Jump to content

Beta Software: Oregon 400t, Ver. 3.15, Beta as of Aug 18, 2009,


strumble
Followers 10

Recommended Posts

I don't know if this is new but I noticed this issue today and sent Oregonbeta@garmin.com the following email.

 

"Set the "PRESSURE" units to "INCHES(HG)" and then try to calibrate the altimeter using known inches of mercury(hg). Inches of mercury have two significant digits anfter the decimal point. The Oregon units only allow you to adjust one digit after the decimal point. Calibration seems like it thinks you're in millibars and appearantly doesn't recognize the fact that you're in inches of mercury (hg). "

 

Small annoyance yes, but an annoyance nontheless.

Link to comment

REPLY:

 

They changed the GPS firmware to 3.7. Some say the tracklogs and positional accuracy are more averaged. I say they simply dialed in the chipset and finally engineered an update that works. I can hold my Oregon in different hands (held all the way out to their perspective sides) while walking, switching back and forth, and I can actually see where I change hands in the tracklog. The averaging can't be too bad if one can get this kind of resolution. Before, the tracklogs would look erratic and simply be innaccurate at closer zoom levels.

Link to comment

REPLY:

 

They changed the GPS firmware to 3.7. Some say the tracklogs and positional accuracy are more averaged. I say they simply dialed in the chipset and finally engineered an update that works. I can hold my Oregon in different hands (held all the way out to their perspective sides) while walking, switching back and forth, and I can actually see where I change hands in the tracklog. The averaging can't be too bad if one can get this kind of resolution. Before, the tracklogs would look erratic and simply be innaccurate at closer zoom levels.

Yogazoo,

 

You must be exaggerating here. GPS accuracy with WAAS is around 10-15 feet. Unless you have a HUGE wingspan AND you had the best WAAS signals you could get, then it's not posible to determine when you switched hands with your GPS based on tracklogs.

 

JetSkier

Link to comment

Not exaggerating a bit JetSkier. I'll post the tracklogs one of these days. I do it on a consistent basis as well. Everyday on my 10 oclock break I walk around the block and perform the hand to hand test zoomed into 20 feet on the map screen. I can see the tracklog undulate as I switch hands while I walk (looks like a snake). Before the 3.7 update it looked like pure crap. I have a "wingspan" of 6 feet.

 

True, GPS accuracy is within 10-15 feet, BUT accuracy is not what I'm talking about here. I'm talking about tracklog recording. Tracklog performance has little to do with accuracy and is mostly a product of processing of the available signals (accurate or not). The 60CSX, with the same 10-15 ft accuracy, laid down some really nice and detailed tracks that would pronounce the slightest of switchbacks.

 

It appears to me now, under relatively little tree cover and/or sky obstructions, the Oregon tracks are quite comparable.

Edited by yogazoo
Link to comment

If you make a track with your gps in one hand and the gps is between a highrise/mountain/canyon and your body, then you take the gps in the other hand, where the gps has a better reception because it has a better view to the sat's, you can see a difference.

 

This is not new, has nothing to do with new beta's or waas.

The 'smooth' track is just software averaged and actually I think it's wrong.

Edited by splashy
Link to comment

So, if you move your GPS above your head and then down to your feet do you see a 5-6' change in elevation? I don't think so. The tracklog is totally based on sampling your position and then "smoothing" out a track. It is totally based on the GPS accuracy of 10-15 feet. I'm not convinced at all.

 

JetSkier

Link to comment

I'm not talking about above, I say if you shield it on one side and have a bad sight to the sat's and on the other side you have a clear sight to the sat's, you might see a difference when you change hands.

If you keep it on top of your head you should have a good reading, if it sees the sat's, but you will look a bit funny. ;)

Edited by splashy
Link to comment
Not exaggerating a bit JetSkier. I'll post the tracklogs one of these days. I do it on a consistent basis as well. Everyday on my 10 oclock break I walk around the block and perform the hand to hand test zoomed into 20 feet on the map screen. I can see the tracklog undulate as I switch hands while I walk (looks like a snake). Before the 3.7 update it looked like pure crap. I have a "wingspan" of 6 feet.

 

True, GPS accuracy is within 10-15 feet, BUT accuracy is not what I'm talking about here. I'm talking about tracklog recording. Tracklog performance has little to do with accuracy and is mostly a product of processing of the available signals (accurate or not). The 60CSX, with the same 10-15 ft accuracy, laid down some really nice and detailed tracks that would pronounce the slightest of switchbacks.

 

It appears to me now, under relatively little tree cover and/or sky obstructions, the Oregon tracks are quite comparable.

 

yogazoo,

 

I have seen the same thing you are referring to.

 

When I first got my 400t I was comparing it's tracklog using unit software 2.97, to my 60CSx, software 3.7. To do so I walked my lot's perimeter at a normal walking pace holding both the 60CSx and the 400t side by side approximatley 1.5' in front of my chest with both units tilted up at approximately a 45 deg. angle, did this 3 seperate times. My lot is 2 acres with 3 sides perfectly straight with the 4th side being road frontage of 437' in a large smooth arc. Corners are clearly marked with pins so walking the perimeter precisely is very repeatable. Along one side I have a rock lined water runoff ditch that is 2 ft. wide right on the property line.

 

The first time I compared the 2 units I stepped across this ditch, don't remember exactly why, but it was a side step of no more than 3 ft. When I reviewed the tracks on the computer zoomed in maximum amount this side step was very apparent on the Oregon's track but did not show up at all on the 60CSx's track. I questioned if it was repeatable so I did it a second time. Yes, it showed up the second time also.

 

Now the bad news was that the 60CSx (3.7) tracking was so much better than the Oregon (2.97) that there was no comparison as to which you would want to follow. The 60CSx did not show the 3 ft. side step but it had all 3 sides perfectly straight and the arced side was nice and smooth just like it is. The Oregon had a significant amount of clutter/perceived error due to sensitivity and erroneous sensitivity in some cases, none of the 3 lines were perfectly straight and the arced side was not smooth either. Looking at both tracks zoomed in on the PC the 60CSx came back to within 3 ft. of my original starting point and the Oregon's track showed it was 18 ft. from my original starting point, My starting and ending point were at the exact same spot standing on top of a pinned corner.

 

I haven't repeated this test since then with any of the newer software updates but it did make me aware that the Oregon did pick up the slightest movement in my case 3' laterally. In real world practical tracking applications this slight amount of movement is probably not good since so much error is also recorded in the track.

 

So knowing this I sort of feel Garmin's pain in trying to determine at what level is sensitivity of recording data in a track a trade off with real world practicality. I am curious now how the latest 3.15 beta is averaging tracklog data.

Link to comment

I'm sticking with v3.10 at the moment. I do have two questions for users of v3.15 beta.

 

1) Is the Oregon tracklog inaccuracy issue now resolved? I don't mean perfect, just mean no longer off by huge margins like earlier software versions. I think the problem was related to slow speeds like walking.

 

2) How is WAAS reception? Does it still require a clear view of the sky and a long wait time or can you get a quick WAAS lock with some tree cover?

 

Thanks

Edited by Low Bat
Link to comment

So, if you move your GPS above your head and then down to your feet do you see a 5-6' change in elevation? I don't think so. The tracklog is totally based on sampling your position and then "smoothing" out a track. It is totally based on the GPS accuracy of 10-15 feet. I'm not convinced at all.

 

JetSkier

 

The variance in the tracklog from switching hand-to-hand is not at all related to accuracy. It's dependant on the smallest coordinate increment the gps can record. Accuracy refers to an absolute position; the tracklog is recording relative positions - hopefully, but not necessarily accurate ones.

 

Oh yes, back to the issue at hand - I just got a new Oregon 200, installed 3.13 beta, and got "System Software Missing". I tried all the tricks I found online to no avail so I exchanged it for a new one the next day. Now I'm using v3.10 (like Low Bat) and have no intention on trying a beta version again. Ever.

 

I have thousands of custom POI's for Waymarking, catagorized with custom symbols, etc., and the unit is working A-1. I could not be more pleased. The Oregon replaces an eTrex Venture Cx, which also handled the POI's well.

Edited by paulmc
Link to comment

I'm sticking with v3.10 at the moment. I do have two questions for users of v3.15 beta.

 

1) Is the Oregon tracklog inaccuracy issue now resolved? I don't mean perfect, just mean no longer off by huge margins like earlier software versions. I think the problem was related to slow speeds like walking.

 

2) How is WAAS reception? Does it still require a clear view of the sky and a long wait time or can you get a quick WAAS lock with some tree cover?

 

Thanks

 

Tracklogs do seem to be much better and more consistent. I've taken two weeks worth of comparison using 3.1x beta (all use the same 3.7 GPS software) and so far the Oregon is holding its own against the 60csx in terms of tracklog accuracy and the tracks don't appear overly averaged like they did in earlier releases.

 

That said the Oregon still records too much stopped time and frequently you will see your speed drop to zero if you are moving slowly under cover. These speed drops are usually accompanied by heading freezes as well.

 

WAAS is no different than before.

Link to comment

If you make a track with your gps in one hand and the gps is between a highrise/mountain/canyon and your body, then you take the gps in the other hand, where the gps has a better reception because it has a better view to the sat's, you can see a difference.

 

This is not new, has nothing to do with new beta's or waas.

The 'smooth' track is just software averaged and actually I think it's wrong.

 

We seemed to have discussed the issue of changing hands at Dec 2008/Feb 2009.

 

I tried the test in the early part of the year -walked the same route every day for exercise with Oregon in my right hand in both directions.

Result the accuracy of the 2mile route varied depending on whether walking South/West or North/East.

 

I tried another test on a 'board walk' 3 feet wide on a Nature reserve in both direstions.[with Oregon in my right hand in both directions] Result the accuracy varied depending on whether walking South/West or North/East.

 

I tried another test along a 4"[inch] wide kerb along my street. [with Oregon in my right hand in both directions] Result the accuracy varied depending on whether walking South/West or North/East.

 

The best way to test the accuracy was the board walk with a bog on both sides because it had precise boundaries and restricted you to a track 3 feet wide.

 

So the principle of changing hands and getting different readings was understandable as the view of the satellites change :(

 

So if you are walking along a cliff edge in Wales [in fog or darkness] make sure the unit has a clear view of the satellites in the south and west!! :D

Link to comment

Spent the day out today with my Oregon 400t and 3.15 Beta loaded. Performed like a champ. I didn't see a ton of difference in 3.10 in overall performance, and didn't test any other betas between 3.10 and 3.15.

 

Compass performed well in auto mode, good accuracy under tree cover, and benchmarked the unit against a USGS marker and it was right on.

Link to comment

Spent the day out today with my Oregon 400t and 3.15 Beta loaded. Performed like a champ. I didn't see a ton of difference in 3.10 in overall performance, and didn't test any other betas between 3.10 and 3.15.

 

Compass performed well in auto mode, good accuracy under tree cover, and benchmarked the unit against a USGS marker and it was right on.

 

Also went for a hike today with the 3.15 beta loaded. Couldn't be happier with the performance! Dead on accurate the entire time. Got to the cache and left the GPS there for about 10 minutes. The distance did not shift by more than 5 feet.

 

Tracks have been as smooth as butter and also dead on accurate.

:)

 

Pics:

GPS

Track

Edited by Tahoe Skier5000
Link to comment

Spent the day out today with my Oregon 400t and 3.15 Beta loaded. Performed like a champ. I didn't see a ton of difference in 3.10 in overall performance, and didn't test any other betas between 3.10 and 3.15.

 

Compass performed well in auto mode, good accuracy under tree cover, and benchmarked the unit against a USGS marker and it was right on.

 

Also went for a hike today with the 3.15 beta loaded. Couldn't be happier with the performance! Dead on accurate the entire time. Got to the cache and left the GPS there for about 10 minutes. It did not shift by more than 5 feet.

 

Tracks have been as smooth as butter and also dead on accurate.

:P

 

Pics:

GPS

Track

 

I need to pull my tracklogs off from yesterday to have a look see... I've wanted to start doing some tracklogs to share and upload to various projects but haven't been happy with the quality of the logs really. Maybe they are improved enough at this point that I would feel comfortable with that.

 

FWIW: The link for the image for your tracklog is broken.

Link to comment

Beta 3.15 I mark a wayoint. Then I go to "Where To?" and it shows that the waypoint is 40 to 60 metres away.

 

Consistently

 

dadgum. I like betas but I shake my head that it seems that Garmin do NO testing what so ever - how can they miss these little things before putting out a beta? Don't programmers test their own code any more?

 

I doubt if it has to do with the code but rather your unit. If I mark a waypoint then go to "Where To" it will show that waypoint at 4 feet away or less. Consistently.

Link to comment

Beta 3.15 I mark a wayoint. Then I go to "Where To?" and it shows that the waypoint is 40 to 60 metres away.

 

Consistently

 

dadgum. I like betas but I shake my head that it seems that Garmin do NO testing what so ever - how can they miss these little things before putting out a beta? Don't programmers test their own code any more?

 

I doubt if it has to do with the code but rather your unit. If I mark a waypoint then go to "Where To" it will show that waypoint at 4 feet away or less. Consistently.

 

I'll confirm this as well. . . no problems like Squirrel mentions for me. I marked several waypoints over the weekend, averaged a few existing and saw no abnormalities. Hmm.

Link to comment

I'm also very satisfied with this release. Accuracy is great and the compass updates way more frequently than before.

 

Only issue I noticed so far is that when changing profiles sometimes some of the settings are not changed correctly. This is especially true for the Lock on road yes/no setting, which often stays on "no" when changing to a profile where it was supposed to be on "yes"

Link to comment
Not exaggerating a bit JetSkier. I'll post the tracklogs one of these days. I do it on a consistent basis as well. Everyday on my 10 oclock break I walk around the block and perform the hand to hand test zoomed into 20 feet on the map screen. I can see the tracklog undulate as I switch hands while I walk (looks like a snake). Before the 3.7 update it looked like pure crap. I have a "wingspan" of 6 feet.

 

True, GPS accuracy is within 10-15 feet, BUT accuracy is not what I'm talking about here. I'm talking about tracklog recording. Tracklog performance has little to do with accuracy and is mostly a product of processing of the available signals (accurate or not). The 60CSX, with the same 10-15 ft accuracy, laid down some really nice and detailed tracks that would pronounce the slightest of switchbacks.

 

It appears to me now, under relatively little tree cover and/or sky obstructions, the Oregon tracks are quite comparable.

 

yogazoo,

 

I have seen the same thing you are referring to.

 

When I first got my 400t I was comparing it's tracklog using unit software 2.97, to my 60CSx, software 3.7. To do so I walked my lot's perimeter at a normal walking pace holding both the 60CSx and the 400t side by side approximatley 1.5' in front of my chest with both units tilted up at approximately a 45 deg. angle, did this 3 seperate times. My lot is 2 acres with 3 sides perfectly straight with the 4th side being road frontage of 437' in a large smooth arc. Corners are clearly marked with pins so walking the perimeter precisely is very repeatable. Along one side I have a rock lined water runoff ditch that is 2 ft. wide right on the property line.

 

The first time I compared the 2 units I stepped across this ditch, don't remember exactly why, but it was a side step of no more than 3 ft. When I reviewed the tracks on the computer zoomed in maximum amount this side step was very apparent on the Oregon's track but did not show up at all on the 60CSx's track. I questioned if it was repeatable so I did it a second time. Yes, it showed up the second time also.

 

Now the bad news was that the 60CSx (3.7) tracking was so much better than the Oregon (2.97) that there was no comparison as to which you would want to follow. The 60CSx did not show the 3 ft. side step but it had all 3 sides perfectly straight and the arced side was nice and smooth just like it is. The Oregon had a significant amount of clutter/perceived error due to sensitivity and erroneous sensitivity in some cases, none of the 3 lines were perfectly straight and the arced side was not smooth either. Looking at both tracks zoomed in on the PC the 60CSx came back to within 3 ft. of my original starting point and the Oregon's track showed it was 18 ft. from my original starting point, My starting and ending point were at the exact same spot standing on top of a pinned corner.

 

I haven't repeated this test since then with any of the newer software updates but it did make me aware that the Oregon did pick up the slightest movement in my case 3' laterally. In real world practical tracking applications this slight amount of movement is probably not good since so much error is also recorded in the track.

 

So knowing this I sort of feel Garmin's pain in trying to determine at what level is sensitivity of recording data in a track a trade off with real world practicality. I am curious now how the latest 3.15 beta is averaging tracklog data.

 

 

 

Update to my above post #19.

 

I repeated the above test using the Oregon 3.15 beta software (had not done this since 2.97 SW update) and the Oregon recorded an outstanding track; comparable to the 60CSx. Was very smooth straight lines where it shoud be and turns were right angles, also as they should be. Looks like Garmin finally got the track averaging/recording of track points correct. Even did the 3' side step at the same location as I discussed in post #19 above and it did not appear now... a good thing. When I completed the loop my ending point was within 5' of my starting point (the 2 are at the same pin). Track had absolutely no clutter at a slow continuous walk.

 

I am very pleased with the tracking capability of the Oregon now.

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...
Followers 10
×
×
  • Create New...