GPS Cycle Computer v3

Search This thread

fabrylama

Member
Jul 3, 2008
49
0
Hi Klaus,
I did some more testing on the HR logging.. if the HR is disabled gcc works perfectly (as always), if the HR is enabled there are random problems: sometimes the phone soft reset itself, sometimes the signal quality drops to 0, and the only fix is the soft reset and one time the gps stopped reading (no data from gps) and the only fix was the soft reset.
These problems don't happen everyday, more like once every couple of rides, and after at least 30 min from the beginning of the log.
I tried changing ROM, and/or radio on the phone, but everything is the same.
Maybe the phone wasn't build to keep the mic-in open for so long and something overflows?:confused:

Anyway you said you could try to automatically close and reopen the wave device every x minutes, I think it's worth a try...
Thanks again

Fabrizio
 

pecorone

New member
May 29, 2012
3
0
Hi pecorone,
why didn't you install the .NETCF the normal way by simple click (execute the .cab)?
Klaus

hi, for install .net framework 3.5 I have to reallocate memory but gcc works now. the problem is that when I turn off my navigator windows ce is reset and i have to reinstall .net framework every time.
is there any solution?
 

Blaustein

Senior Member
May 13, 2009
378
79
Hi pecorone,
I think this is a device specific issue, not related to gcc. Perhaps you can find help in any forum of your device.
Can you install .NETCF on the storage card?
Klaus
 

Blaustein

Senior Member
May 13, 2009
378
79
Hi Fabrizio,
I have analyzed the code and found no error, but there is one thing which is not documented how it works in detail. I have made a shortcut in code which is imaginable to cause the problems in some rare cases.
Please test the attached exe (without this shortcut) before I realize such a botch like regularly close and open the wave device.
Klaus
 

Attachments

  • GpsCycleComputer.zip
    263.5 KB · Views: 28

Blaustein

Senior Member
May 13, 2009
378
79
Hi Alfredo,
attached a new beta 4.10beta1 (only .exe)
New features:
- display statistics and altitude graph of T2F
- performance enhancement on display graph

You can show the statistics of t2f by context menu on button Track2Follow.
Klaus
 

Attachments

  • GpsCycleComputer.zip
    264.7 KB · Views: 50

fabrylama

Member
Jul 3, 2008
49
0
Hi Fabrizio,
I have analyzed the code and found no error, but there is one thing which is not documented how it works in detail. I have made a shortcut in code which is imaginable to cause the problems in some rare cases.
Please test the attached exe (without this shortcut) before I realize such a botch like regularly close and open the wave device.
Klaus

Hi Klaus,
I tested the exe for 3 long rides and so far it worked perfectly.
Thank you very much!
Fabrizio

PS does the new beta with the new features you provided today also implement the HR fix?
 

low.flying.pigs

Senior Member
Feb 3, 2012
202
11
Hello Blaustein / Klaus,

I am running GCC 4.10beta 1. Other than the abnormal termination functionality, I have not encountered any issues with the functionality I have used. Overall the software is excellent and I thank you for it's continued development.

I had an incident whilst cycling recently. The seperate GPS receiver remained on the bike however the Pocket PC detached from the bike and hit the ground quite severely. The impact caused the battery cover to come away and the battery itself to come out of the Pocket PC. Fortunately, there was no real damage done. After reasembling everything and restarting GCC, it asked if I wished to resume logging, I selected yes. Upon doing so, I then noticed the Distance increased immediately out of all proportion. The bike and GPS receiver had not moved significantly at all from the position where I had crashed, so this was unexpected.

I have confirmed this behavior is consistent by cycling a short distance, and then stopping stationary and pulling the battery on the Pocket PC. Even though remaining stationary throughout, the effect was the same. From memory, I think distance was 0.45 miles aproximately at the point of pulling the battery, immediately increasing to 0.78 miles approximately upon resuming despite having remained stationary.

Unless whatever causes this is equipment specific, it should be easy for you to replicate as above. It might be worth some attention before the next beta release.

Let me know if you need any further information or testing.

low.flying.pigs
 

Blaustein

Senior Member
May 13, 2009
378
79
Hello low.flying.pigs,
I am out for the weekend so be patient. Can you do another test: Do the same, but choose not to continue and switch gps on (if it is not on) and check that the position on map is stable and at the right place. Then choose continue from the menu page.
Check the track on the map (also for the first test) if there are some irregular points , which can produce this increase in distance.
Have a nice weekend
Klaus
 
Last edited:
  • Like
Reactions: low.flying.pigs

womisa

Member
Jul 21, 2010
12
0
Herzfrequenz mit Sportstracker?

Hi Blaustein/Klaus

is it possible to support the heartrate belt "Sports Tracker" in your program?
've since moved on Android and had long ceased in this forum. But now I have my "OLD" and my bike made ​​696 OLD Asus rolling again and I have a Bluetooth Sports Tracker chest gained the interesiert ....... but no one!

Now to my question:
Is it possible that this is supported by the GPS cycle computer and logged with the GPS log data in conjunction with the heart rate?
Maybe I can contribute to the realization and info since the beta test,


(sorry ich versuche es in Deutsch),

ich habe inzwischen auf Android gewechselt und war lange nicht mehr in diesem Forum. Aber jetzt habe ich mein "ALTES" Fahrrad und mein ALTEN Asus 696 wieder flott gemacht und habe mir einen Bluetooth Sportstracker Brustgurt zugelegt .......das interesiert aber wohl niemand!

Nun zu meiner Frage:
Ist es möglich, dass dieser vom GPS Cycle Computer unterstützt wird und Logdaten mit den GPS-Daten in Verbindung mit der Herzfrequenz loggt?
Eventuell kann ich da infos zur Realisierung und zum BetaTest beitragen,

MfG
Achim
 

Blaustein

Senior Member
May 13, 2009
378
79
Hi Blaustein/Klaus
is it possible to support the heartrate belt "Sports Tracker" in your program?
Hello Achim,
I suppose you need Bluetooth V4.0 on your phone to support this belt. This also applies to the Polar H7 belt (Bluetooth smart).
Did you try the Sports Tracker application for Windows Phone? Maybe it doesn't work on your Windows Mobile phone at all.
On the support site of Sports Tracker there is the following FAQ:
Is the ST hrm (heart rate monitor) compatible with Windows Phone?

Last Updated: Mar 19, 2012 03:34PM EET

Not at the moment. Because of the closed interface of WP, the ST hrm doesn't work with the WP app. If Microsoft ads the necessary Bluetooth support to it's OS then we are able to add the support.

Unfortunately the same applies to the Polar Wearlink with Bluetooth belt.
So I guess it will be impossible or at least very hard to support this HR belt in GCC too.
Klaus
 

low.flying.pigs

Senior Member
Feb 3, 2012
202
11
Hello low.flying.pigs,
I am out for the weekend so be patient. Can you do another test: Do the same, but choose not to continue and switch gps on (if it is not on) and check that the position on map is stable and at the right place. Then choose continue from the menu page.
Check the track on the map (also for the first test) if there are some irregular points , which can produce this increase in distance.
Have a nice weekend
Klaus

Hello Blaustein / Klaus,

Thank you for your reply.

A cursory inspection of the maps reveals no obvious irregular points.

Choosing continue from the menu page also results in distance increasing abnormally.

On the main screen, nothing else seems to change apart from average speed and the odometer.

The effect on average speed seems to be because upon resuming by either method, the time since abnormal termination is added and included to the elapsed time within the 'stop time'. That I presume is by design, and for an abnormal termination event, is probably desireable. If resuming logging of a 'session' however in another circumstance as one would by selecting continue from the menu page, for example resuming ''yesterdays' log for 'todays' leg of a cycle tour, the overnight elapsed time seriously scews the readings. An option whether to include the elapsed time from the end of the log, to the current time when resuming would resolve the issue. Ordinarily I log rides individually. However, if doing a cycle tour or an extended length route / ride over a period of days as I hope to begin in the near furutre, I think I would prefer to record the route as a whole. I don't know however if there are any limitations on log sizes etc.

The odometer has reset, when it had been up to around 800 miles. From observation, it seems it might be becuase the file is written / updated when GCC is terminated, and when termination is abnormal, the odometer distance is lost. Unfortunately, I did not think to check the odometer file before restarting GCC after the first incident. I notice it records some information about the 'exit', usually being 'normalExit'.

There is no particular hurry to any of this from my point of view, I just thought it better to bring to your attention. Let me know if you need any further information or testing. GCC is an excellent peice of software in my opinion, it would be great to see development continue as and when circumstances allow.

low.flying.pigs
 
Last edited:

Blaustein

Senior Member
May 13, 2009
378
79
Hello low.flying.pigs;
there was a bug: when restartig GCC, load a track and continue, then immediately the shortest distance between start of track and current position was added to the distance. It should now be fixed but I haven't tested it in real - I think you will do it.

In Versions prior 4.9 tracks was limited to 18 hours but since 4.9 the limit is 68 years, which should be enough.

The odometer should never reset. GccState (with current odo) is written on all major actions (switch on, off GPS; start, stop, pause, continue of tracks; load files and exit GCC. One possibility is a problem with reading the file. Do you have a "LoadState" error in GpsCycleComputer.log? If GCC crashes you should only loose the km since last save of GccState, unless it crashes right in the moment of saving and the file is corrupt.

Attached is version 4.10 beta 2
Klaus
 

Attachments

  • GpsCycleComputer_4.10beta2.zip
    264.6 KB · Views: 24
  • Like
Reactions: low.flying.pigs

low.flying.pigs

Senior Member
Feb 3, 2012
202
11
Hello low.flying.pigs;
there was a bug: when restartig GCC, load a track and continue, then immediately the shortest distance between start of track and current position was added to the distance. It should now be fixed but I haven't tested it in real - I think you will do it.

In Versions prior 4.9 tracks was limited to 18 hours but since 4.9 the limit is 68 years, which should be enough.

The odometer should never reset. GccState (with current odo) is written on all major actions (switch on, off GPS; start, stop, pause, continue of tracks; load files and exit GCC. One possibility is a problem with reading the file. Do you have a "LoadState" error in GpsCycleComputer.log? If GCC crashes you should only loose the km since last save of GccState, unless it crashes right in the moment of saving and the file is corrupt.

Attached is version 4.10 beta 2
Klaus

Hello Blaustein / Klaus,

Thank you for the information regarding track limits.

I can confirm GCC 4.10beta 2 resolves the resumed session distance error issue, by whatever method the session is resumed. Excellent work.

What are your thoughts regarding an option to include / exclude the elapsed time since the end of the session being resumed within the 'stop time'? If resuming after a few moments due to an abnormal termination for example, it may be desireable. If resuming after an overnight stop for example, it may well not be.

Regarding the resetting odometer, there were a number of errors in my GpsCycleComputer.log file, including load state errors. I renamed the existing file so GCC would start afresh with a new log file and will carry out some further testing. As far as I can tell so far, something causes the gccstate.txt file to be unreadable / corrupted and it is then overwritten by GCC starting again from zero, there being no other storage of the previous odometer value to restore from. I will let you know any findings. If it just happening when an abnormal termination occurs, that is not to much of issue as that should be rare. If not, it may be more significant. Time will tell.

Something that is perhaps more of an issue are the 'pause', 'stop', 'gps off', and 'X close program' buttons. All work as intended and that is the problem, although no fault of GCC, more the practicalities of cycling and the equipment. I wasn't going to bother mentioning this but since it keeeps happening occasionally, this evening being the latest incident, I thought it better to bring it to your attention.

Occasionally, logging gets interupted unintentionally and goes unnoticed resulting in lost logging. It seems to me a 'press and hold' instead of just 'press' for these functions would largely eliminate the issue (perhaps with the exception 'X close program, it not being part of GCC), at least in most situations I suffer. It is probly more elegant than a popup 'do you really want to' type message.

Incidentally, the 'X close program' button for me if pressed whilst logging, shows a breif message 'GPS is logging do you want to exit stop logging' with a yes or no option. It then promply closes itself without waiting for an answer ending logging regardless.

Again there is no particular hurry to any of this from my point of view, have a good weekend.

Let me know if you need any further information or testing.

low.flying.pigs
 
Last edited:

Blaustein

Senior Member
May 13, 2009
378
79
Hello low.flying.pigs,

Thank you for testing. :)

>>option to include / exclude the elapsed time since the end of the session being resumed within the 'stop time'
There are no plans to do anything like this. I think it will be too difficult to manage and it is only useful for a (hopefully) very seldom malfunction or accident.

>> odometer
V4.10beta2 includes a slight change in reading GccState.txt file so that odo is read (and used) even if the file is corrupt afterwards (odo is the first line). This hopefully reduces the cases where odo is lost. You should never have GccState.txt opened in an editor while using GCC, as this could generate access failures.

>>'pause', 'stop', 'gps off' 'press and hold'
I will think about, but it was relatively complex to implement such a function on the left button. And next time the users complain that the button doesn't work reliably. I think I leave it as it is.

>> 'X close program' button ... promply closes itself without waiting for an answer
I cannot reproduce this issue. Which device and Windows version do you use? On my device and emulator it works as intended.
But I have discovered a bug when using Exit button from menu and then choose 'no' to cancel exiting (it exits anyway).

Klaus
 

low.flying.pigs

Senior Member
Feb 3, 2012
202
11
Hello Blaustein / Klaus

My Pocket PC runs Windows Mobile 2003 SE and GCC 4.10 beta 2.

I can confirm as you noticed that if logging, when selecting exit from the menu, it will await a response, but it will close anyway even if 'no' is selected.

To add to my earlier report, pressing the 'X close program' button on the top right of the main main screen will promptly exit without waiting for an answer, but if you are quick, it will also exit even if 'no' is selected.

Both exit issues seem related, and it seems may therefore actually be the same and have the same solution.

Regarding the odometer, just to confirm I did not have any files open when running / testing GCC.
I will keep observing the odometer for any further problems.

Regarding an option to include as currently is the case, or exclude the elapsed time since the end of the session being resumed within the 'stop time', with respect I disagree about it being only useful for a seldom malfunction or accident. I had already provided examples where both options are preferable, dependent on circumstances. I think, and I may be wrong, that most users on extended rides or tours would want to record only time when 'out and about on the bike', and not 8 hours or whatever of sleep within thier stop time. In the absence of any other solution at the moment, I will just work around this by simply not using the resume functionality at all in this instance, instead logging each day individually, converting the resultant .gcc files to .gpx and then combining all the .gpx files together as a single .gpx for the total route. I will then try to import the combined .gpx into GCC and then save as a .gcc file.

Regarding the press and hold, it is just to stop accidental touches causing unwanted commands, that stop the logging. It could be made an option to choose a 'safety' or not so users have the choice and no reason to complain. Alternatively, a 'do you want to' confirmation request similar to that when exiting would achive the same, avoiding any difficulties implementing press and hold. In the absense of any other solution at the moment, I will either tolerate it or work around this also somehow.

I appreciate the difficulites with software development. If you are willing to provide the current beta sourcecode, I am willing to have an attempt at these matters myself and share any success.

low.flying.pigs
 
Last edited:

Top Liked Posts

  • There are no posts matching your filters.
  • 3
    Hello,
    attached the new beta24:
    - copyright string for maps
    - improved "Input LatLon"; additional waypoint is created if you add a name
    - additional option: voice commands only when leaving track for more than 100m (configurable in map options) [feature not thoroughly tested]

    remark to copyright: I thought that it must be present on the screen for legal reasons, but noone needs this information and noone wants the map be covered by this copyright information. Therefore I made it as small as possible and put in in the lower corner where it disturbs not much. But if you want, I can make it in big bold letters in the middle of the screen - here it is. (It only urges people to disable it completely.)

    to my remark of my new android phone: I think it makes no sense to port (write new) GCC to Android. There are already good applications which are already on a much higher level than GCC.
    I will reduce my engagement in developing GCC to a lower level (in the last few weeks it was already very low).

    Regards, Klaus
    2
    Hello low.flying.pigs
    attached the version 4.10beta7, which should fix this error.
    Klaus
    2
    new final version 4.12

    Hello friends of GCC,

    at the dropbox link
    https://www.dropbox.com/sh/69ps6iqg4ajau3r/_cXdntYYU6
    is the new final (and perhaps the last) version 4.12 of GCC.
    There are also some AddOns for GCC at this dropbox link.

    changes from the last beta are:
    - additional option: voice commands only for important messages i.e. when leaving track for more than 100m (configurable in map options)
    - additional option: beep instead of voice commands (especially for devices without voice capability i.e. S3857)

    Regards
    Klaus
    2
    Hello lfp, hello Cheque,

    I have made a first beta with your suggestions.

    - statistics also for kml import
    - improved time unit display when > 24h
    - backlightOff button now on first position in menue. Reactivation with click on left half screen -> main; on right half -> map
    - improved map display while moving map
    - map label color can be configured in Options - Main screen - Select fore/back/mapLabel-color...

    - bugfix T2F import with large files (>4096 points)

    Regarding backlight: Can't you use the hardware power button? How do you reactivate your device? On my device, when I use the backlight button, the touch screen is still working, but you can't see what you are doing. To reactivate my device I had to use the power button. I have improved this now.

    Regarding small black corners sometimes: there is a very sophisticated algorithm to select the best map files to display. This makes sense for arbitrary maps which could overlap each other. For OSM-style map tiles this algorithm is not optimal and has its weaknesses. It could be done much more structured and with less effort. The black corners show especially when some zoom levels are missing, e.g. you have only zoom level 14 and 16. I have now changed the map tile selection for OSM-style maps. Please test if it has some weaknesses.
    Furthermore I will investigate if I can change the behaviour so that the map is displayed 1:1 (and ev. 3:2) pixel of the map to pixel of the display. This will greatly improve the quality of the map presentation, but requires much effort because some things must go the opposite way as now.
    Klaus
    2
    new version 4.5

    Hello,
    version 4.5 of GCC is now online:
    http://code.google.com/p/gccv2/downloads/list
    Jürgen, a new developer, has added a couple of nice features and fixed some bugs:

    ********* version 4.5 *********
    new features:
    - if track2follow and current position can´t be displayed on the same screen:
    * an additional line indicates the shortest way
    * the distance between T2F and current position is shown
    - New context menu on the map screen:
    * show the start / end position of the track to follow,
    * add an check point (only if logging is active)
    * reset map to current position
    - new energy safe options
    * switch GPS off during power off (tracklog will be interrupted, a popup will inform the user)
    * switch Backlight off after timeout
    - if a track log is activated, and in the file dialog the cancel button is pressed, logging is not started.
    - Input Lat/Long can now additionally be used, to navigate to a specific point (e.g. usefull for Geocaching)
    - in map screen: arrow now in color of track line (configurable)

    Have fun
    Klaus