wxforecaster
Posts: 3
Joined: Mon May 08, 2006 11:14 am

Garmin 18 USB NMEA Decoder Failure

Sun Oct 08, 2006 9:03 pm

Johan,

I've had a recurring problem with V2.0.1 where GPSGate was repeatedly sending the same data infinitely. At first it appeared the software simply has an error in the decoder when the GPS device is locking onto satellites. However, I then tried several instances where I used nRoute and the Garmin USB to lock onto the satellites outside and THEN switched to GPSGATE. In this case, instead of sending the same bogus data repeatedly, it sends incorrect data -- both to nRoute and to my other GPS Serial applications. I'm well aware of the NMEA protocol and can tell you that your software is not converting the data from Garmin to NMEA correctly.

I did some testing and there's definitely a problem with your decoder/encoder to convert the data from Garmin to NMEA.

I have taken 3 snapshots of nRoute. The first is NOT using GPSGate with the software connected via USB. You can see a fully locked signal with great reception.

The second screen shot is a few seconds later after turning on GPSGate and switching the nRoute device over to GPSGate's COM1. Note the gross errors on Accuracy and satellite signal strength (even though this was the SAME EXACT data as before).

The third screen shot is after turning off GPSGate and switching back to the Garmin USB. You'll note the immediate similarities to the accurate data in the first image.

My settings are Virtual COM1/Garmin Emulator (Generic Device) and Virtual COM2. NOTE: Both COM1 and COM2 are receiving the same exact bad data, so this isn't an issue of filtering data for COM2. The screen shots above are with COM1 in use ONLY.

This may be overkill, but in the hopes that you finally won't blame the user and fix the product we paid for, here you go. [;)]

Evan

http://www.interwarn.com/images/nroute_usb1.jpg
<hr noshade size="1">
http://www.interwarn.com/images/nroute_gpsgate_2.jpg
<hr noshade size="1">
http://www.interwarn.com/images/nroute_usb_3.jpg

Andreas2
Posts: 108
Joined: Sun Feb 13, 2005 3:19 pm

RE: Garmin 18 USB NMEA Decoder Failure

Sun Oct 08, 2006 9:22 pm

Hi,

> GPSGate was repeatedly sending the same data infinitely
This sound like a bug. If I remember correctly, Garmin Que behaves poorly when Garmin packets stop arriving, and I believe that's the root of the problem. We'll look into it and try to come up with a better solution to this.

> nRoute and to my other GPS Serial applications
Are the other GPS applications NMEA apps or Garmin apps?

As for the errors in accuracy, we are aware of this. The reason that the problem remains is that Garmin has a proprietary algorithm for converting between Dilution of Position (DOP; used in NMEA) and Estimated Position Error (EPE; used in Garmin). This is not public information and we made a simple approximation that appears to be valid under certain conditions and invalid in others. While it may be possible to improve it, I don't think we'll be able to get an exact conversion for this.

The satellite strength again is a specification issue. The ranges of the signal quality isn't specified and thus, the satellite bars will constitute an approximation rather than an absolute. We'll look in to improving this approximation as well.

Regards,
Andreas Nilsson
Franson Support

wxforecaster
Posts: 3
Joined: Mon May 08, 2006 11:14 am

RE: Garmin 18 USB NMEA Decoder Failure

Mon Oct 09, 2006 3:07 am

The 'repeated' data I would agree is definitely a bug that causes the software to essentially fail when locking onto the satellites. I noticed that in the raw NMEA data that there were azimuths of 1700 and 1242, etc... (obviously they must be between 0 and 360). Hope that helps!

In the 2nd image above with GPSGATE, it looks like the satellite strength is OK but the "green vs. yellow" is not.

I would also of course say that the accuracy is a bug. Maybe it's off a decimal place?? I don't know.

wxforecaster
Posts: 3
Joined: Mon May 08, 2006 11:14 am

RE: Garmin 18 USB NMEA Decoder Failure

Wed Jan 03, 2007 4:45 pm

Johan,

I see you released V2.5, but no mention of fixing this known problem that's affecting a lot of us judging by the threads. Instead of continuing to add new features, how about first fixing those crippling bugs that render the software non-functional to those of us who paid good money for it. It's funny that prior to purchase you swore up and down that this device works fine (need I point to documentation that still exists on this site), and months later, still nothing. I have provided tons of documentation and screen shots to illustrate the decoding problems.

I respectfully ask that this is either fixed or I get a refund.

Evan

mrlenny
Posts: 2
Joined: Fri Mar 23, 2007 5:07 pm
Location: USA

RE: Garmin 18 USB NMEA Decoder Failure

Fri Mar 23, 2007 5:12 pm

We recently upgraded from GPSGate 2.0.72 to version 2.5 after finding out about the memory leak. However, now we have an accuracy issue as well. If I connect the Garmin GPS 18 USB without running GPSGate, the accuracy shown in nRoute is between 10 and 15 feet.

If I running the GPS 18 with GPSGate (which is needed in our situation as we run ArcPad), the accuracy is between 130 and 175 feet.

I uninstalled GPSGate 2.5 and reinstalled GPSGate 2.0 and the accuracy issue is gone, however the memory leak is back. I tried GPSGate 2.0.1 build 139 and it has the accuracy issues as well.

Will there be a fix for this? We are at a stand still right now and have 45 employees complaining about there maps not working correctly.

Thank you,
Leonard Williamson
GIS Coordinator
Arkansas Oklahoma Gas Corporation
lwilliamson@aogc.com

User avatar
johan
GpsGate
Posts: 15423
Joined: Wed Aug 04, 2004 10:40 pm
Location: Sweden
Contact: Website

RE: Garmin 18 USB NMEA Decoder Failure

Fri Mar 23, 2007 5:18 pm

The memory leak is only an issue when you have GpsGate running for a long time configued to use Garmin USB as input, without a Garmin USB plugged in. Or do you have another scenario for a memory leak?

Have you invsetigated in detail what the complain is about? What does "not working correctly" mean? Why do you think it relates to GpsGate?

Please also note that even though the accuracy number may differ, the acual position is the exact same.

Regards,
Johan

Franson Support

mrlenny
Posts: 2
Joined: Fri Mar 23, 2007 5:07 pm
Location: USA

RE: Garmin 18 USB NMEA Decoder Failure

Fri Mar 23, 2007 5:26 pm

The employees are using the antenna and mapping software all day long (8 plus hours). We discovered the memory leak when laptops were locking up due to no available resources around 4:30 each afternoon. Once I upgraded to GPSGate 2.5 this problem went away.

I have only heard of the problems second hand. I have heard the maps are "jumping around". I suspect, yet have not proven that this problem is two-fold: nRoute has an option to lock the GPS position to the roads. With the poor accuracy we are now getting with GPSGate 2.5, the problem is with nRoute switching the position between roads.

Perhaps the accuracy is the same as you mention (10 feet versus 175 feet) however, we did not have these issues with GPSGate 2.0.72.

I am spending today looking into this further. I was able to determine the accuracy issue was linked to GPSGate 2.5 yesterday.

Thank you,
Leonard Williamson
GIS Coordinator
Arkansas Oklahoma Gas Corporation
lwilliamson@aogc.com

User avatar
johan
GpsGate
Posts: 15423
Joined: Wed Aug 04, 2004 10:40 pm
Location: Sweden
Contact: Website

RE: Garmin 18 USB NMEA Decoder Failure

Fri Mar 23, 2007 5:28 pm

Ok, do some research on this to get the details. Then email us at support @ franson . com, and we'll work this out.

Regards,
Johan

Franson Support

Return to “GpsGate Splitter - How to setup and use”