinder the signal), you should be able to get an
> accuracy
>> of about 10 ft. It shouldn't get less accurate over time. Once the GPS
>> gets it's fix at the beginning of the round, it knows where it's at all
> the
>> time, until either the view of the sky is lost (due to trees), or it is
>> turned off.
>>
>> There should be no wander unless there are a lot of trees around, and the
>> view of satellites is compromised.
>>
>> --Tom
>>
>
> Tom, you seem to understand how these things work and you used an
> interesting phrase "once the GPS gets its fix". This device has what seems
> like a 'warm-up time' and it had been my understanding that it was
> gatherering epherimis errors (orbital error data) put out by the
> satellites
> periodically. Is this what you were referring to or is there something
> else
> going on here?
>
> Thanks.
>
> dave
>
>

Date: 12 Feb 2007 11:58:04
From: Dave Lee
Subject: Re: iGolf/GPS Accuracy Update

"Tom K" <tkanitra@optonline.net > wrote in message
news:_SSzh.76$0O1.69@newsfe12.lga...
> That "wander off" that you mention doesn't seem to make any sense from a
GPS
> perspective, unless it's simply not a very good unit.
>
> A GPS calculates it's location based on the location of the satellites in
> orbit, using triangulation. The more satellites in a good pattern, the
more
> accurate the reading (my Garmin actually give you it's level of accuracy
> based on the number is satellites in view and the patter of them). There
> should be no "wandering" off of the location since SA (Selective
> Availability) was turned off during the Clinton years. With a good view
of
> the sky (trees do hinder the signal), you should be able to get an
accuracy
> of about 10 ft. It shouldn't get less accurate over time. Once the GPS
> gets it's fix at the beginning of the round, it knows where it's at all
the
> time, until either the view of the sky is lost (due to trees), or it is
> turned off.
>
> There should be no wander unless there are a lot of trees around, and the
> view of satellites is compromised.
>
> --Tom
>
> "Dave Lee" <DaveLeeNC@ix.netcom.RemovE.com> wrote in message
> news:ijtzh.23236$yx6.4657@newsread2.news.pas.earthlink.net...
> > Earlier I had posted a relatively negative review of GPS accuracy (using
> > the
> > iGolf device) here
> >
> > http://tinyurl.com/yop4a8
> >
> > I've done some additional work with my iGolf device which has given me a
> > different (and much more positive) perspective on its usage.
> >
> > I had a conversation with the iGolf folks and got some interesting info
> > regarding the best way to use the device. I had been using "STABILIZER
ON"
> > and trying to remain at the measured point as long as possible. It turns
> > out
> > that this does not give the best results.
> >
> > The right way to use the device is "STABILIZER OFF". Just walk up to the
> > point to be measured and wait for a screen update (maybe a second or
so).
> > The next reading is your best reading. Particularly with "STABILIZER
OFF"
> > the device will sometimes just "wander off" one yard at a time for a
dozen
> > yards or so. I had found this very disconcerting so had immediately

Tom, I was talking to the keting side of the business so I don't have a
good technical explanation of what was going on.

With "STABILIZER ON" the device doesn't typically "walk off". With it OFF it
typically doesn't walk off either. But maybe 10%-15% of the time things
move 3-5 yards. Less often than this things will move more.

Wish I had a rationale for this. But in general "accurate to around 10 feet"
is somewhere in the ballpark of what I am now seeing. This device gives you
a satellite map, but no accuracy assessment.

dave

Date: 11 Feb 2007 10:23:40
From: Miss Anne Thrope
Subject: Re: iGolf/GPS Accuracy Update

And on top of all that, it will tell you the exact location of the
nearest group of impotent mama's boys.

Date: 11 Feb 2007 13:57:58
From: Blake 7
Subject: Re: iGolf/GPS Accuracy Update

Thought the following decision from the R & A Andrews might be of
interrest:-

Local rule for Distance Measuring Devices

Decision 14-3/0.5 in the 2006-2007 Decisions on the Rules of Golf enables a
Committee to introduce a Local Rule permitting golfers to use a distance
measuring device to obtain information during a round of golf. Without such
a Local Rule, for which is disqualification.

The R & Ais not advocating the intruduction of a Local Rule permitting the
use of distance measuring devices and is not intruducing such a Local Rule
for any of its own amateur championships or matches, or The Open
Championship. It is for each individual Committee to decide whether it wants
to allow the use of distance measuring devices for play on its course or in
certain of the competitions. In the absence of a Local Rule, the use of
such a device is contrary to the Rules of Golf. If the Committee wishes ro
introduce such a Local Rule, the following wording is recommended:

Distance Measuring Devices:
[Specify as appropriate, e.g., In this competition, or For al, play at this
course, etc], a player may obtain distance information by using a device
that measures distance only. However, if, during a stipulated round, a
player uses a distance measuring device that is designed to gauge or measure
other conditions that might affect his play [e.g., gradient, wind-speed,
temperature, etc.], the player is in breach of rule 14-3, for which the
penalty is disqualification, regardles of whether any such additional
functions are actually used.

It should be noted that any Local Rule, regardless of the wording, must
prohibit the use of a distance measuring device that is capable of gauging
or measuring other conditions that might affect play, even if such a
function is not used.

The R & A does not endorse or approve and brand of distance- measuring
device. Manufacturers are not permitted to use terms as "R&A approved" or
the like in any pruduct advertising.

"Dave Lee" <DaveLeeNC@ix.netcom.RemovE.com > wrote in message
news:ijtzh.23236$yx6.4657@newsread2.news.pas.earthlink.net...
> Earlier I had posted a relatively negative review of GPS accuracy (using
> the
> iGolf device) here
>
> http://tinyurl.com/yop4a8
>
> I've done some additional work with my iGolf device which has given me a
> different (and much more positive) perspective on its usage.
>
> I had a conversation with the iGolf folks and got some interesting info
> regarding the best way to use the device. I had been using "STABILIZER ON"
> and trying to remain at the measured point as long as possible. It turns
> out
> that this does not give the best results.
>
> The right way to use the device is "STABILIZER OFF". Just walk up to the
> point to be measured and wait for a screen update (maybe a second or so).
> The next reading is your best reading. Particularly with "STABILIZER OFF"
> the device will sometimes just "wander off" one yard at a time for a dozen
> yards or so. I had found this very disconcerting so had immediately
> stopped
> using this mode. It seems that the best way to use the device is to 'let
> this happen' and ignore the wandering. This is very counter-intuitive but
> ultimately convenient. It means you get your reading quickly and if you
> are
> unsure just back off five paces and re-approach the spot. This actually
> seems to work (even though I do not understand why).
>
> So I did an experiment to measure "accuracy" of the measurement process
> only. Before I had measured the combination of mapping error and
> measurement
> error. I put accuracy in quotes because all I was truly measuring was
> repeatability (like before). But it is my opinion that for this device a
> measurement of accuracy and repeatability are pretty much the same.
>
> I mapped four points that are 100 to 200 yards from my house (surrounding
> my
> house). Then I measured these four distances 70 times over a 5 day period
> (280 measurements). Here is what I found.
>
> 91% of the measurements were within 3 yards of "correct" (where "correct"
> is
> the average of all the measurements for that point)
>
> 97% within 4 yards
>
> 99% within 5 yards
>
> This is very different from my earlier results for two reasons.
>
> 1) My earlier results including measurement error AND mapping error.
>
> 2) My measurement process was very different. In looking at my old data I
> would guess that the measurement error window (removing mapping error)
> using
> the old process was roughly double what I am seeing now.
>
> I still seem to have an unresolved mapping error (seems to be larger than
> measurement error). But I can now see how this technology can actually be
> useful on the course and I may well end up being a user.
>
> dave
>
>