How is RoyalTek Sapphire GPS Doing?

hkelley · 09-28-2002

It's been several months since a number of you wrote about your intentions to try the tiny RoyalTek Mouse GPS...Sapphire (RGM2100) version. - http://www.royaltek.com/proditem.as...ouse\gindex.asp and http://www.tri-m.com/products/royaltek/sapphire.html

I would be interested in any reaction to the unit's effectiveness and performance. How well does it receive and how well does it interface with MP2002.

subterphuge · 09-28-2002

Just got mine a couple days ago. Works GREAT with both MP2002 and the Netstumbler 0.3.30 under Win98SE. (Not at the same time yet, as I haven't bothered trying to install a COM port sharing program...)

But even on the cloudy day that I tested it out for some time with it sitting on my dash, it was quite accurate... within a block of my actual location at all times. (Weather's been shitty, haven't had a chance to test it on a clear sky yet...) But I even got a fairly close fix (within a couple blocks) of my house when running it INDOORS!

It doesn't take more than a couple seconds to get an initial fix either. But what I like best about it, other than its low-profile, is that it requires no independent power source. Just the PS2 port on your laptop with the serial version, or a USB port for that version.

What I need to do is go on a test drive and keep a mental tab of exactly where I'm seeing APs then upload the results to WiGLE and see just how accurate they are, see if its any different than what Maps and Streets or my own recollection is telling me.

But yeah, I'd highly recommend it, esp. if you don't want a bulky stand-alone hand-held GPS sliding all around your dash and attracting attention, or having to get a separate antenna for it. The Sapphire has a mag-mount so you CAN put it on your roof if you want, but I didn't find much difference in performance on the roof versus the dash. Again, this was on a cloudy day, so I haven't had a chance to test it out at peak performance yet. But from what I've seen so far even in less-than-ideal conditions, its performs well enough for my purposes.

I got mine from here:

http://www.sapphire-gps.com/

Good company, they ship right away. I didn't even bother doing the $10 extra overnight delivery and it still arrived in 3 days via USPS.

lincomatic · 09-29-2002

the sapphire is a nice little unit as i mentioned in the garmin gps thread. fast TTFF and none of the heading jitter that some other SiRF-based units have. i don't know if the units have been redesigned or not, but i did have a few problems with my units...
after leaving them on my dash for several months, some of them take forever to get a fix, and have greatly reduced sensitivity. mine were obtained directly from the manufacturer, however.

so my advice is when you're not using it, put it some place a little cooler like under the dash instead of on top. the leadtek SiRF-I-based unit is much bigger (the one that looks like a square box) but it's the hands down king of sensitivity. inside my office w/ the blinds closed it can get 5-6 satellites even when sitting on a desk 7ft from the closest window. most other GPS's i've tried can't even get a fix unless i put them right next to the window. the TTFF is a bit slower w/ the SiRF-I's (sapphire is SiRF-II), but i still like this unit the best for overall performance and reliability. none of the 10 units that i have have had any problems.

hkelley · 09-29-2002

Thank you both for the helpful information. Am I reading the specs correctly and understanding Subterphuge that the accuracy of the Sapphire unit is approx 75-ft. (25 meters)? How does this stack up with other GPS units of similar cost and size?

The Magellan/GPS Companion for my Palm 505 claimes >9 ft. and seems about what I am experiencing. I realize that specs don't always equate to reality but I would like your thoughts on what a reasonable accuracy range should be for Stumbling.

lincomatic · 09-29-2002

personally, i think acquisition performance and sensitivity are the most impt performance metrics for a GPS. if you have to wait 5min every time you get turn on a GPS before it gets the initial fix, then it wastes too much of your time. also, the more satellites the GPS can keep a lock on while driving thru trees, urban canyons, and other obstructions, the more likely you aren't going to completely lose the lock on your position temporarily every time your sky view is obstructed. if you do lose your lock on the satellites, another impt thing is how fast you can get it back after you pass the obstructions. this is called a "hot start," which the SiRF units can usually do in <5-10sec, even if you cut the power off for 1/2 hr (like when you go into a store and come back).

the manufacturer's ratings in accuracy are doodoo as far as i'm concerned, because i've seen such wildly different specs on different units using the same hardware, and so many conditions beyond your control such as the geometry of the satellites relative to each other and you come into play to determine the "accuracy" at any given moment when you're driving. also, don't forget that the GPS is only spitting out a fix every 1sec, and that the fix was calculated using after the data were acquired, so when moving, your accuracy is further decreased by your velocity (and when using a garmin etrex in NMEA mode or a NAVMAN GPS sleeve, you get a fix only once every 2sec).

WAAS support definitely improves accuracy, but if i remember correctly, there are currently only 2 satellites and it's hard to get a clear sky view to them at all times.

as far as stumbling is concerned, the position of the AP is just an estimate based on the signal strength trajectory, so it adds so much error that talking about GPS error relative to stumbling is moot.

just my 2cents.

MarkSC · 09-29-2002

(From the Garmin site)

WAAS consists of approximately 25 ground reference stations positioned across the United States that monitor GPS satellite data. Two master stations, located on either coast, collect data from the reference stations and create a GPS correction message. This correction accounts for GPS satellite orbit and clock drift plus signal delays caused by the atmosphere and ionosphere. The corrected differential message is then broadcast through one of two geostationary satellites, or satellites with a fixed position over the equator. The information is compatible with the basic GPS signal structure, which means any WAAS-enabled GPS receiver can read the signal.

Quote:

Originally posted by lincomatic

WAAS support definitely improves accuracy, but if i remember correctly, there are currently only 2 satellites and it's hard to get a clear sky view to them at all times.

as far as stumbling is concerned, the position of the AP is just an estimate based on the signal strength trajectory, so it adds so much error that talking about GPS error relative to stumbling is moot.

just my 2cents. [/b]

lincomatic · 09-29-2002

Quote:

Originally posted by MarkSC
(From the Garmin site)

The corrected differential message is then broadcast through one of two geostationary satellites, or satellites with a fixed position over the equator.

thanks for the info...i was too lazy to look it up.

right, they're over the equator, which puts them towards the horizon in NA, and there are only 2 of them. probably great in the plains states, but not so hot in urban or hilly areas.

hkelley · 09-29-2002

I am curious as to which mapping software you have interfaced your RoyalTek to. And, do you have a preference?

subterphuge · 09-30-2002

Well, MP2002 comes free with it and seems to be comprehensive enough, including even some dirt roads south of town in a military staging area that I didn't expect to be included. But I did notice that the dirt roads seemed to be a bit out of scale... at one point MP2002 had me suddenly travelling a good quarter-mile off and parallel to the road that I was on, out in the open desert, but quickly put me right back track once I hit pavement, so I attribute that to MP2002's inaccuracy, not the GPS.

In fact, on that very trip when I was stopped at a railroad crossing, the thing had me pin-pointed exactly. I was just sitting there saying to myself: "Target Acquired," and stabbing around desperately on the touchpad, looking for the command in MP2002 to "Launch Missiles".

My next project though is to get GPS going under Linux, and all its associated goodies. Kismet has an interesting approach for the war-driver in mind... if you're within range of an open AP, you can just pull up MapQuest and get your exact coordinates from there, based off of your current GPS data.

But I wouldn't mind nuking the Windoze partition completely and for good, which has been really tempting lately with Linux having all the superior tools out first for such new technology for a change. Marius needs to make an open-source GUI port of Netstumbler for Linux... either that or someone just needs to make a decent graphical front-end for Kismet.

subterphuge · 09-30-2002

Quote:

Thank you both for the helpful information. Am I reading the specs correctly and understanding Subterphuge that the accuracy of the Sapphire unit is approx 75-ft. (25 meters)? How does this stack up with other GPS units of similar cost and size?

Yes, the manual for the Sapphire (RGM-2000) actually states 25-meters accuracy in its official spec, and thats about what I've been experiencing even with an overcast.

Quote:

The Magellan/GPS Companion for my Palm 505 claimes >9 ft. and seems about what I am experiencing. I realize that specs don't always equate to reality but I would like your thoughts on what a reasonable accuracy range should be for Stumbling.

I think 25 meters (a quarter of a football field's length) is well within a reasonable margin of error. Its not like you're ever going to get within 9 feet of an AP anyway, without physically breaking and entering. So to me, in regards to accurate war-driving, what it boils down to is this: based on the recorded map reading (where the AP shows up on a generated map from your GPS logs), can you find the AP within a few seconds of driving from the spot its supposed to be? If the answer is YES, then its close enough.

Now what really interests me though, is the possibility of software triangulation. The idea basically consists of taking various readings of signal strengths and matching it up with the GPS data, and extrapolating a much more accurate idea of where the AP itself is located, not just the GPS position of your vehicle when you first initially detected the AP. This would really help localize the AP regardless of accuracy of the GPS, since an average would reduce the margin of error considerably.

hkelley · 09-30-2002

It's worth a try...I've ordered the RoyalTek Sapphire (USB) and will run it on an IBM ThinkPad X20 alongside my Megellan/Palm 505 to see how the two perform in real life.

I have enjoyed the discussions by the three of you concerning what's really important...acquisition and reacquisition times; sensitivty versus geographic accuracy.

These kinds of discussions make for a good board as compared to some of the inane techie oneupsmanship that usually populate the subject folders.

hkelley · 10-10-2002

I have been running the tiny RoyalTek Sapphire now for about a week with great results. It acquires quickly and tracks without a skip. I have measured its accuracy against several other professional level GPS units and found its readings to be consistently within about 3 to 4 meters even though the advertised spec is about 25 meters.

Question to Lincomatic: Did I read you correctly that MapPoint 2002 doesn't handle USB inputs. MS Streets comes with the Sapphire but the NS1 conversion seems clumbsy. I would like to use the same software for tracking as well as plotting. Like most new laptops mine doesn't provide for RS232 or PS2 thus I am bound by USB. Your thoughts?

lincomatic · 10-10-2002

Quote:

Originally posted by hkelley
Question to Lincomatic: Did I read you correctly that MapPoint 2002 doesn't handle USB inputs. MS Streets comes with the Sapphire but the NS1 conversion seems clumbsy. I would like to use the same software for tracking as well as plotting. Like most new laptops mine doesn't provide for RS232 or PS2 thus I am bound by USB. Your thoughts?

sorry, i really don't have any experience w/ USB GPS's, and i've been using them almost exclusively on pocketpc's, not desktops.
doesn't windows just give you a virtual com port to open when you use a USB GPS?

btw, another of my sapphires is suddenly having problems (this is the 4th one w/ problems) getting a quick fix even after being unplugged for a short period. i'm beginning to suspect that the problem is caused by failure of the backup battery rather than by prolonged heat exposure.
i'll look into it when i have more time and let you guys know what's up w/ that.

itsnotme · 11-23-2002

Quote:

Originally posted by lincomatic
sorry, i really don't have any experience w/ USB GPS's, and i've been using them almost exclusively on pocketpc's, not desktops.
doesn't windows just give you a virtual com port to open when you use a USB GPS?

btw, another of my sapphires is suddenly having problems (this is the 4th one w/ problems) getting a quick fix even after being unplugged for a short period. i'm beginning to suspect that the problem is caused by failure of the backup battery rather than by prolonged heat exposure.
i'll look into it when i have more time and let you guys know what's up w/ that.

Considering I've got a sapphire, I'd be interested in hearing about this problem and how to fix it if there is indeed a way to do so. Maybe you can unscrew it and find the battery inside it? I dont have the sapphire with me at the moment but I think I recall there being a screw on the bottom of it..

call-911 · 01-09-2003

lincomatic,
Any thoughts on this new Onyx model?

I'm hesitant to pick one of the original Sapphires if there's something better out....

I'm either going to go with one of these mouse type GPS units or a Garvin Rhino 120... Anyone have any experience with the 120?

09-28-2002	#1 (permalink)
hkelley Registered Member Join Date: May 2002 Location: Florida Posts: 98	How is RoyalTek Sapphire GPS Doing? It's been several months since a number of you wrote about your intentions to try the tiny RoyalTek Mouse GPS...Sapphire (RGM2100) version. - http://www.royaltek.com/proditem.as...ouse\gindex.asp and http://www.tri-m.com/products/royaltek/sapphire.html I would be interested in any reaction to the unit's effectiveness and performance. How well does it receive and how well does it interface with MP2002.

09-29-2002	#3 (permalink)
lincomatic Squaaawk! WiFi! WiFi! Join Date: Apr 2002 Location: Tinsel Town Posts: 1,682	the sapphire is a nice little unit as i mentioned in the garmin gps thread. fast TTFF and none of the heading jitter that some other SiRF-based units have. i don't know if the units have been redesigned or not, but i did have a few problems with my units... after leaving them on my dash for several months, some of them take forever to get a fix, and have greatly reduced sensitivity. mine were obtained directly from the manufacturer, however. so my advice is when you're not using it, put it some place a little cooler like under the dash instead of on top. the leadtek SiRF-I-based unit is much bigger (the one that looks like a square box) but it's the hands down king of sensitivity. inside my office w/ the blinds closed it can get 5-6 satellites even when sitting on a desk 7ft from the closest window. most other GPS's i've tried can't even get a fix unless i put them right next to the window. the TTFF is a bit slower w/ the SiRF-I's (sapphire is SiRF-II), but i still like this unit the best for overall performance and reliability. none of the 10 units that i have have had any problems. __________________ ~lincomatic

09-29-2002	#5 (permalink)
lincomatic Squaaawk! WiFi! WiFi! Join Date: Apr 2002 Location: Tinsel Town Posts: 1,682	personally, i think acquisition performance and sensitivity are the most impt performance metrics for a GPS. if you have to wait 5min every time you get turn on a GPS before it gets the initial fix, then it wastes too much of your time. also, the more satellites the GPS can keep a lock on while driving thru trees, urban canyons, and other obstructions, the more likely you aren't going to completely lose the lock on your position temporarily every time your sky view is obstructed. if you do lose your lock on the satellites, another impt thing is how fast you can get it back after you pass the obstructions. this is called a "hot start," which the SiRF units can usually do in <5-10sec, even if you cut the power off for 1/2 hr (like when you go into a store and come back). the manufacturer's ratings in accuracy are doodoo as far as i'm concerned, because i've seen such wildly different specs on different units using the same hardware, and so many conditions beyond your control such as the geometry of the satellites relative to each other and you come into play to determine the "accuracy" at any given moment when you're driving. also, don't forget that the GPS is only spitting out a fix every 1sec, and that the fix was calculated using after the data were acquired, so when moving, your accuracy is further decreased by your velocity (and when using a garmin etrex in NMEA mode or a NAVMAN GPS sleeve, you get a fix only once every 2sec). WAAS support definitely improves accuracy, but if i remember correctly, there are currently only 2 satellites and it's hard to get a clear sky view to them at all times. as far as stumbling is concerned, the position of the AP is just an estimate based on the signal strength trajectory, so it adds so much error that talking about GPS error relative to stumbling is moot. just my 2cents. __________________ ~lincomatic

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

09-28-2002	#2 (permalink)
subterphuge /dev/war Join Date: Jul 2002 Location: On the Road, Where the WEP is Weak and the Signal is Strong Posts: 90	Just got mine a couple days ago. Works GREAT with both MP2002 and the Netstumbler 0.3.30 under Win98SE. (Not at the same time yet, as I haven't bothered trying to install a COM port sharing program...) But even on the cloudy day that I tested it out for some time with it sitting on my dash, it was quite accurate... within a block of my actual location at all times. (Weather's been shitty, haven't had a chance to test it on a clear sky yet...) But I even got a fairly close fix (within a couple blocks) of my house when running it INDOORS! It doesn't take more than a couple seconds to get an initial fix either. But what I like best about it, other than its low-profile, is that it requires no independent power source. Just the PS2 port on your laptop with the serial version, or a USB port for that version. What I need to do is go on a test drive and keep a mental tab of exactly where I'm seeing APs then upload the results to WiGLE and see just how accurate they are, see if its any different than what Maps and Streets or my own recollection is telling me. But yeah, I'd highly recommend it, esp. if you don't want a bulky stand-alone hand-held GPS sliding all around your dash and attracting attention, or having to get a separate antenna for it. The Sapphire has a mag-mount so you CAN put it on your roof if you want, but I didn't find much difference in performance on the roof versus the dash. Again, this was on a cloudy day, so I haven't had a chance to test it out at peak performance yet. But from what I've seen so far even in less-than-ideal conditions, its performs well enough for my purposes. I got mine from here: http://www.sapphire-gps.com/ Good company, they ship right away. I didn't even bother doing the $10 extra overnight delivery and it still arrived in 3 days via USPS.

09-29-2002	#4 (permalink)
hkelley Registered Member Join Date: May 2002 Location: Florida Posts: 98	Thank you both for the helpful information. Am I reading the specs correctly and understanding Subterphuge that the accuracy of the Sapphire unit is approx 75-ft. (25 meters)? How does this stack up with other GPS units of similar cost and size? The Magellan/GPS Companion for my Palm 505 claimes >9 ft. and seems about what I am experiencing. I realize that specs don't always equate to reality but I would like your thoughts on what a reasonable accuracy range should be for Stumbling.

09-29-2002	#8 (permalink)
hkelley Registered Member Join Date: May 2002 Location: Florida Posts: 98	I am curious as to which mapping software you have interfaced your RoyalTek to. And, do you have a preference?

09-30-2002	#9 (permalink)
subterphuge /dev/war Join Date: Jul 2002 Location: On the Road, Where the WEP is Weak and the Signal is Strong Posts: 90	Well, MP2002 comes free with it and seems to be comprehensive enough, including even some dirt roads south of town in a military staging area that I didn't expect to be included. But I did notice that the dirt roads seemed to be a bit out of scale... at one point MP2002 had me suddenly travelling a good quarter-mile off and parallel to the road that I was on, out in the open desert, but quickly put me right back track once I hit pavement, so I attribute that to MP2002's inaccuracy, not the GPS. In fact, on that very trip when I was stopped at a railroad crossing, the thing had me pin-pointed exactly. I was just sitting there saying to myself: "Target Acquired," and stabbing around desperately on the touchpad, looking for the command in MP2002 to "Launch Missiles". My next project though is to get GPS going under Linux, and all its associated goodies. Kismet has an interesting approach for the war-driver in mind... if you're within range of an open AP, you can just pull up MapQuest and get your exact coordinates from there, based off of your current GPS data. But I wouldn't mind nuking the Windoze partition completely and for good, which has been really tempting lately with Linux having all the superior tools out first for such new technology for a change. Marius needs to make an open-source GUI port of Netstumbler for Linux... either that or someone just needs to make a decent graphical front-end for Kismet.

09-30-2002	#11 (permalink)
hkelley Registered Member Join Date: May 2002 Location: Florida Posts: 98	It's worth a try...I've ordered the RoyalTek Sapphire (USB) and will run it on an IBM ThinkPad X20 alongside my Megellan/Palm 505 to see how the two perform in real life. I have enjoyed the discussions by the three of you concerning what's really important...acquisition and reacquisition times; sensitivty versus geographic accuracy. These kinds of discussions make for a good board as compared to some of the inane techie oneupsmanship that usually populate the subject folders.

10-10-2002	#12 (permalink)
hkelley Registered Member Join Date: May 2002 Location: Florida Posts: 98	I have been running the tiny RoyalTek Sapphire now for about a week with great results. It acquires quickly and tracks without a skip. I have measured its accuracy against several other professional level GPS units and found its readings to be consistently within about 3 to 4 meters even though the advertised spec is about 25 meters. Question to Lincomatic: Did I read you correctly that MapPoint 2002 doesn't handle USB inputs. MS Streets comes with the Sapphire but the NS1 conversion seems clumbsy. I would like to use the same software for tracking as well as plotting. Like most new laptops mine doesn't provide for RS232 or PS2 thus I am bound by USB. Your thoughts?

01-09-2003	#15 (permalink)
call-911 Registered Member Join Date: Dec 2002 Posts: 12	lincomatic, Any thoughts on this new Onyx model? I'm hesitant to pick one of the original Sapphires if there's something better out.... I'm either going to go with one of these mouse type GPS units or a Garvin Rhino 120... Anyone have any experience with the 120?