|
Cellular forums Home > Archive > GPS > May 2007 > SiRF III: is sensitivity better than accuracy?
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
SiRF III: is sensitivity better than accuracy?
|
|
|
| I recently bought a Holux 240 largely because it contained the SiRF
III chipset with high sensitivity and low power consumption. I have
been trying for a week to find out how to enable WAAS support in this
device. Turns out that SiRF does not claim it *has* WAAS support!
Basically SiRF said that the OEMs (Holux) can modify the design to
suit themselves and can disable features such as WAAS.
I have not been able to get Holux to respond to requests for support
and their web site provides nothing useful in trying to use any WAAS
capabilities this unit may have. Holux provides a program called GPS
Viewer which sends a "$PSRF108,01" command when WAAS is selected in
the user interface. But this command seems to be ignored by the
device and no indication shows that WAAS capability has been
enabled.
I have been searching for a several days for information on how to use
WAAS on SiRF III based receivers and it seems there is nothing
consistent that works with any number of units. I found one post on
gpspassion that said he got a GlobalSat SiRF III unit to work, but no
details and no one else confirmed it.
It would appear that since SiRF has no control over how the OEMs
implement their receivers, there is a lot of variation in the end
products using this technology. What everyone seems to tout about the
SiRF III units is their good sensitivity. But if they don't support
WAAS, the accuracy will not be good. So which is better, sensitivity
or accuracy?
| |
| Andreas van Hooijdonk 2007-05-10, 10:33 am |
| "nemo" <gnuarm@gmail.com> wrote in message
news:1178806676.226686.108650@n59g2000hsh.googlegroups.com...
> I recently bought a Holux 240 largely because it contained the SiRF
> III chipset with high sensitivity and low power consumption. I have
> been trying for a week to find out how to enable WAAS support in this
> device. Turns out that SiRF does not claim it *has* WAAS support!
> Basically SiRF said that the OEMs (Holux) can modify the design to
> suit themselves and can disable features such as WAAS.
>
> I have not been able to get Holux to respond to requests for support
> and their web site provides nothing useful in trying to use any WAAS
> capabilities this unit may have. Holux provides a program called GPS
> Viewer which sends a "$PSRF108,01" command when WAAS is selected in
> the user interface. But this command seems to be ignored by the
> device and no indication shows that WAAS capability has been
> enabled.
>
> I have been searching for a several days for information on how to use
> WAAS on SiRF III based receivers and it seems there is nothing
> consistent that works with any number of units. I found one post on
> gpspassion that said he got a GlobalSat SiRF III unit to work, but no
> details and no one else confirmed it.
>
> It would appear that since SiRF has no control over how the OEMs
> implement their receivers, there is a lot of variation in the end
> products using this technology. What everyone seems to tout about the
> SiRF III units is their good sensitivity. But if they don't support
> WAAS, the accuracy will not be good. So which is better, sensitivity
> or accuracy?
>
I think that bad sensitivity automatically results in bad accuracy in most
cases.
--
Andreas van Hooijdonk
http://www.gps-practice-and-fun.com
| |
| Holger Issle 2007-05-10, 10:33 am |
| On 10 May 2007 07:17:56 -0700, nemo wrote:
> Holux provides a program called GPS
> Viewer which sends a "$PSRF108,01" command when WAAS is selected in
> the user interface. But this command seems to be ignored by the
> device and no indication shows that WAAS capability has been
> enabled.
I have the same on my Holux 236. However, I only wanted to try it...
> What everyone seems to tout about the
> SiRF III units is their good sensitivity. But if they don't support
> WAAS, the accuracy will not be good.
Sorry, I need to disagree. Most cases the track is in sync with the
road and offroad maps I have, in a ~5m radius. What else would I need?
I am not going to touch down a plane using that, I am following a road
and have other senses at my head.
And in those cases where the reception is not good enough to have good
accuracy, there would for sure not be any augmentation service
available or the signal would be useless as it is not constructed to
correct errors from reflection but from run time changes in the
athmosphere.
I had it on my Map76S before, and turned it on to see if it works, and
then turned it off again....
--
Ciao,
Holger (GUS-KOTAL, GUS#1100)
90-92 Honda CB400 10 Mm | 93-95 Yamaha TDM 850 26 Mm
95-97 KTM 620 LC4 13 Mm | seit 97 BMW R1100GS 50 Mm (Die Renndrecksau!)
cu @ http://www.issle.de
| |
| Dan Anderson 2007-05-10, 12:33 pm |
| nemo wrote:
& #91;snip]
> It would appear that since SiRF has no control over how the OEMs
> implement their receivers, there is a lot of variation in the end
> products using this technology. What everyone seems to tout about the
> SiRF III units is their good sensitivity. But if they don't support
> WAAS, the accuracy will not be good. So which is better, sensitivity
> or accuracy?
The sensitivity means it will work in urban canyons for road
navigation. If it's off by sixty feet, it probably doesn't matter.
The user will get the travel directions needed.
The sensitivity means it picks up multipath that will make it
somewhat poorer in accuracy but does produce a position
solution as opposed to none at all. Loss of lock tends to
upset your casual user, especially when trying to get
somewhere in a city.
I don't have a lot of experience with the SiRF III yet, but
I've seen a Garmin GPSmap 76S produce a relatively more
accurate track receiving four satellites in a poor pattern
in a mountain canyon than a GPSmap 76Cx receiving eight
satellites. Four of those were not directly visible so
the 76S didn't receive them but it looked like the 76Cx
picked up bounced signals which resulted in a less accurate
track. For most users it wasn't off far enough to matter.
If you really want accuracy, you don't want to pick up
multipath signals.
--
Dan
(email change 2001 to 2004)
(www.gpsmap.net)
| |
|
| There are different versions of firmware for the Sirf III. Supposedly,
the early versions did not support WAAS, but the more recent versions
do. You may have to enable WAAS with the SirfDemo program. At any
rate, you can use SirfDemo to read your firmware number and ask Sirf
if that version supports WAAS. Remember earlier caveats and don't
issue Sirf commands to change the baud rate.
Sirf has to respect non-disclosure agreements, so they won't answer
questions if you mention the name "Holux". With my Holux 236, the
Sirf chip is "stock", and it is probably the same for the 240. Other
than setting the default configuration, Holux has no reason to use
custom firmware.
Just a double check. Are you in a location where you can track the
WAAS satellites?
| |
| laura halliday 2007-05-10, 12:33 pm |
| On May 10, 9:28 am, John <goo...@coyotebush.net> wrote:
> There are different versions of firmware for the Sirf III. Supposedly,
> the early versions did not support WAAS, but the more recent versions
> do. You may have to enable WAAS with the SirfDemo program. At any
> rate, you can use SirfDemo to read your firmware number and ask Sirf
> if that version supports WAAS. Remember earlier caveats and don't
> issue Sirf commands to change the baud rate.
>
> Sirf has to respect non-disclosure agreements, so they won't answer
> questions if you mention the name "Holux". With my Holux 236, the
> Sirf chip is "stock", and it is probably the same for the 240. Other
> than setting the default configuration, Holux has no reason to use
> custom firmware.
>
> Just a double check. Are you in a location where you can track the
> WAAS satellites?
I have a Holux 213 (cheapest I could find on eBay :-), and WAAS
works as advertised. I routinely see PRN 48 (Galaxy 15) and
PRN 51 (Anik F1R) in the output data.
As far as I can tell from the SiRF documentation, the device
is a totally standard SiRF III.
Laura Halliday VE7LDH "Que les nuages soient notre
Grid: CN89mg pied a terre..."
ICBM: 49 16.05 N 122 56.92 W - Hospital/Shafte
| |
|
| On May 10, 12:28 pm, John <goo...@coyotebush.net> wrote:
> There are different versions of firmware for the Sirf III. Supposedly,
> the early versions did not support WAAS, but the more recent versions
> do. You may have to enable WAAS with the SirfDemo program. At any
> rate, you can use SirfDemo to read your firmware number and ask Sirf
> if that version supports WAAS. Remember earlier caveats and don't
> issue Sirf commands to change the baud rate.
No, the SiRFDemo program will not work with this unit. At least I
have not figured out how to do that. It has a control to put the
receiver in binary mode, but it does not let you control any of the
command parameters. This control appears to change the baud rate to
57600 (the data source setup changes to 57600). I believe this sets
the rate on the SiRF chip to 57600, but the Holux bluetooth chip does
not know the rate has changed and remains at 38400. The result is the
unit sends out garbage and will no longer respond to commands. Then I
have to let the battery drain for this to reset.
> Sirf has to respect non-disclosure agreements, so they won't answer
> questions if you mention the name "Holux". With my Holux 236, the
> Sirf chip is "stock", and it is probably the same for the 240. Other
> than setting the default configuration, Holux has no reason to use
> custom firmware.
>
> Just a double check. Are you in a location where you can track the
> WAAS satellites?
Yup! I had a Magellan Merigold that would pick up WAAS very easily
and report location error estimates down to 7 feet! Unfortunatly the
screen died and it is no longer usable.
I continued with my tests today and I did find a way to put the Holux
240 into binary mode. I used a tool called NMEA GPS Monitor. It
allows you to send and receive binary commands without trying to
understand or interpret them. It does have the great feature of
calculating check sums however!
So I put the Holux 240 into binary mode and received valid SiRF status
messages. I sent this command to select the DGPS source as SBAS
satellite. It still reports that DGPS is *not* using differential
corrections.
Just to verify everything I said about SiRFDemo, I ended up putting
the receiver into the corrupt state and I have to wait for the battery
to drain. When it comes back up I will try some more tests to see if
I can get DGPS WAAS to work.
With all the problems I have had trying to get this working (not to
mention an issue with a dispreputable eBay seller who sent me a broken
Palm) I have not geocached in three weeks. I am getting tired of
playing with the gadgets and would like to be able to use the durn
things!!!
BTW, NMEA GPS Monitor is a *GREAT* programfor the PC!!! It not only
provides this excellent communications mode, it displays a lot of data
(NMEA only of course, read the name) and is FREE! It is not available
for PDAs or I would just use that on the palm. It is not open source
or I would port it!!!
| |
|
| On May 10, 12:59 pm, laura halliday <marsga...@hotmail.com> wrote:
> On May 10, 9:28 am, John <goo...@coyotebush.net> wrote:
>
>
>
>
> I have a Holux 213 (cheapest I could find on eBay :-), and WAAS
> works as advertised. I routinely see PRN 48 (Galaxy 15) and
> PRN 51 (Anik F1R) in the output data.
>
> As far as I can tell from the SiRF documentation, the device
> is a totally standard SiRF III.
Yes, but that is a very different animal from the bluetooth
receivers. The bluetooth receivers have an extra chip to handle the
bluetooth functions and acts as a middleman in all communications.
For whatever reason, the idividual vendors seem to modify the code on
the SiRF chip to add commands I would expect. But I doubt that Holux
has done this since they don't bother to provide any documentation.
What good would additional commands be without docs?
I think Holux is simply too lazy to care if their users can use WAAS
and don't bother to add any NMEA commands to enable it. I expect SiRF
binary commands will work, but I have not been able to get it working
so far. Again, it's not like I am going to get any support from
Holux!
The only question left in my mind is just how much of an improvement
you will get from having WAAS enabled and working. The people from
SiRF seem to be saying it will only improved accuracy by 1 to 2
meters. But they only support that statement under perfect conditions
and then try to extrapolate to the real world. If I ever get WAAS to
work on this unit, I will do some real world testing.
| |
| sat_alight@yahoo.com 2007-05-10, 10:33 pm |
| Greater sensitivity does not equate to better accuracy. In fact, they
can be opposing forces. Greater sensitivity can simply mean the
receiver accepts noiser data, which means less accurate data.
Don't ignore the antenna quality either. No matter how good a receiver
is, if the antenna is poorly designed the receiver will perform
poorly.
My experience with Holux products has been poor. The ability for the
one I used to hang onto WAAS when you are moving is nearly non-
existent.
> I recently bought a Holux 240 largely because it contained the SiRF
> III chipset with high sensitivity and low power consumption. I have
> been trying for a week to find out how to enable WAAS support in this
> device. Turns out that SiRF does not claim it *has* WAAS support!
> Basically SiRF said that the OEMs (Holux) can modify the design to
> suit themselves and can disable features such as WAAS.
>
> I have not been able to get Holux to respond to requests for support
> and their web site provides nothing useful in trying to use any WAAS
> capabilities this unit may have. Holux provides a program called GPS
> Viewer which sends a "$PSRF108,01" command when WAAS is selected in
> the user interface. But this command seems to be ignored by the
> device and no indication shows that WAAS capability has been
> enabled.
>
> I have been searching for a several days for information on how to use
> WAAS on SiRF III based receivers and it seems there is nothing
> consistent that works with any number of units. I found one post on
> gpspassion that said he got a GlobalSat SiRF III unit to work, but no
> details and no one else confirmed it.
>
> It would appear that since SiRF has no control over how the OEMs
> implement their receivers, there is a lot of variation in the end
> products using this technology. What everyone seems to tout about the
> SiRF III units is their good sensitivity. But if they don't support
> WAAS, the accuracy will not be good. So which is better, sensitivity
> or accuracy?
| |
| Jack Yeazel 2007-05-10, 10:33 pm |
|
sat_alight@yahoo.com wrote:
>
> Greater sensitivity does not equate to better accuracy. In fact, they
> can be opposing forces. Greater sensitivity can simply mean the
> receiver accepts noiser data, which means less accurate data.
-All too true... The first Garmin SIRF receivers had poor
accuracy, but they were able to upgrade their firmware to
fix the problem... Here are our test results:
http://gpsinformation.us/gps60c/g76Creview.html#NOTE
This also indicates that the non-SIRF GPSmap 76C is still
relatively inaccurate under moderate tree cover...
--
Jack
Get general GPS information at:
http://www.gpsinformation.net/
| |
|
| On May 10, 5:57 pm, sat_ali...@yahoo.com wrote:
> Greater sensitivity does not equate to better accuracy. In fact, they
> can be opposing forces. Greater sensitivity can simply mean the
> receiver accepts noiser data, which means less accurate data.
That does not sound correct. If a receiver has more sensitivity, that
means it will pull the signal out of the noise better (actually it
makes less noise on its input and so can hear lower level signals).
What you are describing is just gain; amplify the input, both signal
and noise. Gain is necessary, but not sufficient to have a good
receiver.
> Don't ignore the antenna quality either. No matter how good a receiver
> is, if the antenna is poorly designed the receiver will perform
> poorly.
Yep, but from what I can see, most units have very similar if not
identical patch antennas. I was told by two GPS module reps to avoid
the quad helix antennas, but they did not give specifics.
> My experience with Holux products has been poor. The ability for the
> one I used to hang onto WAAS when you are moving is nearly non-
> existent.
Can you give specifics? What unit did you have? Where did you use
it? Did losing WAAS make a difference in the accuracy of the fix?
Actually, after reading what the SiRF reps are saying, I am starting
to believe that WAAS is not really a significant improvement.
I did finally get the Holux 240 to work with WAAS. I had is
programmed correctly using GPS NMEA Monitor and sending binary
commands. But I could not really see what was going on since it does
not display any results in binary mode. Dennis Gr=F6ning sent me an
email with very detailed instructions on how to use SiRFDemo to set a
bluetooth receiver to work with SBAS. Then I could see the results
and realized that I needed to move the antenna to get a better
signal. I am now picking up SV135 and have a DGPS coorection.
Woohoo! or not...
In response to Dan Anderson; higher sensitivity does not mean it works
better in urban canyons or any other canyons. The problem there is
multipath, where a signal travels a path to the antenna *other* than a
straight line which is what the calculations assume. In fact,
multipath if often a lower level that is better off ignored or comes
in at a low angle which the antenna could ignore. There are some
fancy receivers that make use of multipath to actually boost the
received signal, but I doubt that those techniques are used in
handheld or bluetooth GPS receivers. They tend to be processor
intensive and use extra power.
| |
| sat_alight@yahoo.com 2007-05-11, 4:33 am |
|
>
> That does not sound correct. If a receiver has more sensitivity, that
> means it will pull the signal out of the noise better (actually it
> makes less noise on its input and so can hear lower level signals).
> What you are describing is just gain; amplify the input, both signal
> and noise. Gain is necessary, but not sufficient to have a good
> receiver.
Ok. You'll have to read about GPS signal processing. I'm not writing
that sarcasticly, just worth the read if you want to understand it.
>
> Yep, but from what I can see, most units have very similar if not
> identical patch antennas. I was told by two GPS module reps to avoid
> the quad helix antennas, but they did not give specifics.
There's a reason there are $300 patch antennas and $20 patch antennas
and $2 patch antennas. They vary widely in quality. The cheaper they
get, the less the designer cares about accuracy.
>
> Can you give specifics? What unit did you have? Where did you use
> it? Did losing WAAS make a difference in the accuracy of the fix?
I used their CF GPS receiver. GR-270 I think. Took forever to get WAAS
and very little to lose it.
WAAS makes a difference in accuracy. But WAAS wasn't designed to
improve accuracy. It was designed to improve integrity.
Whether the accuracy improvement is significant or not depends on what
you're using the position for.
> Actually, after reading what the SiRF reps are saying, I am starting
> to believe that WAAS is not really a significant improvement.
Depends on what you are using it for. Most likely you won't care or
even know if you're getting 7M or 3M accuracy. But practically
speaking, even WAAS-enabled receivers aren't using WAAS a lot of the
time because the line-of-sight to the WAAS satellites is blocked.
Marine and airborne users would be the exception.
> I did finally get the Holux 240 to work with WAAS. I had is
> programmed correctly using GPS NMEA Monitor and sending binary
> commands. But I could not really see what was going on since it does
> not display any results in binary mode. Dennis Gr=F6ning sent me an
> email with very detailed instructions on how to use SiRFDemo to set a
> bluetooth receiver to work with SBAS. Then I could see the results
> and realized that I needed to move the antenna to get a better
> signal. I am now picking up SV135 and have a DGPS coorection.
> Woohoo! or not...
>
> In response to Dan Anderson; higher sensitivity does not mean it works
> better in urban canyons or any other canyons. The problem there is
> multipath, where a signal travels a path to the antenna *other* than a
> straight line which is what the calculations assume. In fact,
> multipath if often a lower level that is better off ignored or comes
> in at a low angle which the antenna could ignore. There are some
> fancy receivers that make use of multipath to actually boost the
> received signal, but I doubt that those techniques are used in
> handheld or bluetooth GPS receivers. They tend to be processor
> intensive and use extra power.
Low-level multipath is the reason it was recommended to you that you
stay away from helical antennas. Plus, the ergonomics suck.
| |
|
| On May 11, 1:38 am, sat_ali...@yahoo.com wrote:
>
>
> Ok. You'll have to read about GPS signal processing. I'm not writing
> that sarcasticly, just worth the read if you want to understand it.
Perhaps you can point me to some references you have read?
>
>
> There's a reason there are $300 patch antennas and $20 patch antennas
> and $2 patch antennas. They vary widely in quality. The cheaper they
> get, the less the designer cares about accuracy.
My experience has been that with any product you buy there is a wide
range of pricing. Often the pricing reflects how much money is
desired to go into the pockets of the sellers. Can you give specifics
of what is better about a $300 antenna than a $20 and what exactly can
be done to make an antenna for only $2?
The Sarantel antennas are machine manufactured and cost over $10, even
in quantity. What do you mean by "accuracy", accuracy of what? I
can't recall ever seeing a spec for an antenna called "accuracy".
>
>
> I used their CF GPS receiver. GR-270 I think. Took forever to get WAAS
> and very little to lose it.
>
> WAAS makes a difference in accuracy. But WAAS wasn't designed to
> improve accuracy. It was designed to improve integrity.
>
> Whether the accuracy improvement is significant or not depends on what
> you're using the position for.
So did you *see* a difference in accuracy when you used WAAS? I am
being told by the receiver manufacturers that it can only improve the
accuracy by up to 2 meters. That is not very much compared to the
other errors that are typical in normal operation.
>
> Depends on what you are using it for. Most likely you won't care or
> even know if you're getting 7M or 3M accuracy. But practically
> speaking, even WAAS-enabled receivers aren't using WAAS a lot of the
> time because the line-of-sight to the WAAS satellites is blocked.
> Marine and airborne users would be the exception.
I am at N 40 degrees on the US East coast and I can see SV122 at 10
degrees elevation, SV135 at 17 degrees, SV138 at 34 degrees and that
is indoors. I don't seem to have any trouble finding SBAS satellites
unless I am behind a large hill. So I don't agree that WAAS
satellites are typically blocked.
>
>
> Low-level multipath is the reason it was recommended to you that you
> stay away from helical antennas. Plus, the ergonomics suck.
You say a lot of things that don't make sense. The difference between
patch and helical antenna patterns is rather moot in most practical
applications because of the lack of consistent orientation. As to the
ergonomics, that greatly depends on the application, no? The design I
worked on actually *required* the helical antenna because the unit
would be attached to the users chest or side of his shoulder. Had a
patch antenna been used the end product would have been larger
(although seeing the Holux 240 makes me think we just didn't do enough
research into available antennas) and the antenna would have been
pointed at the horizon most of the time.
I am pretty confident that in another 5 years, the signal processing
will be low power enough to provide for signal enhancement in
multiplath environments. You will actually *look* for multipath as a
way to boost gain!
| |
| Jack Erbes 2007-05-11, 12:33 pm |
| nemo wrote:
<snip>
> It would appear that since SiRF has no control over how the OEMs
> implement their receivers, there is a lot of variation in the end
> products using this technology. What everyone seems to tout about the
> SiRF III units is their good sensitivity. But if they don't support
> WAAS, the accuracy will not be good. So which is better, sensitivity
> or accuracy?
Sounds to me like you should have bought a SiRF III model that was also
WAAS or DGPS capable. It does not appear to be something that you can
just turn on yourself and reap any benefit from it as the OEMs need to
write their software to take advantage of WAAS and to give the user some
control over it. I don't think you've been defrauded, you got what they
described in their advertising.
"Is sensitivity better than accuracy?" Sensitivity is a component of or
factor in accuracy so that question does not really have an answer.
SiRF III accuracy on my 76Cx, as measured by the EPE displayed on the
Satellite page, is excellent with WAAS disabled. I live in North
America so if I turn WAAS on the EPE gets smaller. I normally have WAAS
on all the time, it does not make sense to not have it on.
Like most GPS users, I seldom know know my exact location (expressed in
a coordinate system) with any great degree of precision so my feelings
for accuracy come from my "seat of the pants" impressions as I proceed.
For example, sometimes when my 76Cx gives the final turning prompt and
states the distance, it appears to be a little off on the distance in
feet or yards. I've not taken the time or trouble to determine if the
problem resulted from the quality of my fix of from the accuracy in the
placement of the the mapping vector that the GPS is steering me to.
And I often see tracks that are offset from but parallel to a road
vector on a map for long distances. And I also see tracks that appear
to be directly on top of the road vectors on the mapping.
But in the final analysis, when you zoom down to the highest zoom levels
offered, you are probably going to find that your track is not directly
on top of the map vectors most of the time. But you don't know if the
error is in the mapping or in the GPS. When I have a 10 or 15 foot EPE
and am on a 40 foot roadway, it does not surprise me at all to find my
fix location off to the side of the road.
Improved sensitivity contributes to accuracy. And that improving
sensitivity is a key part of the process of improving accuracy.
Increasing sensitivity can add more satellites to the solution. So
increased sensitivity, along with good software to take advantage of the
additional inputs and sort out the resulting ambiguities and confusion,
should make for more accuracy.
And I can see where too much sensitivity and a lot of reflected signals
could degrade the quality of a fix. Software should see that problem
and give less consideration to the reflected signals in fix
determination but, at the same time, it should increase the EPE because
of the confusion that exists.
But I seriously doubt that the average consumer would see anything but a
steady improvement in fix quality and accuracy as they progressed from
SiRF II to SiRF III based receivers.
Jack
--
Jack Erbes in Ellsworth, Maine, USA - jackerbes at adelphia dot net
(also receiving email at jacker at midmaine.com)
| |
|
| On May 11, 1:28 pm, Jack Erbes <jacker...@adelphia.net> wrote:
> nemo wrote:
>
> <snip>
>
>
> Sounds to me like you should have bought a SiRF III model that was also
> WAAS or DGPS capable. It does not appear to be something that you can
> just turn on yourself and reap any benefit from it as the OEMs need to
> write their software to take advantage of WAAS and to give the user some
> control over it. I don't think you've been defrauded, you got what they
> described in their advertising.
I am confused. I bought a unit that was stated to have WAAS
capability and has specs for accuracy using WAAS. In addition, I have
found out how to enable it and I am watching it collect data using a 3D
+DGPS fix right now. The problem is that for bluetooth receivers,
there is no standard software that enables WAAS in the SiRF III
chipset. It requires a specific set of commands sent in SiRF binary
mode. The most effective way to do this that I found was to use the
SiRFDemo program, but to *NOT* use the "Switch to SiRF Protocol" menu
item as it put the bluetooth receiver in a corrupt state.
So, no, I have not been "defrauded", but Holux and likely most of the
other bluetooth GPS manufacturers make it very hard to use WAAS with
their units.
> "Is sensitivity better than accuracy?" Sensitivity is a component of or
> factor in accuracy so that question does not really have an answer.
My point was that if SiRF units did not support WAAS which is claimed
to give more accuracy, what is the point of getting higher
sensitivity? Nearly every unit designed in the last year or two has
very sensitive chipsets. The SiRF chipset still has the best claimed
sensitivity, but only by a small margin. So at this point I would not
buy a SiRF based unit because of the claimed "high" sensitivity. On
the other hand, I am being told that WAAS does not really provide a
significant increase in accuracy. So I would say neither are good
criteria to select a current GPS receiver. Instead battery life, unit
size and support are things to consider. I chose the Holux 240 based
on its small size. I would have preferred a Numerix based unit
because of lower power drain, but I am not sure that is true. I have
since found out that the Holux 240 uses a single chip GPS receiver
which has a lower power drain than the two chip SiRF III devices.
> SiRF III accuracy on my 76Cx, as measured by the EPE displayed on the
> Satellite page, is excellent with WAAS disabled. I live in North
> America so if I turn WAAS on the EPE gets smaller. I normally have WAAS
> on all the time, it does not make sense to not have it on.
I also found that I got better accuracy on my Merigold receiver when a
WAAS satellite was used. But I can't say that any of this was a valid
test. for one, I was using the EPE from the unit. This is not a true
error measurement, but just what it says, an estimate. I believe this
is based partly on the DOP which is in tern based on the configuration
of the satellites used in the fix. Also, anytime you get software
involved, especially software that you don't know the details of, it
can affect readings by using averaging. In the end I have no data
that shows that WAAS is necessary or sufficient to get an accurate
fix.
> Like most GPS users, I seldom know know my exact location (expressed in
> a coordinate system) with any great degree of precision so my feelings
> for accuracy come from my "seat of the pants" impressions as I proceed.
Bingo!
> For example, sometimes when my 76Cx gives the final turning prompt and
> states the distance, it appears to be a little off on the distance in
> feet or yards. I've not taken the time or trouble to determine if the
> problem resulted from the quality of my fix of from the accuracy in the
> placement of the the mapping vector that the GPS is steering me to.
Yes, that is why additional software needs to be excluded when trying
to determine the GPS accuracy.
> And I often see tracks that are offset from but parallel to a road
> vector on a map for long distances. And I also see tracks that appear
> to be directly on top of the road vectors on the mapping.
Again, more software (or data) just muddies the waters.
> But in the final analysis, when you zoom down to the highest zoom levels
> offered, you are probably going to find that your track is not directly
> on top of the map vectors most of the time. But you don't know if the
> error is in the mapping or in the GPS. When I have a 10 or 15 foot EPE
> and am on a 40 foot roadway, it does not surprise me at all to find my
> fix location off to the side of the road.
>
> Improved sensitivity contributes to accuracy. And that improving
> sensitivity is a key part of the process of improving accuracy.
>
> Increasing sensitivity can add more satellites to the solution. So
> increased sensitivity, along with good software to take advantage of the
> additional inputs and sort out the resulting ambiguities and confusion,
> should make for more accuracy.
>
> And I can see where too much sensitivity and a lot of reflected signals
> could degrade the quality of a fix. Software should see that problem
> and give less consideration to the reflected signals in fix
> determination but, at the same time, it should increase the EPE because
> of the confusion that exists.
How does the software know which signals are reflected?
> But I seriously doubt that the average consumer would see anything but a
> steady improvement in fix quality and accuracy as they progressed from
> SiRF II to SiRF III based receivers.
I am not interested in comparing the SiRF III to the SiRF II. I am
talking about the choice of SiRF III vs. other current units.
Regardless, my original point was about a unit that I thought did not
provide WAAS capability. It does provide that capability, the maker
just makes it very hard to make use of it.
| |
| Jack Erbes 2007-05-11, 10:33 pm |
| nemo wrote:
<snip>
> I am confused. I bought a unit that was stated to have WAAS
> capability and has specs for accuracy using WAAS. In addition, I have
> found out how to enable it and I am watching it collect data using a 3D
> +DGPS fix right now. The problem is that for bluetooth receivers,
> there is no standard software that enables WAAS in the SiRF III
> chipset. It requires a specific set of commands sent in SiRF binary
> mode. The most effective way to do this that I found was to use the
> SiRFDemo program, but to *NOT* use the "Switch to SiRF Protocol" menu
> item as it put the bluetooth receiver in a corrupt state.
>
> So, no, I have not been "defrauded", but Holux and likely most of the
> other bluetooth GPS manufacturers make it very hard to use WAAS with
> their units.
Maybe I was confused then. I looked at a link for the Holux 240 and the
specs did not mention WAAS or DGPS. For accuracy it mentioned only the
non-DPGS accuracy of the unit. So I assumed that it was not sold as a
WAAS capable unit and you were trying to turn it on through a back door
of sorts. I just went to the Holux page and now I see see that an
EGNOS/WAAS accuracy spec is listed.
I don't care for Bluetooth in general because it is a labyrinth of
sellers and implementations that seem to have taken something that
should be simple and made it complicated and hard to use. Maybe that is
part of the problem here, that the OEMs have not developed a software
interface that is adequate to the task under Bluetooth.
>
> My point was that if SiRF units did not support WAAS which is claimed
> to give more accuracy, what is the point of getting higher
> sensitivity? Nearly every unit designed in the last year or two has
> very sensitive chipsets. The SiRF chipset still has the best claimed
> sensitivity, but only by a small margin. So at this point I would not
> buy a SiRF based unit because of the claimed "high" sensitivity. On
> the other hand, I am being told that WAAS does not really provide a
> significant increase in accuracy. So I would say neither are good
> criteria to select a current GPS receiver. Instead battery life, unit
> size and support are things to consider. I chose the Holux 240 based
> on its small size. I would have preferred a Numerix based unit
> because of lower power drain, but I am not sure that is true. I have
> since found out that the Holux 240 uses a single chip GPS receiver
> which has a lower power drain than the two chip SiRF III devices.
Increased sensitivity, without WAAS, could add more satellites to the
fix calculation process and that should improve accuracy. I don't know
how to evaluate or qualify for "significant" but you are likely to do
better with 5 satellites than you are with 3 or 4. On my 76Cx, I see 6,
8, or 10 bird fixes regularly and 11 or 12 occasionally. And the EPE
number generally gets smaller as the number of birds increases and when
I am using WAAS.
>
> I also found that I got better accuracy on my Merigold receiver when a
> WAAS satellite was used. But I can't say that any of this was a valid
> test. for one, I was using the EPE from the unit. This is not a true
> error measurement, but just what it says, an estimate. I believe this
> is based partly on the DOP which is in tern based on the configuration
> of the satellites used in the fix. Also, anytime you get software
> involved, especially software that you don't know the details of, it
> can affect readings by using averaging. In the end I have no data
> that shows that WAAS is necessary or sufficient to get an accurate
> fix.
>
> Yes, that is why additional software needs to be excluded when trying
> to determine the GPS accuracy.
Well its not something that is typically or easily done for consumer
grade GPSRs. And most consumers are happy with the results and not
trying to hard to understand the hows and whys of it all.
>
> Again, more software (or data) just muddies the waters.
But not always. I think the people studying the issues at a more
technical level are, or at least should be, able to use the results from
present studies to improve the software routines.
<snip>
> How does the software know which signals are reflected?
<snip>
I don't know for sure but I can guess. If the software sees a
continuing trend from a specific satellite and then receives a series or
sequence of signals that vary from that norm, the software might exclude
that input from the fix determination process or reduce a weighting
factor if there is such a thing.
WAAS adds the feedback from the ground reference stations to the fix
evaluation and determination process so if you already had a fix, it
should get better, no? I know nothing about the actual software
routines used so I'm purely guessing here.
Jack
--
Jack Erbes in Ellsworth, Maine, USA - jackerbes at adelphia dot net
(also receiving email at jacker at midmaine.com)
| |
| Larry G 2007-05-12, 7:33 am |
| On May 11, 8:04 pm, Jack Erbes <jacker...@adelphia.net> wrote:
> nemo wrote:
>
> <snip>
>
>
>
> Maybe I was confused then. I looked at a link for the Holux 240 and the
> specs did not mention WAAS or DGPS. For accuracy it mentioned only the
> non-DPGS accuracy of the unit. So I assumed that it was not sold as a
> WAAS capable unit and you were trying to turn it on through a back door
> of sorts. I just went to the Holux page and now I see see that an
> EGNOS/WAAS accuracy spec is listed.
>
> I don't care for Bluetooth in general because it is a labyrinth of
> sellers and implementations that seem to have taken something that
> should be simple and made it complicated and hard to use. Maybe that is
> part of the problem here, that the OEMs have not developed a software
> interface that is adequate to the task under Bluetooth.
>
>
>
>
>
>
>
> Increased sensitivity, without WAAS, could add more satellites to the
> fix calculation process and that should improve accuracy. I don't know
> how to evaluate or qualify for "significant" but you are likely to do
> better with 5 satellites than you are with 3 or 4. On my 76Cx, I see 6,
> 8, or 10 bird fixes regularly and 11 or 12 occasionally. And the EPE
> number generally gets smaller as the number of birds increases and when
> I am using WAAS.
>
>
>
>
>
>
>
>
> Well its not something that is typically or easily done for consumer
> grade GPSRs. And most consumers are happy with the results and not
> trying to hard to understand the hows and whys of it all.
>
>
>
> But not always. I think the people studying the issues at a more
> technical level are, or at least should be, able to use the results from
> present studies to improve the software routines.
>
> <snip>> How does the software know which signals are reflected?
>
> <snip>
>
> I don't know for sure but I can guess. If the software sees a
> continuing trend from a specific satellite and then receives a series or
> sequence of signals that vary from that norm, the software might exclude
> that input from the fix determination process or reduce a weighting
> factor if there is such a thing.
>
> WAAS adds the feedback from the ground reference stations to the fix
> evaluation and determination process so if you already had a fix, it
> should get better, no? I know nothing about the actual software
> routines used so I'm purely guessing here.
>
> Jack
>
> --
> Jack Erbes in Ellsworth, Maine, USA - jackerbes at adelphia dot net
> (also receiving email at jacker at midmaine.com)- Hide quoted text -
>
> - Show quoted text -- Hide quoted text -
>
> - Show quoted text -
Okay ... so let's cut to the chase...
for a given unit - how would you calibrate it in terms of accuracy -
to determine for yourself - regardless of antenna or WAAS or any
claims made by the manufacturer or perceived by yourself - what
procedure could you do to assure yourself that you have some idea of
what it really is (or is not)?
BTW - accuracy per se is not a concept if your unit is not getting a
signal... at all...
You could have the most accurate unit built in the entire world - but
if it has a crappy antenna - what good is it?
| |
| laura halliday 2007-05-12, 3:33 pm |
| On May 11, 1:21 pm, nemo <gnu...@gmail.com> wrote:
> I am not interested in comparing the SiRF III to the SiRF II. I am
> talking about the choice of SiRF III vs. other current units.
>
> Regardless, my original point was about a unit that I thought did not
> provide WAAS capability. It does provide that capability, the maker
> just makes it very hard to make use of it.
Sounds ike the real problem is Bluetooth.
Seriously, though, if the device doesn't perform to your
expectations (or to published specifications), why don't
you return it to the vendor and get one that does?
Through my employers I have access to a couple of
reference locations that were surveyed to within a few
centimeters, as part of an earlier GPS-related project,
so I have no difficulty telling if my GPS is accurate or not.
One current project I'm doing uses gpsOne assisted
GPS. This is a can of worms that makes SiRF III and
Bluetooth look like a walk in the park. :-)
Laura Halliday VE7LDH "Que les nuages soient notre
Grid: CN89mg pied a terre..."
ICBM: 49 16.05 N 122 56.92 W - Hospital/Shafte
| |
| Jack Erbes 2007-05-12, 3:33 pm |
| Larry G wrote:
<snip>
> Okay ... so let's cut to the chase...
>
> for a given unit - how would you calibrate it in terms of accuracy -
> to determine for yourself - regardless of antenna or WAAS or any
> claims made by the manufacturer or perceived by yourself - what
> procedure could you do to assure yourself that you have some idea of
> what it really is (or is not)?
By going to know locations and comparing my GPSR's results to the
location. You can use the control data and monuments shown on various
USGS (and NOAA?) maps and charts. A little more info here:
http://mapserver.maptech.com/mapser...ntrol_Data.html
And I would look for National Geodetic Survey marks. You can find them
from the links on this page:
http://www.ngs.noaa.gov/
Looking for survey marks is pastime or hobby for some people. And I
could see where it could be both very interesting and rewarding one.
Maybe one of the folks that knows more about it will chime in here.
> BTW - accuracy per se is not a concept if your unit is not getting a
> signal... at all...
> You could have the most accurate unit built in the entire world - but
> if it has a crappy antenna - what good is it?
They are about the right size for a paperweight.
GPS is here forever. The allure of knowing where you are has a mystical
allure to nearly all of us. It is more addictive than the Internet. I
could live without the Internet.
Jack
--
Jack Erbes in Ellsworth, Maine, USA - jackerbes at adelphia dot net
(also receiving email at jacker at midmaine.com)
| |
| David L. Wilson 2007-05-12, 3:33 pm |
|
"Jack Erbes" <jackerbes@adelphia.net> wrote in message
news:4645fd81$0$1512
8$4c368faf@roadrunne
r.com...
> Larry G wrote:
> <snip>
....[color=darkred]
> By going to know locations and comparing my GPSR's results to the
> location. You can use the control data and monuments shown on various
> USGS (and NOAA?) maps and charts. A little more info here:
In reality this is generally not practical. To characterize the accuacy of
a GPS, one cannot do it but making short visits to known location. What is
required is atleast a couple days of measurement (and specs are based on 30
days). In fact, one learns more by taking a couple days at an unknown
point and looking at that distribution about the mean position. This will
tell you more about the accuracy as a distribution that visits (relatively
brief) to reference locations with a consumer GPS.
Report of accuracy from both a single short visit to a reference point or by
the reported accuracy in the display are really of no value.
| |
| Jack Erbes 2007-05-12, 10:33 pm |
| David L. Wilson wrote:
<snip>
>
> In reality this is generally not practical. To characterize the accuacy of
> a GPS, one cannot do it but making short visits to known location.
You have to be joking, right?
I am a retired Navy cryptologist and also spent about 15-20 years
working as a machinist. I know what anal retentive means, I've seen in
numerous forms. It is possible cross the line from a useful discussion
of accuracy into the realm of endless debate for no useful purpose.
I am talking about going to a monument and at known location with a
handheld consumer grade GPS, holding it directly over or setting it on
the crossed lines on the bronze marker, letting it set for a few minutes
or so, punching the mark button, and then going home and pondering the
results.
Something else I might consider trying is to approach the monument at a
walking pace, hold the GPSR over the monument immediately on arrival,
and creating a mark.
Those are the kind of tests that would relate to the way I use a GPS
receiver.
> What is
> required is atleast a couple days of measurement (and specs are based on 30
> days). In fact, one learns more by taking a couple days at an unknown
> point and looking at that distribution about the mean position. This will
> tell you more about the accuracy as a distribution that visits (relatively
> brief) to reference locations with a consumer GPS.
You are talking about something being done for some other reasons or
purpose than I am. Wouldn't you want to do it again for every
conceivable kind of weather condition? And in every season? I know a
little about radio wave propagation and things that affect it. A couple
of days is only going to be a start.
> Report of accuracy from both a single short visit to a reference point or by
> the reported accuracy in the display are really of no value.
I think you are arguing just for the sake of the argument. A single
short visit to a precisely known location will tell most of us what we
want to know about the accuracy of out handheld, consumer grade, GPS
receivers.
Jack
| |
| David L. Wilson 2007-05-12, 10:33 pm |
|
"Jack Erbes" <jacker@midmaine.com> wrote in message
news:46463894$0$9961
$4c368faf@roadrunner
.com...
> David L. Wilson wrote:
> <snip>
>
> You have to be joking, right?
>
> I am a retired Navy cryptologist and also spent about 15-20 years working
> as a machinist. I know what anal retentive means, I've seen in numerous
> forms. It is possible cross the line from a useful discussion of accuracy
> into the realm of endless debate for no useful purpose.
>
> I am talking about going to a monument and at known location with a
> handheld consumer grade GPS, holding it directly over or setting it on the
> crossed lines on the bronze marker, letting it set for a few minutes or
> so, punching the mark button, and then going home and pondering the
> results.
So am I and I am not joking at all. The error correlation time is of
several minutes and in addition you would only be sample a small sample of
the satellite configuration. This is not anal retentative but the view of a
PhD mathematician who taught statistics for over 10 years ans with a lot of
experience onGPS accuracy.
( http://www.erols.com/dlwilson/gps.htm )
The numbers you obtain will not be significant for any conclusion. One guy
doing it later will get and answer different than another. I have seen guys
sit there for 15 minutes and claim there GPS has 1 m accuracy when it does
not--they had luck of the draw.
> I think you are arguing just for the sake of the argument. A single short
> visit to a precisely known location will tell most of us what we want to
> know about the accuracy of out handheld, consumer grade, GPS receivers.
Far from it. It will lead many to mistaken beliefs and has in the past.
--
David L. Wilson AC4IU
dwilson314@comcast.net
| |
|
| On May 12, 6:58 pm, Jack Erbes <jac...@midmaine.com> wrote:
> I think you are arguing just for the sake of the argument. A single
> short visit to a precisely known location will tell most of us what we
> want to know about the accuracy of out handheld, consumer grade, GPS
> receivers.
I know what you are saying, but that sort of measurement will not be
measuring the GPS error, it will be measuring one particular set of
errors at the time of test. It will tell you little about the
inherent error of that particular intrument. Say you take a
measurement and get an error of 40 feet? The next day you test
another device and get an error of 2 feet. What did that tell you?
Pretty much nothing.
There are lots of sources of errors in GPS measurements. Some vary in
non-correllated ways, other are more consistent and vary very slowly.
To see the sum total of all the errors you have to take many
measurements over a period of time and determine the distribution. It
is that distribution that characterizes the ability of the instrument
to measure a location, not any one error measurement.
I can see this very clearly using a couple of different tools. If you
can hook your GPS up to a PC, you can use USAPhotoMaps, GPS NMEA
Monitor or even the SiRFDemo program (if you have a SiRF III unit) to
get a plot in 2D of all fixes for as long as you run the test. I have
done this for hours (not days) and typically it shows a *very* wide
range of measurements (up to 100 foot/30 meters radius or more), but
with most close to a central point (inside the 10 or even 5 meter
circle). That is why the accuracy is rated for 50% or 95% of the
fixes to be within the specified distance of the "true" location.
| |
|
|
"nemo" <gnuarm@gmail.com> wrote in message
news:1178817300.013508.63750@o5g2000hsb.googlegroups.com...
: BTW, NMEA GPS Monitor is a *GREAT* programfor the PC!!! It not only
: provides this excellent communications mode, it displays a lot of data
: (NMEA only of course, read the name) and is FREE! It is not available
: for PDAs or I would just use that on the palm. It is not open source
: or I would port it!!!
It (NMEA GPS Monitor) seems to only support the Serial interface and
not USB?
| |
| Alan White 2007-05-13, 4:33 am |
| On Sat, 12 May 2007 17:58:48 -0500, Jack Erbes <jacker@midmaine.com>
wrote:
>I am talking about going to a monument and at known location with a
>handheld consumer grade GPS, holding it directly over or setting it on
>the crossed lines on the bronze marker, letting it set for a few minutes
>or so, punching the mark button, and then going home and pondering the
>results.
Which will tell you very little except the recorded position at the time
you punched the button.
David is quite correct.
--
Alan White
Mozilla Firefox and Forte Agent.
Twenty-eight miles NW of Glasgow, overlooking Lochs Long and Goil in Argyll, Scotland.
Webcam and weather:- http://windycroft.gt-britain.co.uk/weather
| |
| Larry G 2007-05-13, 7:33 am |
| On May 12, 1:41 pm, laura halliday <marsga...@hotmail.com> wrote:
> On May 11, 1:21 pm, nemo <gnu...@gmail.com> wrote:
>
>
>
> Sounds ike the real problem is Bluetooth.
>
> Seriously, though, if the device doesn't perform to your
> expectations (or to published specifications), why don't
> you return it to the vendor and get one that does?
>
> Through my employers I have access to a couple of
> reference locations that were surveyed to within a few
> centimeters, as part of an earlier GPS-related project,
> so I have no difficulty telling if my GPS is accurate or not.
>
> One current project I'm doing uses gpsOne assisted
> GPS. This is a can of worms that makes SiRF III and
> Bluetooth look like a walk in the park. :-)
>
> Laura Halliday VE7LDH "Que les nuages soient notre
> Grid: CN89mg pied a terre..."
> ICBM: 49 16.05 N 122 56.92 W - Hospital/Shafte
I haven't had my 76csx very long but I noticed when I leave it hooked
up to my PC that it is able to maintain contact with the
satellites ... and it ... creates a track log ... which, when one
looks at it.. it appears that the unit has been "walking" around.
So the unit is laying down tracks.. each one with a lat/long .. ...
and it varies... and if I leave it on for a few days... the track
looks a lot ... as I said.. that the unit has been "walking" ....
so that was part of my thinking with respect to asking the calibration
question.
If the unit itself .. generates a "mark" in track mode.. one.. every
second or so (not sure what the interval reallyis)... the lat/long
"mark" that it generates for the track log is ... not the same.. but
different...
The last time I looked.. it appeared that you could draw an imaginary
circle around the track and the circle would be 50 foot or less in
diameter....
I assume this is "normal" for the 76csx... but to be honest.. even
though I've used the track log capability on the basic etrex and the
Rino.. I've not noticed this before.
and I am clueless about how other Garmin models and other brand GPS
function with respect to this... so that's why I asked the calibration
question.
| |
| David L. Wilson 2007-05-13, 7:33 am |
|
"Larry G" <gross.larry@gmail.com> wrote in message
news:1179050567.116937.176410@w5g2000hsg.googlegroups.com...
> On May 12, 1:41 pm, laura halliday <marsga...@hotmail.com> wrote:
[color=darkred]
> I assume this is "normal" for the 76csx... but to be honest.. even
> though I've used the track log capability on the basic etrex and the
> Rino.. I've not noticed this before.
It is normal.
Also, s problem with looking at plotted track logs is that one cannot see
the points that are on top of points. One tends to notice more the rare
extreme errors and notice less the higher probabilites of small errors.
| |
|
| On May 13, 1:18 am, "Sam" <nos...@nospam.com> wrote:
> "nemo" <gnu...@gmail.com> wrote in message
>
> news:1178817300.013508.63750@o5g2000hsb.googlegroups.com...
> : BTW, NMEA GPS Monitor is a *GREAT* programfor the PC!!! It not only
> : provides this excellent communications mode, it displays a lot of data
> : (NMEA only of course, read the name) and is FREE! It is not available
> : for PDAs or I would just use that on the palm. It is not open source
> : or I would port it!!!
>
> It (NMEA GPS Monitor) seems to only support the Serial interface and
> not USB?
Yes, it looks that way. With a bluetooth interface, the bluetooth
drivers emulate a serial port. Does USB do the same thing?
| |
|
| Just another update on what I have found with the Holux 240 SiRF III
receiver use of SBAS that I figured out with some advice from Dennis
Gr=F6ning. Seems that the SiRFDemo program won't display the actual
corrections from the SiRF III receivers, but they are calculated and
are sent over the comm port.
Dennis suggested that I turn message 27 off (DGPS Status) and turn
message 29 on (Nav Lib DGPS Data). I had tried this before, but saw no
difference in the dsiplay, which when MID 27 is used shows a constant
14.27 meter correction for all SVs.
Looking at the logged commands I saw that MID 29 does indeed show that
corrections are enabled and that different corrections are shown for
different SVs. However, I did not see a correcton for *every* SV being
used in the fix.
If I restart SiRFDemo with message 27 disabled and 29 enabled, nothing
shows up in the DGPS view window as if SiRFDemo does not report
message 29. I believe in one of the GPSPassion forum threads, Carl
from SiRF said that the SiRF III does not report corrections even when
they are calculated and applied. I guess they *are* reported, but not
with MID 27.
If you want full info on how to use SiRFDemo to control and monitor a
bluetooth GPS receiver, go to http://www.gpspassion.com/forumsen/topic.asp?=
TOPIC_ID=3D83158&whichpage=3D1
and check out both pages. That is the thread I will be updating as I
find more info.
| |
| laura halliday 2007-05-13, 3:33 pm |
| On May 13, 4:54 am, "David L. Wilson" <dwilson...@comcast.net> wrote:
> "Larry G" <gross.la...@gmail.com> wrote in message
>
> news:1179050567.116937.176410@w5g2000hsg.googlegroups.com...
>
>
> It is normal.
>
> Also, s problem with looking at plotted track logs is that one cannot see
> the points that are on top of points. One tends to notice more the rare
> extreme errors and notice less the higher probabilites of small errors.
I've done animations of fix data to visualize dynamic
behaviour, which show this clearly. They can be entertaining,
though this probably means I need to get out more. :-)
There are probably better ways to do it, but I generate
individual frames as TIFF images with a C program,
convert to JPEG and add some annotations with
ImageMagick, then stitch the series of frames in to
a movie with MJPEGTools.
Laura Halliday VE7LDH "Que les nuages soient notre
Grid: CN89mg pied a terre..."
ICBM: 49 16.05 N 122 56.92 W - Hospital/Shafte
| |
| Jack Erbes 2007-05-13, 3:33 pm |
| David L. Wilson wrote:
<snip>
>
>
> Far from it. It will lead many to mistaken beliefs and has in the past.
>
Your statement that "Report (sic) of accuracy from both a single short
visit to a reference point or by the reported accuracy in the display
are really of no value." is what I do not agree with. Without a GPS you
knew nothing about your location. You got a GPS and you had some idea
of your location. You went to a monument, and you had some level of
comparison. There are still ambiguities and questions but you knew
more, it was helpful.
Nothing is exact and nothing is useless. Let's move on to how many
angels can dance on the head of a pin.
Or how about English grammar as a comparison? You statement has some
inaccuracies in that area. Does make the statement say nothing? Does
it take away the easily determined context? No. Just because higher
levels of accuracy can be attained, it does not make a poorer result
meaningless.
Jack
| |
| Kevin 2007-05-13, 10:33 pm |
| On Sat, 12 May 2007 21:48:46 -0400, "David L. Wilson"
<dwilson314@comcast.net> wrote:
>
>"Jack Erbes" <jacker@midmaine.com> wrote in message=20
> news:46463894$0$9961
$4c368faf@roadrunner
.com...
>
>So am I and I am not joking at all. The error correlation time is of=20
>several minutes and in addition you would only be sample a small sample =
of=20
>the satellite configuration. This is not anal retentative but the view =
of a=20
>PhD mathematician who taught statistics for over 10 years ans with a lot=
of=20
>experience onGPS accuracy.
>( http://www.erols.com/dlwilson/gps.htm )
>
How many years experience do you have with web link accuracy? Hint,
check the last three links on the page that you cite.
--=20
E-mail: kevjay@attglobal.net
| |
| David L. Wilson 2007-05-13, 10:33 pm |
|
"Jack Erbes" <jacker@midmaine.com> wrote in message
news:46476171$0$4717
$4c368faf@roadrunner
.com...
> David L. Wilson wrote:
>
> <snip>
> Your statement that "Report (sic) of accuracy from both a single short
> visit to a reference point or by the reported accuracy in the display are
> really of no value." is what I do not agree with. Without a GPS you knew
> nothing about your location. You got a GPS and you had some idea of your
> location. You went to a monument, and you had some level of comparison.
> There are still ambiguities and questions but you knew more, it was
> helpful.
>
> Nothing is exact and nothing is useless. Let's move on to how many angels
> can dance on the head of a pin.
The point is that you really do not know more. The error is a probability
distribution. You end up measuring a few values which tell you essentially
nothing about the probability distribution because the sample is too small
to be meaningful. The only use of it is to make wrong conclusions, which
posting to here in the past show from doing such.
> Or how about English grammar as a comparison? You statement has some
> inaccuracies in that area. Does make the statement say nothing? Does it
> take away the easily determined context? No. Just because higher levels
> of accuracy can be attained, it does not make a poorer result meaningless.
*The stated public spec is more meaningful that a short visit measurement
(which thus gives no additional real information other than your unit is
probably not defective)*.
The main problem is that the error is a probability distribuition--not a
single number. One needs to take a large enough same to see that
distribution or large enough to characterize its parameter(s). Anything
else is an exercise in not understanding statistical nature of the error.
If one understands this, one knows the short time measurements are of little
value. As I have said before, long term measurement at a non-reference
point will actually tell you more about your error than a short term
measurement at a reference point as it will get at the dsitribution of an
essentially unbiased measurement.
| |
| Jack Erbes 2007-05-13, 10:33 pm |
| Kevin wrote:
> On Sat, 12 May 2007 21:48:46 -0400, "David L. Wilson"
> <dwilson314@comcast.net> wrote:
>
> How many years experience do you have with web link accuracy? Hint,
> check the last three links on the page that you cite.
I don't have any experience with web link accuracy. And you attributed
that quote to me but I did not post it, some one else did.
Jack
| |
| Jack Erbes 2007-05-13, 10:33 pm |
| David L. Wilson wrote:
> "Larry G" <gross.larry@gmail.com> wrote in message
> news:1179050567.116937.176410@w5g2000hsg.googlegroups.com...
>
>
> It is normal.
>
> Also, s problem with looking at plotted track logs is that one cannot see
> the points that are on top of points. One tends to notice more the rare
> extreme errors and notice less the higher probabilites of small errors.
MapSource is not a very good program for looking at tracks with. It is
okay for seeing your travel on the Garmin mapping and it does an okay
job on calculating speeds and elapsed times and distances.
Beyond that, you might want to try some of the third party applications
that are much more capable at distinguishing individual track points and
other details. There are several good ones here, both freewares and
sharewares:
http://gpsinformation.net/#3rdparty
My personal favorite is GPS Utility:
http://www.gpsu.co.uk/
Jack
| |
| Jack Erbes 2007-05-13, 10:33 pm |
| Larry G wrote:
<snip>
> I haven't had my 76csx very long but I noticed when I leave it hooked
> up to my PC that it is able to maintain contact with the
> satellites ... and it ... creates a track log ... which, when one
> looks at it.. it appears that the unit has been "walking" around.
>
> So the unit is laying down tracks.. each one with a lat/long .. ...
> and it varies... and if I leave it on for a few days... the track
> looks a lot ... as I said.. that the unit has been "walking" ....
> so that was part of my thinking with respect to asking the calibration
> question.
That "walking" is the distance accumulated from one track to the next.
It is a rare (if ever) occurrence to have two consecutive trackpoints
with the same location so they are always going to be some distance apart.
> If the unit itself .. generates a "mark" in track mode.. one.. every
> second or so (not sure what the interval reallyis)... the lat/long
> "mark" that it generates for the track log is ... not the same.. but
> different...
It generates a track point in track memory. And if you have that option
turned on it will save all track points collected to a *.gpx file on the
microSD card. There are user settings for the collection rate and you
can collect trackpoints either on time intervals or on distances.
> The last time I looked.. it appeared that you could draw an imaginary
> circle around the track and the circle would be 50 foot or less in
> diameter....
Is that the blue (cyan?) circle seen around the location marker? If so,
that represents the EPE (Estimated Position Error, the value can be seen
on the Satellite page) as a circular area. In theory, you could be
anywhere within that circle. You'll notice that the size of the circle
changes with the EPE value. It is particularly noticeable when the GPS
is first started up because the in initial fix will be a large number
and it will decrease pretty rapidly.
> I assume this is "normal" for the 76csx... but to be honest.. even
> though I've used the track log capability on the basic etrex and the
> Rino.. I've not noticed this before.
It is normal. The 60/76 Cx/CSx (and all the "x" series models that have
microSD cards) are really very capable as far as track recording. If
you have the Log Track To Data Card check box checked, every single
track point collected by the unit will be saved to a *.gpx file on the
microSD card. There is one file for each the GPS is turned on and there
will be separate track segments for each no fix to fix transition or
each time the GPS is turned on.
The *.gpx files cannot be seen from the GPS. You can to download them
to your hard drive (from the GPS in the USB Mass Storage Mode or from
the microSD card in a reader) and open them with MapSource or with any
of a number of third party applications that will read *.gpx files.
I really like the fact that I can set a high trackpoint collection rate
and, let the GPS overwrite the 10,000 trackpoint memory capacity when it
is full, and still recover all track data from the *.gpx file. And as
long as there is space on the data card, I can go for long periods of
time without having to download tracks. The *.gpx files contain the
location, date and time, and elevation for each track point.
> and I am clueless about how other Garmin models and other brand GPS
> function with respect to this... so that's why I asked the calibration
> question.
I don't know of any consumer grade GPSRs that are calibrated by the
users. In the pure sense of the word "calibrate", there are no options
for the user to collect or study the GPS data and made any changes to
the receiver to improve or change performance based on their
observations. The fix locations are derived from the satellite data and
the routines in the software, what you see is what you get at any given
moment. There are some calibration requirements and options for the
models with internal compasses and barometers but not for the GPS function.
Jack
| |
| Jack Erbes 2007-05-14, 7:33 am |
| Alan White wrote:
> On Sat, 12 May 2007 17:58:48 -0500, Jack Erbes <jacker@midmaine.com>
> wrote:
>
>
> Which will tell you very little except the recorded position at the time
> you punched the button.
>
> David is quite correct.
Okay, let's say I visited one day and got a two foot error, then visited
a second time on another day and got a 40 foot error. Is there no way
to express that result within this holy grail of "probability
distribution" that you guys worship at?
When you say "the sample is too small to be meaningful" who or what is
it too small for? Is there no room in this science for coarse results,
that can be bettered with persistent and continued study?
If you were thrown out of an airplane (lets assume a parachute) over a
horizon to horizon expense of featureless wasteland and had a GPS
receiver with you, would being able to determine your approximate
location with it be meaningless? No.
If if your GPS had a waypoint to an oasis that was within walking
distance and gave you a range and bearing to it, would you discard that
because you knew that your position had an error of hundreds of feet or
more?
If you took two steps toward the location and your GPS receiver's
batteries died would you lay down and await rescue or death?
And if I said that that my location, relative to the visited feature,
was determined accurate to within a few hundred feet or less, would that
not be true?
Jack
--
Jack Erbes in Ellsworth, Maine, USA - jackerbes at adelphia dot net
(also receiving email at jacker at midmaine.com)
| |
|
| On May 14, 7:51 am, Jack Erbes <jacker...@adelphia.net> wrote:
> Okay, let's say I visited one day and got a two foot error, then visited
> a second time on another day and got a 40 foot error. Is there no way
> to express that result within this holy grail of "probability
> distribution" that you guys worship at?
>
> When you say "the sample is too small to be meaningful" who or what is
> it too small for? Is there no room in this science for coarse results,
> that can be bettered with persistent and continued study?
>
> If you were thrown out of an airplane (lets assume a parachute) over a
> horizon to horizon expense of featureless wasteland and had a GPS
> receiver with you, would being able to determine your approximate
> location with it be meaningless? No.
>
> If if your GPS had a waypoint to an oasis that was within walking
> distance and gave you a range and bearing to it, would you discard that
> because you knew that your position had an error of hundreds of feet or
> more?
>
> If you took two steps toward the location and your GPS receiver's
> batteries died would you lay down and await rescue or death?
>
> And if I said that that my location, relative to the visited feature,
> was determined accurate to within a few hundred feet or less, would that
> not be true?
I think you don't get what the others are discussing. They are
talking about how to determine the accuracy of the GPS. No one is
saying that a reading by the GPS is meaningless. They are saying that
a small number of measurements with a GPS does not tell you anything
about the *accuracy* of the GPS. The displayed EPE value of the GPS
also does not really tell you anything about the *real* error of the
measurement because it is not measureing that. The EPS is telling you
something about the condition of the constellation and how many SVs
you are receiving which has an impact on the accuracy, but does not
determine it.
Of course any reading from a GPS tells you a lot more than *nothing*
about your position. We are simply trying to understand how much
accuracy a given receiver has and how best to measure that accuracy.
| |
| Alan White 2007-05-14, 10:33 am |
| On 14 May 2007 06:14:39 -0700, nemo <gnuarm@gmail.com> wrote:
>I think you don't get what the others are discussing. They are
>talking about how to determine the accuracy of the GPS. No one is
>saying that a reading by the GPS is meaningless. They are saying that
>a small number of measurements with a GPS does not tell you anything
>about the *accuracy* of the GPS. The displayed EPE value of the GPS
>also does not really tell you anything about the *real* error of the
>measurement because it is not measureing that. The EPS is telling you
>something about the condition of the constellation and how many SVs
>you are receiving which has an impact on the accuracy, but does not
>determine it.
>
>Of course any reading from a GPS tells you a lot more than *nothing*
>about your position. We are simply trying to understand how much
>accuracy a given receiver has and how best to measure that accuracy.
Exactly.
--
Alan White
Mozilla Firefox and Forte Agent.
Twenty-eight miles NW of Glasgow, overlooking Lochs Long and Goil in Argyll, Scotland.
Webcam and weather:- http://windycroft.gt-britain.co.uk/weather
| |
| Alan White 2007-05-14, 10:33 am |
| On Mon, 14 May 2007 07:51:10 -0400, Jack Erbes <jackerbes@adelphia.net>
wrote:
>...
>When you say "the sample is too small to be meaningful"
>...
I didn't.
Jack, you are one of the most helpful people that I've ever encountered
on Usenet but this time I think that you've misunderstood what we're
talking about. We're not arguing that the position information is
meaningless but that the assessment of the accuracy of the GPSR achieved
with one position reading is meaningless.
It is quite correct to say that to assess the accuracy of a GPSR you
need to average its recorded position for thirty days and then compare
that averaged position with the actual position as accurately surveyed.
For everyday use, any of the positions recorded over that thirty day
period would be sufficient for you to know where you were.
--
Alan White
Mozilla Firefox and Forte Agent.
Twenty-eight miles NW of Glasgow, overlooking Lochs Long and Goil in Argyll, Scotland.
Webcam and weather:- http://windycroft.gt-britain.co.uk/weather
| |
| Dan Anderson 2007-05-14, 12:33 pm |
| laura halliday wrote:
[snip]>
> I've done animations of fix data to visualize dynamic
> behaviour, which show this clearly. They can be entertaining,
> though this probably means I need to get out more. :-)
Yes, get out and put that navigation tool to another use!
> There are probably better ways to do it, but I generate
> individual frames as TIFF images with a C program,
> convert to JPEG and add some annotations with
> ImageMagick, then stitch the series of frames in to
> a movie with MJPEGTools.
--
Dan
(email change 2001 to 2004)
(www.gpsmap.net)
| |
| Dan Anderson 2007-05-14, 3:33 pm |
| Larry G wrote:
> I haven't had my 76csx very long but I noticed when I leave it hooked
> up to my PC that it is able to maintain contact with the
> satellites ... and it ... creates a track log ... which, when one
> looks at it.. it appears that the unit has been "walking" around.
>
> So the unit is laying down tracks.. each one with a lat/long .. ...
> and it varies... and if I leave it on for a few days... the track
> looks a lot ... as I said.. that the unit has been "walking" ....
>
> so that was part of my thinking with respect to asking the calibration
> question.
>
> If the unit itself .. generates a "mark" in track mode.. one.. every
> second or so (not sure what the interval reallyis)... the lat/long
> "mark" that it generates for the track log is ... not the same.. but
> different...
>
> The last time I looked.. it appeared that you could draw an imaginary
> circle around the track and the circle would be 50 foot or less in
> diameter....
>
> I assume this is "normal" for the 76csx... but to be honest.. even
> though I've used the track log capability on the basic etrex and the
> Rino.. I've not noticed this before.
>
> and I am clueless about how other Garmin models and other brand GPS
> function with respect to this... so that's why I asked the calibration
> question.
Most handheld receivers generate a position every second. As time passes,
the position calculated will vary. If you keep recording the position
over a long period of time, you will get a curve something like a "bell
curve". This is the probability distribution that has been talked about.
The width of the distribution is what the "accuracy" is all about.
The wider the distribution curve, the lower the accuracy.
In case you don't know... You set the the track logging parameters to
"Distance", "Time", or "Auto". If "Time" is chosen the interval is
listed and is user settable.
--
Dan
(email change 2001 to 2004)
(www.gpsmap.net)
| |
| Dan Anderson 2007-05-14, 3:33 pm |
| Jack Erbes wrote:
> Larry G wrote:
> <snip>
>
>
> That "walking" is the distance accumulated from one track to the next.
> It is a rare (if ever) occurrence to have two consecutive trackpoints
> with the same location so they are always going to be some distance apart.
[snip]
The "walking" is simply the unit generating different position values
as time passes. Larry has the receiver sitting on a desk that's
not moving.
The recorded trackpoints will usually be some distance apart
only if track logging is set to "Auto" or "Distance". If
track logging is set to "Time" and the receiver is stationary,
then successive same locations may be common (somewhat
dependent on time interval).
For some data I've collected recording locations at a one
second rate and a fairly good view of the sky:
Of 54,805 trackpoints,
92% were the same as the previous location
7% changed one of the two horizontal coordinates from
the previous location
< 1% changed both horizontal coordinates
from the previous location
For some data I've collected recording locations at a one
second rate and a poor view of the sky:
Of 75,874 points,
71% were the same as the previous location
23% changed one of the two horizontal coordinates from
the previous location
6% changed both horizontal coordinates
from the previous location
The standard deviation of the above points is 19 feet.
Number of Times Coordinates Changed:
Easting Northing 2d Distance
Count Percent Count Percent Count Percent
Less than 1: 63332 83.5 83 61870 81.5 82 53959 71.1 71
By 1 meter: 10477 13.8 97 11336 14.9 96 17966 23.7 95
By 2 meters: 1568 2.1 99 1549 2.0 99 2544 3.4 98
By 3 meters: 108 0.1 99 625 0.8 99 713 0.9 99
By 4 meters: 68 0.1 100 79 0.1 99 197 0.3 99
By 5 meters: 40 0.1 100 82 0.1 100 66 0.1 99
By 6 meters: 32 0.0 100 25 0.0 100 51 0.1 100
By 7 meters: 18 0.0 100 43 0.1 100 31 0.0 100
By 8 meters: 17 0.0 100 18 0.0 100 29 0.0 100
By 9 meters: 19 0.0 100 22 0.0 100 32 0.0 100
10 or more: 194 0.3 100 224 0.3 100 285 0.4 100
--
Dan
(email change 2001 to 2004)
(www.gpsmap.net)
| |
| laura halliday 2007-05-14, 3:33 pm |
| I've just uploaded an animation of GPS fixes to
<http://www.youtube.com/watch?v=K2tAZdahHk4>
The MPEG encoding shows some artifacts, but it will
do for 5 minutes work. :-)
Laura Halliday VE7LDH "Que les nuages soient notre
Grid: CN89mg pied a terre..."
ICBM: 49 16.05 N 122 56.92 W - Hospital/Shafte
| |
| Jack Erbes 2007-05-14, 10:33 pm |
| nemo wrote:
<snip>
> I think you don't get what the others are discussing. They are
> talking about how to determine the accuracy of the GPS. No one is
> saying that a reading by the GPS is meaningless. They are saying that
> a small number of measurements with a GPS does not tell you anything
> about the *accuracy* of the GPS. The displayed EPE value of the GPS
> also does not really tell you anything about the *real* error of the
> measurement because it is not measureing that. The EPS is telling you
> something about the condition of the constellation and how many SVs
> you are receiving which has an impact on the accuracy, but does not
> determine it.
>
> Of course any reading from a GPS tells you a lot more than *nothing*
> about your position. We are simply trying to understand how much
> accuracy a given receiver has and how best to measure that accuracy.
Okay, if that is what they are talking about now I'll drop out of the net.
I was replying to Larry G's "cut to the chase post" when he asked "what
procedure could you do to assure yourself that you have some idea of
what it really is (or is not)?"
I read that question to mean a "seat of the pants" or layman's kind of
procedure. Something that could be done by anyone that was curious
about their consumer grade GPS. So I said I would go to a monument or
other known location and see what result I got.
The ensuing posts lost track of that question, poo-poohed my suggestion,
and really did not offer any other procedure that goes to answering the
question that was asked.
They took it from trying to "have some idea of what it (the accuracy)
really is" to a need to try to "characterize the accuracy of a GPS" and
I don't think that was what was asked for.
I still consider what I suggested as practical and feel it would
contribute something to a person's having some feel for the accuracy of
the GPS receiver. And I did not agree that the test I suggested, and
the EPE value displayed by the receiver, were "really of no value".
Jack
| |
| Laurie Forbes 2007-05-14, 10:33 pm |
|
"nemo" <gnuarm@gmail.com> wrote in message
news:1179023417.369659.67040@e51g2000hsg.googlegroups.com...
> There are lots of sources of errors in GPS measurements. Some vary in
> non-correllated ways, other are more consistent and vary very slowly.
> To see the sum total of all the errors you have to take many
> measurements over a period of time and determine the distribution. It
> is that distribution that characterizes the ability of the instrument
> to measure a location, not any one error measurement.
>
> I can see this very clearly using a couple of different tools. If you
> can hook your GPS up to a PC, you can use USAPhotoMaps, GPS NMEA
> Monitor or even the SiRFDemo program (if you have a SiRF III unit) to
> get a plot in 2D of all fixes for as long as you run the test. I have
> done this for hours (not days) and typically it shows a *very* wide
> range of measurements (up to 100 foot/30 meters radius or more), but
> with most close to a central point (inside the 10 or even 5 meter
> circle). That is why the accuracy is rated for 50% or 95% of the
> fixes to be within the specified distance of the "true" location.
I've done just that (using SiRFDemo) - the results are instructive and as
you describe. Certainly spot readings or readings over a short period
convey little useful accuracy related information.
Laurie Forbes
| |
|
| On May 14, 2:42 pm, laura halliday <marsga...@hotmail.com> wrote:
> I've just uploaded an animation of GPS fixes to
> <http://www.youtube.com/watch?v=K2tAZdahHk4>
>
> The MPEG encoding shows some artifacts, but it will
> do for 5 minutes work. :-)
Interesting how youtube makes something simple into something
complex. Even if you have a dozen programs to play mpeg files, you
can't watch them on youtube if you don't have flash.
| |
| Alan White 2007-05-15, 4:33 am |
| On Mon, 14 May 2007 17:39:22 -0500, Jack Erbes <jacker@midmaine.com>
wrote:
>...
>I still consider what I suggested as practical and feel it would
>contribute something to a person's having some feel for the accuracy of
>the GPS receiver. And I did not agree that the test I suggested, and
>the EPE value displayed by the receiver, were "really of no value".
The specification for the accuracy of a consumer grade GPSR is better
than 15 metres r.m.s. for 95% of the time over a thirty day period. How
does recording one position at a known point contribute something to a
person's having some feel for the accuracy of the GPSR if at that time
the user is in the 5% when the accuracy is worse than 15 metres?
--
Alan White
Mozilla Firefox and Forte Agent.
Twenty-eight miles NW of Glasgow, overlooking Lochs Long and Goil in Argyll, Scotland.
Webcam and weather:- http://windycroft.gt-britain.co.uk/weather
| |
| Larry G 2007-05-15, 7:33 am |
| On May 13, 10:14 pm, Jack Erbes <jac...@midmaine.com> wrote:
> David L. Wilson wrote:
>
>
>
>
> MapSource is not a very good program for looking at tracks with. It is
> okay for seeing your travel on the Garmin mapping and it does an okay
> job on calculating speeds and elapsed times and distances.
>
> Beyond that, you might want to try some of the third party applications
> that are much more capable at distinguishing individual track points and
> other details. There are several good ones here, both freewares and
> sharewares:
>
> http://gpsinformation.net/#3rdparty
>
> My personal favorite is GPS Utility:
>
> http://www.gpsu.co.uk/
>
> Jack
Jack - I had looked at this and others but they did not appear to use
maps....
right now I'm using NG TOPO... backroad and mid Atlantic ..... how
would the NG
software compare to the GPS utility software?
| |
| Larry G 2007-05-15, 7:33 am |
| On May 14, 1:43 pm, Dan Anderson <danderson2...@hotmail.com> wrote:
> Larry G wrote:
>
>
>
>
>
>
>
>
>
>
>
>
>
> Most handheld receivers generate a position every second. As time passes,
> the position calculated will vary. If you keep recording the position
> over a long period of time, you will get a curve something like a "bell
> curve". This is the probability distribution that has been talked about.
> The width of the distribution is what the "accuracy" is all about.
> The wider the distribution curve, the lower the accuracy.
>
> In case you don't know... You set the the track logging parameters to
> "Distance", "Time", or "Auto". If "Time" is chosen the interval is
> listed and is user settable.
>
> --
> Dan
> (email change 2001 to 2004)
> (www.gpsmap.net)- Hide quoted text -
>
> - Show quoted text -
very helpful... explanations... thank you
I noticed the options for auto, time, and distance... and my
inclination was to leave it alone... not knowing what effect the time
or distance setting would have on it's operations in the field.
| |
| Jack Erbes 2007-05-15, 7:33 am |
| Larry G wrote:
<snip>
> Jack - I had looked at this and others but they did not appear to use
> maps....
>
> right now I'm using NG TOPO... backroad and mid Atlantic ..... how
> would the NG
> software compare to the GPS utility software?
I've not used the NG software. But with GPS Utility I can open the USGS
topo *.tif (tiff?) images and display my tracks and waypoints over them.
The USGS topos are a free download and are the same maps you have in
the NG package I think. They may add some features to the NG package
but I think the detail is the same. GPSU even comes with some sample
files and a good tutorial for calibrating maps. I don't remember that
the map images can be edited from GPSU though.
I bought GPSU and have found it meets all my needs for looking at data
and moving it around and I've received a great number of upgrades with
no additional fees. The developer, Alan Murphy, is an occasional
visitor here and very interactive with users. He seems to have the goal
of keeping GPSU as compatible and up to date with as possible. The
trial version is worth looking at.
Ozi Explorer is another good bet for those that want to work on their
own map editing. I think it has quite a bit of versatility as far as
editing the map image files. I did not ever actually use Ozi Explorer
to do that though, I was just looking at tracks and data as temporary
overlays, not as edited into the map. I did buy Ozi Explorer-CE and use
it on a iPAQ PDA for marine and topo navigation until I got my Magellan
Meridians equipped with marine and topo mapping.
I am still waiting for a good rugged waterproof handheld (like my 76Cx)
that has an open architecture for mapping.
Jack
--
Jack Erbes in Ellsworth, Maine, USA - jackerbes at adelphia dot net
(also receiving email at jacker at midmaine.com)
| |
| Jack Erbes 2007-05-15, 12:33 pm |
| Larry G wrote:
<snip>
> I noticed the options for auto, time, and distance... and my
> inclination was to leave it alone... not knowing what effect the time
> or distance setting would have on it's operations in the field.
I've not done anything where I wanted to collect data points on a timed
basis. Whichever method you use, you can always see both the time and
distance values when you look at the track later, That applies to the
*.gpx file tracks, *not* the Saved Tracks. With the saved tracks some
points may be lost to the 700 point maximum. And you don't get the date
and time data on Saved Tracks, just the lat/long and elevation.
With a 10,000 point track memory you can collect points for a 100 mile
journey at the .01 mile setting. That is where mine is set most of the
time. With the Overwrite when full option on and the Log Track to Data
Card box checked, if you go more than 100 miles the GPS will start over
writing the oldest points but all points collected will still be in that
day's *.gpx file on the microSD card.
When the track log is full and overwriting, the percentage used just
stays at 99% until you clear it. I'm not sure but I think if you set it
to stop writing when full (to preserve the track in track memory for
example), it still continues to record points to the *.gpx file.
I seldom need 100 points per mile for anything but the next setting (.1
miles) will only give you 10 points per mile and that is a little
coarser than I like for some things.
I love tracks, the 76Cx has about as good a track recording capability
as you can get. It is almost as good as the Meridian Color that it
replaced.
Jack
--
Jack Erbes in Ellsworth, Maine, USA - jackerbes at adelphia dot net
(also receiving email at jacker at midmaine.com)
| |
| laura halliday 2007-05-15, 3:33 pm |
| On May 14, 9:20 pm, nemo <gnu...@gmail.com> wrote:
> On May 14, 2:42 pm, laura halliday <marsga...@hotmail.com> wrote:
>
>
>
> Interesting how youtube makes something simple into something
> complex. Even if you have a dozen programs to play mpeg files, you
> can't watch them on youtube if you don't have flash.
While I consider Flash to be the enemy of all that is
free and right and holy, not to mention Just Plain Wrong,
I accept that YouTube have way more disk space
and net bandwidth than I do.
Please do not invent issues where there are none.
Laura Halliday VE7LDH "Que les nuages soient notre
Grid: CN89mg pied a terre..."
ICBM: 49 16.05 N 122 56.92 W - Hospital/Shafte
| |
| Jack Yeazel 2007-05-15, 10:33 pm |
|
Jack Erbes wrote:
> I was replying to Larry G's "cut to the chase post" when he asked "what
> procedure could you do to assure yourself that you have some idea of
> what it | | |