Author Topic: Recording track status: moving time = total time  (Read 1984 times)

Offline svartbjorn

  • Premium
  • Apprentice of Locus
  • **
  • Posts: 187
    • View Profile
Recording track status: moving time = total time
« on: March 16, 2011, 22:27:03 »
Locus version 0.9.28.
Settings: Recording interval = 5m and 5s. "Record only during move": unchecked. GPS "TIme between GPS location" = 0. For the Title field, "Track record" is selected.
While recording a track and making stops after start, pressing "i" brings up the status window of the track being recorded.
Three issues:
  1) After pressing REC, the time in the Title field doesn't start running until you actually start moving initially. This time field is later displaying the total time, so it should start running at the moment REC is pressed.
  2) In the "i" status popup window, the "Speed > 0" fields for "Average speed" and "Track time" incorrectly include the stop times. So the "All" the "Speed > 0" fields are displaying the same time, except for an initial difference equal to the time from REC is pressed until you actually start moving the first time. This difference remain the same through the whole recording, ignoring stops.
  3) While stopped, pressing "i" repeatably to bring up an updated "i" window, doesn't update the "All" fields. It should since the time is running.
« Last Edit: January 01, 1970, 01:00:00 by Guest »
 

Offline svartbjorn

  • Premium
  • Apprentice of Locus
  • **
  • Posts: 187
    • View Profile
Re: Recording track status: moving time = total time
« Reply #1 on: March 17, 2011, 22:42:06 »
Latest version 1.0.0 issue #1 is solved, but issue #2 and #3 are still a problem.
« Last Edit: January 01, 1970, 01:00:00 by Guest »
 

Offline menion

  • Administrator
  • Professor of Locus
  • *****
  • Posts: 11302
  • Thanked: 284 times
    • View Profile
    • http://www.asamm.com
  • Device: SGS7
Re: Recording track status: moving time = total time
« Reply #2 on: March 18, 2011, 07:41:23 »
I'll look onto point 2, thanks. Anyway actually don't know simple solution for num three because you can imagine that "i" button show stats of actually recorded tracks so it measure time, distance from first to last point. I'll think about it more, promise
« Last Edit: January 01, 1970, 01:00:00 by Guest »
Ideas, wishes, problems
Advanced topics, public discussion, sharing of knowledges, testing beta versions: you're here!
 

Offline svartbjorn

  • Premium
  • Apprentice of Locus
  • **
  • Posts: 187
    • View Profile
Re: Recording track status: moving time = total time
« Reply #3 on: March 18, 2011, 21:30:03 »
If "i" shows stats of actually recorded track so far, that I guess explains why "moving time" = "total time". Because on the menu Settings/Track there are two entries "Interval (meters)" and "Interval (sseconds)". You do not want to set "Interval(meters" too low like 0 or 1m, since you are then detected as moving all the time because of the inaccuracy of your GPS position. So you need to set something like 5m, 10m , 20m. Then there won't be any new sample points as long as you are not moving, right?

So I guess you would need to treat the "i" of the recording track differently than stored tracks. You have the running time available, right, as shown in the Title window.

MyTracks is presenting this information correctly. When you have this issue fixed, then it is time to uninstall MyTracks from my phone.
« Last Edit: January 01, 1970, 01:00:00 by Guest »
 

Offline svartbjorn

  • Premium
  • Apprentice of Locus
  • **
  • Posts: 187
    • View Profile
Re: Recording track status: moving time = total time
« Reply #4 on: March 20, 2011, 18:03:34 »
Menion, I think the basic problem here is that the criteria for storing data points in track log is wrong.
We have:
- Settings/Track/Interval(meters) = INT-M
- Settings/Track/Interval(seconds) = INT-S
- Settings/Track/Record only during move = RMOV

RMOV checked (on): Store data points only when moving, i.e. the criteria for storing data points should be INT-M AND INT-S. This seems to be working correctly.

RMOV unchecked (off): Store data points even when not moving, i.e. the criteria for storing data points should be INT-S only. This doesn't seem to be the case, rather the AND-criteria above incorrectly seems to be used here also. So basically the RMOV selection seems to be ignored. INT-M must used as a FILTER to determine if the position of the new data point should be a copy of the previous point or a new GPS position. The filter is necessary to remove the small movements due to the inaccuracy of the GPS position.

If you correct the storing criteria as suggested above, you do not need to do anything about the "i" of the recording track.

What do you think, Menio? Does this make sense?
« Last Edit: January 01, 1970, 01:00:00 by Guest »
 

Offline menion

  • Administrator
  • Professor of Locus
  • *****
  • Posts: 11302
  • Thanked: 284 times
    • View Profile
    • http://www.asamm.com
  • Device: SGS7
Re: Recording track status: moving time = total time
« Reply #5 on: March 21, 2011, 20:56:31 »
hmm you need to know how actual algorithm works ...

- Settings/Track/Interval(meters) = INT-M
 test last stored location with actual location from GPS

- Settings/Track/Interval(seconds) = INT-S
 same as previous but for time

- Settings/Track/Record only during move = RMOV
  test if location object returned from gps have some speed, so if speed > 0

And finally how do I check if save location or not. I'm just testing all three criteria. If all success, point is stored. If you pause or start record (after pause), and last point != actual point, point is saved also. So I think, this is correct behavior isn't it? Settings RMOV was added only due to problems that cause GPS when you stay on on place and GPS is betting better or worst accuracy and you're moving on map but without any speed value
« Last Edit: January 01, 1970, 01:00:00 by Guest »
Ideas, wishes, problems
Advanced topics, public discussion, sharing of knowledges, testing beta versions: you're here!
 

Offline svartbjorn

  • Premium
  • Apprentice of Locus
  • **
  • Posts: 187
    • View Profile
Re: Recording track status: moving time = total time
« Reply #6 on: March 21, 2011, 22:11:02 »
You are saying that all three criteria need to be true before you store a point. Yes, that is how it looks to me, and that is what doesn't seem correct to me - because: you have said before that the "i" status window of a track being recorded is based on stored points. And you are confirming that a point is only stored if I have moved, i.e INT-M is true. So how is then the non-moving time registered? If I stop, the "i" status doesn't change, but when I then move again, the "Track time" field of "All" and "Speed > 0" jump up by the same amount of time. Could this be correct if that time update is done only for the "All" field? The "Speed > 0" field should not jump since I haven't moved, right? Isn't that a solution? (since storing points when not moving just for the purpose of catching the non-moving time is waste of memory).

From your last response I do not understand what RMOV actually is used for since we have the INT-M criteria that should take care of that problem (GPS inaccuracy). Or am I missing something here?
« Last Edit: January 01, 1970, 01:00:00 by Guest »
 

Offline menion

  • Administrator
  • Professor of Locus
  • *****
  • Posts: 11302
  • Thanked: 284 times
    • View Profile
    • http://www.asamm.com
  • Device: SGS7
Re: Recording track status: moving time = total time
« Reply #7 on: March 22, 2011, 07:13:12 »
Hmm looks that you're really interested in "All" and "Speed >0" fields. Looks that best should be remove this fields :)

Hmm so, if you set min distance to 0, all points will be saved (based on min time value of course). So you can have stored points with zero speed!

RMOV criteirum - this can have value if you set INT-M for 0 or! ... just try to step out and wait without move. You'll see that you GPS is moving on map a little, but have no speed at all. So RMOV fix this also.

Do you really think that actual state do not have sence? Maybe "All" and "Speed>0" is a little weird. And RMOV is maybe a little unclear how it works. Anway I think that all three together make sence and you can thanks to it set recording exactly as you want.

Also count that "i" screen is just "temporal" ... I really want to create some screen for recording with all possible info, but it takes some time and I don't have much of it now for this "not very needed" features. Anyway for some improvements on this topic, I'm open!
« Last Edit: January 01, 1970, 01:00:00 by Guest »
Ideas, wishes, problems
Advanced topics, public discussion, sharing of knowledges, testing beta versions: you're here!
 

Offline svartbjorn

  • Premium
  • Apprentice of Locus
  • **
  • Posts: 187
    • View Profile
Re: Recording track status: moving time = total time
« Reply #8 on: March 22, 2011, 14:46:07 »
I think the "All" and "Speed > 0" are of interest for many people. And this is a feature you find in most other GPS apps, so a brilliant app like Locus should support it too. When I go out for exercise, my true moving speed (= "Average speed" under "Speed > 0") is of great interest, but I also would like to know how long I have been out. So please keep all this - just make the statistics correct.

I am sure you know the best way to do this correctly, Menion. I am just reporting what I find. As it is today, it just isn't correct. I have also tried setting INT-M = 0, but if I remember correctly, the small movements due to GPS inaccuracy caused the distance to accumulate (with RMOV unchecked). I can try checking the RMOV box, but then I understand only moving points are stored and we are back to the basic problem? And as I said in my previous post, storing identical points just to get the stop time info, seems a bit unnecessary.

Anyway, Menion, it is just fine if you look at this later when you have time. This is not critical, just good to know it is coming.

BTW - I paid for the PRO version to support you and to get all the new features in the future. Very best regards. Keep on with your great work!
« Last Edit: January 01, 1970, 01:00:00 by Guest »