Jump to content

HCx inaccuracy after upgrade to 2.60/2.60


mikser

Recommended Posts

I wouldn't call that "drift"… not in the sense that others have expressed, where their positions are off by hundreds of feet.

 

Yours looks close to normal variation.

 

It's also not fair to summarize any error as the sum of two errors, as you're showing with your 75' spread. Your tracks are showing 50-55' position error at their worst, and within 5-10' at their best. But even your worst (the area you indicated) seems to be momentary, and reel back in, which could easily be due to any combination of satellite coverage/interference + GPSr orientation + foliage + terrain + your speed.

Edited by speede541
Link to comment

I created another thread about a week ago "Vista HCx solution...". I reported that my all-but-useless Vista HCx was finally accurate in tracking caches after upgrading to 2.70 / 2.60 AND doing a hard reset. While I don't claim to have any info on speed, stopped time or distance, I can truly say that the accuracy is now incredible. Over two weekends, all caches (five) have been within several feet of accurate. One that was 121 feet off last Saturday was reading one foot when held over the cache. Two small park roads crossed at a 90 degree angle were right on (w/ NA NT 2008) at the maximun zoom. The track log w/ the unit held in my hand (hanging by my side most of the time for the five mike hike in fairly dense woods w/ 99.5% tree cover) was right on. There were no drop outs for the first time in months and out/back routes overlapped each other. Accuracy read 17 feet +/- 5 feewt for most of the hike, but it didn't seem to have a negative impact on the readings.

Link to comment

Blindmantoo, I guess many of us are interested in achieving the kind of accuracy you are now reporting with 2.70/2.60 + reset.

 

Take a look at the tracklog image I posted at the bottom of Page 7 of this thread. It shows the same route repeated four times. Would your tracklog show the four tracks not up to 75 ft apart as mine does, but all four clustered within several feet of each other? I'm trying to get a feel for how your setup would perform under conditions similiar to mine.

 

Thanks,

 

Don

Link to comment

Yesterday I downgraded the chipset software from 2.60 to 2.30 using this link:

 

http://www.garmin.com/software/GPSChipsetT...onFile__230.rgn

 

Does anyone have any suggestions about how to find chipset 2.40, 2.50 and 2.60?

 

You can easy get region file V2.60.

First downgrade to V2.30, than erase all in dir c:\documents and settings\(your name)\Local Settings\Temp (first enable "show hidden files" somehow!)

Now upgrade your GPS through webupdater. When it ends, you will have new dir in c:\documents and settings\(your name)\Local Settings\Temp - "Garmin updates" or similar. In that dir you will have file Update.rgn - It is Region file V2.60 :D

 

I'm not at home and can't attach my vista HCx region file till tomorow. But if you can wait, I'll send it tomorow...

---------------------------------

---------------------------------

 

IMPORTANT!!! If any of you have Vista HCx upgraded at home to V2.40 or 2.50 and NOT upgraded to region file V2.60 yet, there is a possibility that you have that earliler version of regon files stored in c:\documents and settings\(your name)\Local Settings\Temp !!!! Look at your dirs, mailby some of you still have older Update.rgn file. Attach your update files and I'll compare it, but say if it is for Vista HCx or Legend HCX!! I have only Vista...

 

Regards,

L.A.

Link to comment

Since the 2.70/2.60 upgrade I've now actually experienced the big 0.1mile drift that people have reported here on two separate occasions. Standing within 100 feet of a cache as reported by others, mine would be a steady 0.1mile+ off. Powering the unit down bought it back to the same reading as others.

Link to comment

Here is a tracklog image of a recent walk with 2.70/2/30. (including hard reset).

 

Is the 75ft. spread normal GPS error, or "drift"? With 2.70/2.60 I also see similar spreads, although also "drifts" of 150ft or more. Is one of our issues here a matter of how "normal" errors are defined?

 

230-Test-Track-9-6-2008b.jpg

 

I apologize for re-posting this image. As most of you probably know, we have two active threads here on the Forum dealing with the identical topic.

 

Don

Link to comment

Here is a tracklog image of a recent walk with 2.70/2/30. (including hard reset).

 

Is the 75ft. spread normal GPS error, or "drift"? With 2.70/2.60 I also see similar spreads, although also "drifts" of 150ft or more. Is one of our issues here a matter of how "normal" errors are defined?

u

 

I apologize for re-posting this image. As most of you probably know, we have two active threads here on the Forum dealing with the identical topic.

 

Don

 

Although the error here is large, I don't know if it qualifies as the "bug" initially described in these discussions. It appears that the unit corrects it self over time. It could be simply caused by satellite locations or atmospheric phenomena. Remember the symptoms initially described are that even when good satellite reception is restored the error remains until the unit is power cycled, at which point the error drops in one large step.

Link to comment

I haven't followed this entire thread, but I keep seeing it and reading about the errors some are seeing.

 

Is it possible this might be related to having the routing set to CAR or TRUCK or something instead of PEDESTRIAN? These settings seem to affect by bearing pointer a bit -- I'm using a Legend HCx with software version 2.40, GPS SW version 2.40 and Bravo version B.

 

I haven't tried any upgrades as of yet, I'm watching this thread for a while first! :D

Link to comment
Is it possible this might be related to having the routing set to CAR or TRUCK or something instead of PEDESTRIAN?

 

no

 

According to Garmin tech support: " ... it appears ... that either the unit lost reception or you may have experienced what we call multipath where these units being High Sensitive GPS devices are continually trying to update the satellite information may have picked up a satellite bouncing off something as small as a leaf, a hill, a tree, a path or sidewalk, whatever it has hit."

Link to comment
Is it possible this might be related to having the routing set to CAR or TRUCK or something instead of PEDESTRIAN?

 

no

 

According to Garmin tech support: " ... it appears ... that either the unit lost reception or you may have experienced what we call multipath where these units being High Sensitive GPS devices are continually trying to update the satellite information may have picked up a satellite bouncing off something as small as a leaf, a hill, a tree, a path or sidewalk, whatever it has hit."

 

Let's be clear- the bug we are talking about is that the unit doesn't recover its accuracy after what ever the problem that caused it is removed. It requires a power cycle. Any of these highly sensitive units can get confused during multipath.

Link to comment
Let's be clear- the bug we are talking about is that the unit doesn't recover its accuracy after what ever the problem that caused it is removed. It requires a power cycle. Any of these highly sensitive units can get confused during multipath.

 

So, what, exactly, is multipath? Is the implication that the unit is simultaneously plotting itself into more than one location?

Link to comment

Let's be clear- the bug we are talking about is that the unit doesn't recover its accuracy after what ever the problem that caused it is removed. It requires a power cycle. Any of these highly sensitive units can get confused during multipath.

 

Here is an example of what I've been calling "drift". The return track on the top is well over two hundred feet away from the bottom (and correct) track and puts me into the reservoir (no I can't walk on water). At first, I was going to post this image as an example of the drift returning to normal without a power cycle. However, now that I look at the tracks, I can't remember whether or not I power-cycled just before the upper track returns to normal. The Mapsource track properties shows that one segment to the right of the scale moving 345ft. in five seconds. Is this the result of a power-cycle? I don't recall.

 

Drift-Example-7-2-08.jpg

 

Don

Link to comment
Let's be clear- the bug we are talking about is that the unit doesn't recover its accuracy after what ever the problem that caused it is removed. It requires a power cycle. Any of these highly sensitive units can get confused during multipath.

 

So, what, exactly, is multipath? Is the implication that the unit is simultaneously plotting itself into more than one location?

 

Kind of like a house of mirrors where your image appears to be in many places at once; in the case of a GPS signal, the signal is bouncing off of nearby objects (rocks, trees, houses, buildings, you) and then to the unit, thus the signals are taking 'multipaths.' I wouldn't think multipath is a fault of a GPS unit, how could it tell if the signal is straight from the satellite or bouncing off of something like a rock wall (multipath)? It seems the more sensitive the unit the more likely this will happen.

 

BTW; I've upgraded to v2.7 and my GPS has been all over the place, sometimes up to 500ft! Yesterday, I had to restart at every geocache just to get close. When driving on mountain roads, the unit showed me deep in the forest! Very frustrating to say the least.

Link to comment
Let's be clear- the bug we are talking about is that the unit doesn't recover its accuracy after what ever the problem that caused it is removed. It requires a power cycle. Any of these highly sensitive units can get confused during multipath.

 

So, what, exactly, is multipath? Is the implication that the unit is simultaneously plotting itself into more than one location?

 

Kind of like a house of mirrors where your image appears to be in many places at once; in the case of a GPS signal, the signal is bouncing off of nearby objects (rocks, trees, houses, buildings, you) and then to the unit, thus the signals are taking 'multipaths.' I wouldn't think multipath is a fault of a GPS unit, how could it tell if the signal is straight from the satellite or bouncing off of something like a rock wall (multipath)? It seems the more sensitive the unit the more likely this will happen.

 

BTW; I've upgraded to v2.7 and my GPS has been all over the place, sometimes up to 500ft! Yesterday, I had to restart at every geocache just to get close. When driving on mountain roads, the unit showed me deep in the forest! Very frustrating to say the least.

 

Pictures of my tracks sooo far off

log%201.jpglog%202.jpg

Link to comment
Kind of like a house of mirrors where your image appears to be in many places at once; in the case of a GPS signal, the signal is bouncing off of nearby objects (rocks, trees, houses, buildings, you) and then to the unit, thus the signals are taking 'multipaths.' I wouldn't think multipath is a fault of a GPS unit, how could it tell if the signal is straight from the satellite or bouncing off of something like a rock wall (multipath)? It seems the more sensitive the unit the more likely this will happen.

 

Thanks for the explanation. If that is the issue, it sounds like a) there might not be an easy fix for it and :D some of the criticism directed at Garmin might be overly harsh.

 

By the way, the Garmin tech person reminded me that, for proper functioning, the unit should be horizontal and away from the body. If there is a problem with multipath, as you described it, I suppose carrying the unit in a front shirt pocket might exacerbate the problem. However, hiking with the unit stored in the top of a fanny or back pack would make it difficult to periodically check the unit while walking.

Link to comment
Kind of like a house of mirrors where your image appears to be in many places at once; in the case of a GPS signal, the signal is bouncing off of nearby objects (rocks, trees, houses, buildings, you) and then to the unit, thus the signals are taking 'multipaths.' I wouldn't think multipath is a fault of a GPS unit, how could it tell if the signal is straight from the satellite or bouncing off of something like a rock wall (multipath)? It seems the more sensitive the unit the more likely this will happen.

 

Thanks for the explanation. If that is the issue, it sounds like a) there might not be an easy fix for it and :D some of the criticism directed at Garmin might be overly harsh.

 

By the way, the Garmin tech person reminded me that, for proper functioning, the unit should be horizontal and away from the body. If there is a problem with multipath, as you described it, I suppose carrying the unit in a front shirt pocket might exacerbate the problem. However, hiking with the unit stored in the top of a fanny or back pack would make it difficult to periodically check the unit while walking.

 

It seems maybe a little of both but mostly justified frustration/anger towards Garmin; I can't judge very well based on my experience since my Vista Hcx is the only GPS I've owned and therefore can’t compare to other units. However, based on what others have said here there is an issue that is specific to the eTrex series and/or perhaps the chip set. I do believe that we are in the process of going through a troubleshoot cycle to isolate the issue; 'we' being the owners because Garmin is apparently claiming that there is not an issue. Others here have much more knowledge about these issues.

 

I agree about the shirt pocket; my tracks shown in the pictures in my previous post where taken with the unit mostly in my pocket and under pretty good tree cover. It’s been posted here that the body can block or at least interfere a lot and that just makes logical sense as well. I’m not sure how much affect there is with the setting vertically. Honestly I don’t have the time either to go back and rerun the routes with the unit horizontal and mounted away from my body. I’m sure that corporations such as Garmin bank on the fact that consumers have limited time to troubleshoot these issues and therefore will try to make us go away by claiming that we are somehow using the unit incorrectly.

 

One thing I do know is that I want a GPS unit that will work well without much effort. I’m not talking about basic setup and programming such as for a new cell phone but trying to solve the issues noted under here are very different then the performance claims made by the manufacturer when I purchase the unit. In my mind, my Vista Hcx is far enough outside that box to be labeled defective. BTW, Re Garmin's response: It’s a personal GPS unit – here the heck am I suppose to carry it on my person that is level and away from my body?

Edited by _Shaddow_
Link to comment

Try carrying the GPSr in one hand and covering up the antenna portion with the other hand. When I do this I see the location error reading on the satellite page go up, but when I watch the track being drawn on the map page, I am unable to notice any clear change in the track direction.

 

This tells me that while the body does attentuate the received signal some, it doesn't appear to have a substantial effect on the track accuracy. And it would certainly not seem to be the cause of the drift error in the hundreds of feet that we are observing.

 

Don

Link to comment

Try carrying the GPSr in one hand and covering up the antenna portion with the other hand. When I do this I see the location error reading on the satellite page go up, but when I watch the track being drawn on the map page, I am unable to notice any clear change in the track direction.

 

This tells me that while the body does attentuate the received signal some, it doesn't appear to have a substantial effect on the track accuracy. And it would certainly not seem to be the cause of the drift error in the hundreds of feet that we are observing.

 

Don

 

Seems like the difference of accuracy vs precision.

 

I understand the little circle on the GPS to be a measure of accuracy of incoming signals or basically how much the signal is 'bouncing around.' Precision on the other hand is about where to place little circle on the map. If that circle isn't located anywhere anywhere near where I actually am located then the accuracy means nothing - it could be that the unit is accurately placing me in the wrong location.

Link to comment

Try carrying the GPSr in one hand and covering up the antenna portion with the other hand. When I do this I see the location error reading on the satellite page go up, but when I watch the track being drawn on the map page, I am unable to notice any clear change in the track direction.

 

This tells me that while the body does attentuate the received signal some, it doesn't appear to have a substantial effect on the track accuracy. And it would certainly not seem to be the cause of the drift error in the hundreds of feet that we are observing.

 

Don

 

Seems like the difference of accuracy vs precision.

 

I understand the little circle on the GPS to be a measure of accuracy of incoming signals or basically how much the signal is 'bouncing around.' Precision on the other hand is about where to place little circle on the map. If that circle isn't located anywhere anywhere near where I actually am located then the accuracy means nothing - it could be that the unit is accurately placing me in the wrong location.

 

Ok... sorry, today must be reverse day: I have accuracy and precision mixed up. But the concept is the same; that circle is often determined to be in the wrong location. So the unit is averaging to find the center of the circle but all of the points must be up to 500ft off (the error shown on my tracks).

accuracy%20vs%20precision.jpg

Link to comment

Although I am sure they contribute to the problem, multipath issues, by themselves, typically do not result in the kind of drift that has been reported here (and in the corresponding Colorado thread). Typically, multipath issues result in random excursions about your actual position. When you are standing still, this shows up as a "spider web" or "ball of yarn" track which is centered on your actual location. When you are moving, this typically shows up as higher EPE and a track which is somewhat displaced from reality -- but still consistent with the reported EPE.

 

Symptoms of the drift problem include a very high EPE (in the neighborhood of 100 feet) and a cumulative error which is much larger than can be accounted for by that EPE (500 feet or more). Typically, the recorded track slowly diverges from the actual track, with the error becoming larger and larger over time and distance along the track. Even when reception conditions improve, the unit takes a very long time to return to the correct location, if it ever does. Power-cycling the unit causes it to immediately revert to the correct location. The fact that this occurs is the best proof that the problem is not due (entirely) to multipath issues. If the problem was a simple multipath error, the position reported before and after the power cycle would be approximately the same.

Link to comment

 

Thanks, Tom, for one of the most accurate (or should I say "precise" :rolleyes: ) summaries of the problem we've been discussing in this and several other related threads.

 

I would only add that for a number of people the problem seems to occur fairly infrequently, that is, once every several hours of use. This has made troubleshooting more difficult. Thus, for example, people have tried various software up- downgrades, concluded the problem was "solved", only to have it reappear later.

 

Don

Link to comment

The apparent randomness is very bothersome; if I know when to expect the error then I could at least trust my unit. As it is, I can only assume I am within 300-500' at all times, unless I want to go through the trouble of cycling off/on each time I read from the unit (which also adds the work of splicing chunks of tracks together after I got home...)

 

Don, can you provide links to the other threads? I can do a search but you seem knowledgeable about the topic and it might take me a while to find the right threads and also to be sure I've got them all.

Link to comment

The apparent randomness is very bothersome;

 

I think we all agree with you about the apparent randomness.

 

Here are some other threads on this forum discussing this problem:

 

HCx inaccuracy after upgrade to 2.60/2.60

 

New Garmin Vista Hcx Software 2.70, There has just been a new release for the vista

 

Vista HCx Solution!, Be sure to do a hard reset after updating

 

accuracy of new vista hcx

 

Vista HCx drift problems, Also causes altimeter problems

 

Colorado location errors, I've seen my first significant issue.

 

Note that some of threads are more current than others. I'm guessing that together they have had well over 100,000 views and many hundreds of replies. Obviously, many people view and reply to several threads, but it would still indicate that this is a topic of considerable interest.

 

BTW, you're probably aware that you can track these threads by clicking on the Options drop-down menu in the upper right hand corner of the thread page.

 

Don

Link to comment

Hi Guys,

 

I realize ive said it before but I run 2.70/2.30 and have no trouble. I did lose signal for a short time in a 40 foot rock cliff canyon. My tracks are fairly consistent only at times varying maybe to 50 foot and then correcting. Both of my units are now running this and neither are showing any problems after extensive testing using the units metal detecting, walking, in the car, on the atv etc. Anyone else using these versions?

 

Kyle

Link to comment

While using version 2.60/2.60 on a weekend hiking trip in Wisconsin (hiked less than 5 miles), I experienced the "drift" issue numerous times. Since then I have been running version 2.60/2.30 and on a recent week long hike in Killarney Provincial Park (50+ miles) I did not experience one single "drift" issue. I repeat.....not a single drift issue even though the conditions were far more challenging. Isolated incident....I think not...on a day hike in South Lake Tahoe running version 2.60/2.30...10 miles and no drift issue. I did a day hike in a local park where I experienced many of the drift issues in the past.....once again no drift issues.

 

Am I happy that the odometer doesn't work well with version 2.60/2.30...of course not. But at least I am comfortable that the unit is correctly reporting my location and I can always get the mileage from the track log.

 

EFIALTIS....what kind of track discrepancies were you seeing running x/2.60 vs. x/2.30? When you say the results running x/2.60 were much, much worse it makes me wonder what type of accuracy you are expecting to achieve. While running 2.60/2.30 I have seen occasional track discrepancies up to 60 ft but the discrepancy quickly corrects itself without having to power cycle the unit (this is acceptable to me). Running 2.60/2.60 I was seeing track discrepancies up to .2 miles which would not get corrected without a power cycle (this was not acceptable to me).

 

My experiences indicate that the "drift" issue is tied to the GPS chipset software and I wouldn't expect that the GPS firmware would have any impact, negative or positive, on the accuracy issue. Maybe I will upgrade to 2.70 and test it for myself, but then again, why would I want to press my luck? My unit is working much better now and for that I am happy…though Garmin’s customer supports still leaves a lot to be desired….but that is another story.

Link to comment

''.....Three days ago i hiked Snezka in Chech republic.

I recorded the track when going up and again when going down.

When i was going down i recorded 150m drift between the other track for about half an hour and then the drift corrected without power cycle.

Except this drift the rest of the tracks were identical.

I used version 2.70/2.30.

I also tested the unit back home with version 2.70/2.60. The drift was greater and the odometer was recording meters even when standing still....

So the update doesn't correct the problem.......''

 

The problem seems to be random.The previous Sunday i hiked a mountain near the borders. The hike was 8.30 hours with a lot of tree passage but i didn't experience any problem at all.The accuracy was from 3m to 15m in heavy wood everything o.k. it corrected itself after exiting the wood's.

The major problem isn't the accuracy itself but the sudden leaps i.e you walk with 3m (or 15m it doesn't matter) accuracy and suddenly it jumps to 30m or 40m and it doesn't recover unless you powercycle.

With x/2.60 the leaps are greater and happen more often and the odometer under the drift condition keeps recording while standing still like beating seconds on a clock if you understand what i mean...

So i agree that the problem has something to do with the GPS chipset firmware and i don't think Garmin is going to fix it...

Link to comment

So i agree that the problem has something to do with the GPS chipset firmware and i don't think Garmin is going to fix it...

Yes. And by the way, it is unlikely that a software update fixes the issue (for instance 2.60 -> 2.70, currently). Only a firmware (chipset micro-software) is likely to fix it (currently 2.60).
Link to comment

Hi Kyle,

 

I use the same combination and am quite happy with it. Betatest of my long cache currently in review proved that after the update the coordinates were right finally.

RaF

 

Hi Guys,

 

I realize ive said it before but I run 2.70/2.30 and have no trouble. I did lose signal for a short time in a 40 foot rock cliff canyon. My tracks are fairly consistent only at times varying maybe to 50 foot and then correcting. Both of my units are now running this and neither are showing any problems after extensive testing using the units metal detecting, walking, in the car, on the atv etc. Anyone else using these versions?

 

Kyle

Link to comment

Here is (yet another!) example of what I call a drift recorded with eTrex Vista Hcx (2.60/2.60).

On the first picture, you can see a shift of 69m (230ft) between the two traces on the same path (return trip).

The drift, based on previous posts, is almost of the same magnitude: 50 to 70m (200 to 230ft).

The lower track is correctly on the actual path, but the upper one is drifted-up.

I will now test 2.70/2.60 AFTER a hard-reset...

 

Note: for those who are curious and/or interested, these snapshots are from a guest Windows XP running in a virtual machine (VirtualBox) installed on a host Linux OS (Ubuntu).

This is only way for me to run Mapsource as I have a desktop running Linux.

 

Capture1.png

 

What is interesting is that the altitude profile is also messed up (see below) by almost the same value (50m of 200ft).

 

Capture2.png

Edited by TransFXB
Link to comment

I've updated to 2.80/2.80 and am still having the drift issue though not as often. In general, I’ve found that the accuracy is up but also I purchased the karabiner clip so now carry it on my pack shoulder strap rather then in my pocket – that alone may be the reason for the better accuracy.

 

Anyone else using 2.80/2.80 and how is it working for you?

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...