Downloading from Edge305

[tdcarter]

Hi,
Using the upgrade version of TF- 2.85. Upgraded from 2.7... I think.

When I download tracks from my Edge 305, ocassionaly the program dies out.

Occasionaly it works, but a different problem then... the last few tracks are broken up so that every trackpoint is a seperate active track.

TF.log shows...
-----------------
DirectDraw : Hardware supports colorkeying
DirectDraw : Hardware supports source colorkey
Creating Surfaces
BackBuffer: Width=186, Height=363, was created in Video Memory
TileBuffer: Width=586, Height=763, was created in Video Memory
Creating Resource Tiles....
Creating Sprites
Successfully Created Surfaces
Initializing Memory Pool
Total Physical Memory: 1071755264
Physical Memory Free: 694571008
Memory Load: 35 percent
Tile Cache Size = 1301
650 surfaces created in (local) video memory
651 surfaces created in system memory
Initializing Threads
Initializing Index File
Sorting Index File.  Number of entries: 30429
Time to sort index file: 0.02 seconds
Initialization Complete

Releasing other Surfaces
Creating Surfaces
BackBuffer: Width=1274, Height=563, was created in Video Memory
TileBuffer: Width=1674, Height=963, was created in Video Memory
Creating Resource Tiles....
Creating Sprites
Successfully Created Surfaces
handling chunked data, bufsize = 8384
writing out response.txt
found chunked in data
curpos = 246446298
curpos = 246446335
chunksize = 4096
moved past chunked line, outpos = 80
chunksize = 4093
moved past chunked line, outpos = 4176
chunksize = 0
processed chunks
Number of Records = 10509
Starting new track - header received, curTrack = 0
Starting new track - header received, curTrack = 1
Starting new track - header received, curTrack = 2
Starting new track - header received, curTrack = 3
Starting new track - header received, curTrack = 4
Starting new track - header received, curTrack = 5
------------------------------

Thoughts?
I was going to try and remove and re-install TF, but my topofusion.msi file is missing and I can't uninstall automated cleanly.

TC

PS I can load the data into Training Center with no problem, and see the data in the unit itself cleanly via history.

[tdcarter]

More info...
When it does download, here is the log entry...
--------------
handling chunked data, bufsize = 8384
writing out response.txt
found chunked in data
curpos = 238057690
curpos = 238057727
chunksize = 4096
moved past chunked line, outpos = 80
chunksize = 4093
moved past chunked line, outpos = 4176
chunksize = 0
processed chunks
Number of Records = 10509
Starting new track - header received, curTrack = 0
Starting new track - header received, curTrack = 1
Starting new track - header received, curTrack = 2
Starting new track - header received, curTrack = 3
Starting new track - header received, curTrack = 4
Starting new track - header received, curTrack = 5
Starting new track - header received, curTrack = 6
Starting new track - header received, curTrack = 7
Starting new track - header received, curTrack = 8
Starting new track - header received, curTrack = 9
Starting new track - header received, curTrack = 25
Starting new track - header received, curTrack = 569
Starting new track - header received, curTrack = 688
Starting new track - header received, curTrack = 1214
Starting new track - header received, curTrack = 1593
Starting new track - header received, curTrack = 1693
Starting new track - header received, curTrack = 1776
Starting new track - header received, curTrack = 2447
Starting new track - header received, curTrack = 3522
Recieved last point. Finshing up
curPont = 1, curTrack = 3974
Transfer complete!
Checking for deleted (zero point) tracks, numtracks = 3975
-------------------

Is the chunked data the problem?
The save/merge screen shows 9 saved tracks and thousands of active tracks.

TC

[Krein]

Thomas,

Thanks for the report.  Looks like we need still more work on the Edge.  I don't have one available right now (getting a forerunner 205/305 soon, which may suffer the same problem).  If the forerunner doesn't shed some light I'll get an Edge or see if you or some other folks are willing to help test.

It should be a fairly simple fix.  I just really need the unit in my hands.

Thanks.

[Krein]

Thomas--

Try the new beta version at:

http://topofusion.com/beta.php

(For registered users)

Demo users can download:

http://topofusion.com/GPSMaster.ocx

Save it to "C:\Program Files\TopoFusion" overwriting the existing file.

Let me know how if it works.  My hunch is that it will fix things for the Edge also.  I assume that you're using the heart rate strap with your edge 305?  It's the heart rate data that was messing things up.

BTW-I can get the HR data now and I'm starting to plan on how to integrate it into TopoFusion.

[tdcarter]

Scott,
I did turn on HR data, also I have tha cadence sensor.

Tried doing the update to beta, no joy.
Get error message of wrong GPSMASTER.ocx. Looking for 2.85, found 2.86.

Yes I did overwrite, I even tried the beta users version.
TC

PS I didn't beleive you earlier when you said it was a problem in TF,  after I saw the message of chunked data. So I did a hard reset of my Edge. Guees what, well, you already know.

[Krein]

Sorry, my mistake on the beta files.

Try it now and you shouldn't get the version message.  It should be 2.86.

It's possible that with cadence enabled it is killing it too.  Grr...

Give it a shot, first.

Scott,
Working beautiful at the moment.

My cadence and HR units are back in St. Louis, and I'm in Ketchikan.

Hopefully I'll be home later this week and able to try 'em.

Thanks for the quick revision.
TC

PS So will the new Fore units have to be 'reset' before you can se and download the active track?

[Krein]

Cool, thanks for trying it.  I'll wait to hear more about the addition of HR and Cadence.

Yep, it looks like you have to "reset" before the garmin track protocol can "see" the tracks on the Forerunners.  I still need to find out what the total capacity of the "history" is.  I captured a ~3 hour ride that was about 9000 points, and even after doing another rider it still downloaded at full resolution.

[tdcarter]

Scott,
Okay, got home and tied the Edge to the bike and myself.

And as you thought, something dosen't like something in either the HR or cadence.

Log file shows...
-------------------
Number of Records = 4757
Starting new track - header received, curTrack = 0
Starting new track - header received, curTrack = 1
Starting new track - header received, curTrack = 2
Recieved last point. Finshing up
curPont = 1, curTrack = 7
Transfer complete!
Checking for deleted (zero point) tracks, numtracks = 8
Checking for deleted (zero point) tracks, numtracks = 8
-------------------

First time I tried to download, it blew away. I fired it back up before I thought of looking at the log.

This time when it did succesfully downloaded, the track was broke into a few different segments. It wasn't a long ride, tomorrow I'll try and get some more data trials for analysis

WOW! 3 hour ride gets ya 9,000 points. I did 2,500 miles and only got 4,500 points.
TC

PS If I download to TC before TF, I don;'t have to do the reset. I guess TC is doing a reset for us? Oui?

[tdcarter]

Hey folks,
Not sure if this is indiciticve of a problem on my end/my windows installation. Or if this is releated to the whole issue of downloading from the Edge 305 w/ cadence and HR data.

But... I dug out my Forerunner 201 and used i the other day (seems the Edge dosen't like preassurized planes) and when I download from the 201, it also breaks the track up into lots of individual points. Ya know, like a 1000 active logs.

Also.. again not sure what it points too, but I keep getting some datum points that have lat/long of 180 and 180, respectively.  

TC

[trouble]

It looks like the problem might be with the cadence information. I just added the cadence sensor to my Forerunner 305 - (I use it while riding also) and voila...a whole bunch of really short seperate tracks for the ride.  v2.86 had taken care of the HR info creating lots of little mini-tracks but it looks like the addition of the cadence sensor creates more problems.

I've got v2.86 and had been having no problems with pulling off the tracks until I enabled the cadence sensor. Obviously, I can just disable the cadence sensor for now but I thought I'd let you know.

[Krein]

I really appreciate the report on the cadence sensor.  Looks like I might have to buy one, too!

Too bad Garmin make their protocol spec public.  Grr...

Actually I think I might be able to conjure up a fix for this.  I'll let you guys know if I have a new beta test.

[tdcarter]

Thanks for the updates you have done.

If you need some sample data, or a beta beta tester... holler at me.

TC

[Krein]

Hey Thomas,

I've been away, racing bikes and traveling for the past month.  This is long overdue, but if you could try downloading with the Edge + Cadence using the new beta (v2.87) I'd most appreciate it.

It most likely will not work (many tiny tracks as before), but if you can send me the TopoFusion.log output I should be able to get it to work in about 2 minutes.

Thanks a lot.

[tdcarter]

Scott,
Okay, installed the latest beta.

Same ressonse ion the download log...lots of broiken out track points and after merging the tracks I get some of those 180x180 lat/long cordinates.

I won't post the whole log, as it's 1.2 meg. If you want I can zip and send it to ya.

The first few lines, till it starts getting repetitve, are...
-----------------------------
TopoFusion 2.88 Log Begin
Windows Major Version: 5
Windows Minor Version: 1
Windows Build Number : 2600
Windows Platform ID  : 2
Windows CSD Version  : Service Pack 2
Initializing Winsock
Initializing DirectDraw
Color Info : BBP = 32
Color Info : Sizes : R=8, G=8, B=8
Color Info : Shifts : R=16, G=8, B=0
DirectDraw : Driver "ATI MOBILITY RADEON X600  " (ati2dvag.dll)
DirectDraw : Total Video Memory : 318133312
DirectDraw : Free Video Memory : 318133312
DirectDraw : Hardware supports blits
DirectDraw : Hardware supports asynchronous blits
DirectDraw : Hardware supports colorfill on blits
DirectDraw : Hardware supports stretchblit
DirectDraw : Hardware supports colorkeying
DirectDraw : Hardware supports source colorkey
Creating Surfaces
BackBuffer: Width=186, Height=363, was created in Video Memory
TileBuffer: Width=586, Height=763, was created in Video Memory
Creating Resource Tiles....
Creating Sprites
Successfully Created Surfaces
Initializing Memory Pool
Total Physical Memory: 1071755264
Physical Memory Free: 678125568
Memory Load: 36 percent
Tile Cache Size = 1270
634 surfaces created in (local) video memory
636 surfaces created in system memory
Initializing Threads
Initializing Index File
Sorting Index File.  Number of entries: 3962
Time to sort index file: 0.00 seconds
Initialization Complete

Releasing other Surfaces
Creating Surfaces
BackBuffer: Width=1274, Height=563, was created in Video Memory
TileBuffer: Width=1674, Height=963, was created in Video Memory
Creating Resource Tiles....
Creating Sprites
Successfully Created Surfaces
Unit type = EDGE305 Software Version 2.60
Supported protocols/data types:
P - 0
L - 1
A - 10
T - 1
A - 100
D - 110
A - 201
D - 202
D - 110
D - 210
A - 302
D - 311
D - 304
A - 500
D - 501
A - 600
D - 600
A - 601
D - 601
A - 700
D - 700
A - 800
D - 800
A - 801
D - 801
A - 902
A - 903
A - 907
D - 907
D - 908
D - 909
D - 910
A - 918
D - 918
A - 1000
D - 1009
A - 906
D - 1011
A - 1002
D - 1008
A - 1003
D - 1003
A - 1004
D - 1004
A - 1005
D - 1005
A - 1006
D - 1006
A - 1007
D - 1007
A - 1008
D - 1012
A - 1009
D - 1013
newTrackLoc = 22
Number of Records = 11967
Starting new track - header received, curTrack = 0
packet[17] = 0
packet[18] = 0
packet[19] = 0
packet[20] = 0
packet[21] = 255
packet[22] = 1
packet[23] = 255
Bogus Cadence or HR point? Lat = 180.000000
packet[17] = 0
packet[18] = 0
packet[19] = 0
packet[20] = 0
packet[21] = 255
packet[22] = 1
packet[23] = 255
Bogus Cadence or HR point? Lat = 180.000000
packet[17] = 89
packet[18] = 4
packet[19] = 105
packet[20] = 0
packet[21] = 255
packet[22] = 0
packet[23] = 255
Bogus Cadence or HR point? Lat = 180.000000
packet[17] = 89
packet[18] = 4
packet[19] = 105
packet[20] = 0
packet[21] = 255
packet[22] = 0
packet[23] = 255
Bogus Cadence or HR point? Lat = 180.000000
packet[17] = 0
packet[18] = 0
packet[19] = 0
packet[20] = 0
packet[21] = 0
packet[22] = 0
packet[23] = 255
-----------------------------------
etc...... etc.......etc.....
and then..........
-----------------------------------
Recieved last point. Finshing up
curPont = 0, curTrack = 162
Transfer complete!
Checking for deleted (zero point) tracks, numtracks = 162
Releasing other Surfaces
Creating Surfaces
BackBuffer: Width=730, Height=305, was created in Video Memory
TileBuffer: Width=1130, Height=705, was created in Video Memory
Creating Resource Tiles....
Creating Sprites
Successfully Created Surfaces
---------------------------

Let me know what else I can do.
Hope you enjoyed the ride, I was also out all of last week on a ride. Found out that if your using the cadence and HR monitor, the stored data gets dumped and it just save the summed data for the overall run. Oh well, another thread and not a TF problem.

Thanks,
TC

[Krein]

Thomas,

Thanks for testing.  Made some progress in 2.88, but it looks like there was still one more step (hopefully).

I think it should work this time.  I don't know what those bogus points with 180's are.  This time we shouldn't get so many track segments all split up.  Hopefully none of the 180 points will make it either.

It's still 2.88 since it was one line of code that I changed and I didn't want to do a whole new version.

I'm crossing my fingers this time.

What was your week long ride?  Where?  Details!

Thanks.

[tdcarter]

Hi,
Thanks for asking... but my week of riding probably wasn't anywhere near as thrilling as yours. The most exciting thing was what didn't happen.
Spent the week riding (and camping) the Katy trail across Missouri. The female companion (who is not into camping) is still speaking to me after the ride. It's some of the small victories that are important.

For TF and the beta 2.88B

SUCCESS!!!!!!!

YEAH!!!!

The download didn't die out part way through. I even tried it a few times to make sure.

Ohhh crud... it did just die during a download.

Okay, here are the starts and ends of the log,
------------------------------------------------------

Releasing other Surfaces
Creating Surfaces
BackBuffer: Width=1594, Height=963, was created in Video Memory
TileBuffer: Width=1994, Height=1363, was created in Video Memory
Creating Resource Tiles....
Creating Sprites
Successfully Created Surfaces
Unit type = EDGE305 Software Version 2.60
Supported protocols/data types:
P - 0
L - 1
A - 10
T - 1
A - 100
D - 110
A - 201
D - 202
D - 110
D - 210
A - 302
D - 311
D - 304
A - 500
D - 501
A - 600
D - 600
A - 601
D - 601
A - 700
D - 700
A - 800
D - 800
A - 801
D - 801
A - 902
A - 903
A - 907
D - 907
D - 908
D - 909
D - 910
A - 918
D - 918
A - 1000
D - 1009
A - 906
D - 1011
A - 1002
D - 1008
A - 1003
D - 1003
A - 1004
D - 1004
A - 1005
D - 1005
A - 1006
D - 1006
A - 1007
D - 1007
A - 1008
D - 1012
A - 1009
D - 1013
newTrackLoc = 22
Number of Records = 11967
Starting new track - header received, curTrack = 0
length = 24
Bogus Cadence or HR point? Lat = 180.000000
length = 24
Bogus Cadence or HR point? Lat = 180.000000
length = 24
Bogus Cadence or HR point? Lat = 180.000000
length = 24
Bogus Cadence or HR point? Lat = 180.000000
length = 24
packet[17] = 0
packet[18] = 0
packet[19] = 0
packet[20] = 0
packet[21] = 0
packet[22] = 0
packet[23] = 255
length = 24
packet[17] = 74
packet[18] = 0
packet[19] = 64
packet[20] = 0
packet[21] = 0
packet[22] = 0
packet[23] = 255
length = 24
packet[17] = 129
packet[18] = 227
packet[19] = 64
packet[20] = 0
packet[21] = 32
packet[22] = 0
packet[23] = 255
length = 24

(etc...etc...etc...)

length = 24
packet[17] = 7
packet[18] = 189
packet[19] = 70
packet[20] = 107
packet[21] = 66
packet[22] = 0
packet[23] = 255
length = 24
packet[17] = 67
packet[18] = 189
packet[19] = 70
packet[20] = 112
packet[21] = 79
packet[22] = 0
packet[23] = 255

Caught Exception:
 Function: Unhandled
 Exception: Access Violation


--# FV EIP----- RetAddr- FramePtr StackPtr Symbol

  0 .V 1003cea5 1003cef3 0d71f554 0d71f4ec getPacketUSBGarmin +373 bytes    Sig:  ?getPacketUSBGarmin@@YAHPAEPAK_N@Z
    Decl: int cdecl getPacket(unsigned char *,unsigned long *,bool)
    Line: C:\GPSMaster\checkout\GPSMaster\serial.cpp(480) +17 bytes
    Mod:  GPSMAS~1[C:\PROGRA~1\TOPOFU~1\GPSMAS~1.OCX], base: 10000000h
    Sym:  type: CV, file: C:\PROGRA~1\TOPOFU~1\GPSMAS~1.OCX

  2 .V 1003e479 1002b746 0d71fa9c 0d71f4ec downloadGarmin +1449 bytes    Sig:  ?downloadGarmin@@YAHFPAUlayer@@_N11@Z
    Decl: int cdecl gpsComm(short,short,short,short,char *,bool,bool,bool)
    Line: C:\GPSMaster\checkout\GPSMaster\master.cpp(3304) +17 bytes
    Mod:  GPSMAS~1[C:\PROGRA~1\TOPOFU~1\GPSMAS~1.OCX], base: 10000000h
    Sym:  type: CV, file: C:\PROGRA~1\TOPOFU~1\GPSMAS~1.OCX

  4 .V 1002c233 100a6871 0d71ff14 0d71f4ec GPSThread +179 bytes    Sig:  ?GPSThread@@YAIPAX@Z
    Decl: unsigned int stdcall _AfxThreadEntry(void *)
    Line: thrdcore.cpp(112) +5 bytes
    Mod:  GPSMAS~1[C:\PROGRA~1\TOPOFU~1\GPSMAS~1.OCX], base: 10000000h
    Sym:  type: CV, file: C:\PROGRA~1\TOPOFU~1\GPSMAS~1.OCX

  6 .V 1008d0d9 7c80b50b 0d71ffb4 0d71f4ec beginthreadex
    Decl: __beginthreadex
    Mod:  GPSMAS~1[C:\PROGRA~1\TOPOFU~1\GPSMAS~1.OCX], base: 10000000h
    Sym:  type: CV, file: C:\PROGRA~1\TOPOFU~1\GPSMAS~1.OCX

  7 .V 7c80b50b 00000000 0d71ffec 0d71f4ec GetModuleFileNameA +436 bytes    Sig:  GetModuleFileNameA
    Decl: GetModuleFileNameA
    Mod:  kernel32[C:\WINDOWS\system32\kernel32.dll], base: 7c800000h
    Sym:  type: -exported-, file:

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

So there is the latest error... and it looks like the 180 lat is still showing up, are you just dumping it now? Now that I think about it... I do believe there are some 'unwanted points' that are thrown out in the download into Garmins programs. Hmmm.... I'll have to look into those logs.

Okay... I also tried download from the older 201. Seems to be working beautiful. Tried three downloads, no faults.

Looks like there was considerable progress on the last round. Muchas Gracias.

So where was your week of riding at?

TC

PS Now that all the tracks are on the download screen without me having to do a bunch of merge/deletes, I see that I exceeded the 10,000 points after three days of riding. So I guess it makes sense that it dumped the detail data of the previous 2 days rides. Course I'd rather see it cut it down to maybe the 250 ponts of the older series then just dump it all. Sigh.

[tdcarter]

Hey... that PS just clicked.

There are tracks in my units memory that have no detail data. Will that cause a problem for TF? I don't see those days showing up in the save track dialog box.

TC

[Krein]

Quote (tdcarter @ June 30 2006,12:37)

Hi, Thanks for asking... but my week of riding probably wasn't anywhere near as thrilling as yours. The most exciting thing was what didn't happen. Spent the week riding (and camping) the Katy trail across Missouri. The female companion (who is not into camping) is still speaking to me after the ride. It's some of the small victories that are important. For TF and the beta 2.88B SUCCESS!!!!!!! So where was your week of riding at? TC PS Now that all the tracks are on the download screen without me having to do a bunch of merge/deletes, I see that I exceeded the 10,000 points after three days of riding. So I guess it makes sense that it dumped the detail data of the previous 2 days rides. Course I'd rather see it cut it down to maybe the 250 ponts of the older series then just dump it all. Sigh.

Thanks for the report, including the crash.  I'll see if I can figure out what was wrong.  Sometimes the USB downloads are a little flaky, so I'm not too worried about it dying partway.  But access violation is no good!

How many miles on the Katy Trail?  What kind of setup did you run (panniers, bob, etc).  Bummer that you lost data due to the ridiculous 10k (slightly more in 205/305s, but only marginally) limit on Garmin units.  It really burns.  Glad you and the missus came out of the trip on good terms.  That's pretty rare, I think.  You've got a keeper...

My ride was only slightly more than 3 days, but it felt like a week and I'm still recovering from it.  It was 'the Grand Loop', a 360 mile, unsupported race in Grand Junction/Moab.

The long and gory story is here:

http://forums.mtbr.com/showthread.php?t=199904

[Krein]

Quote (tdcarter @ June 30 2006,1:00)

Hey... that PS just clicked. There are tracks in my units memory that have no detail data. Will that cause a problem for TF? I don't see those days showing up in the save track dialog box. TC

Shouldn't cause any problems, and there's no way that I know of (no documented way, that is) to retrieve that data, either.

Yep, the least it should do is simplify them into saved tracks so that you have *something* from the ride.

[tdcarter]

Okay... old problem is still there. I was hoping it would go away when I rebuilt the laptop.... but it hasn't. It still exists with the 2.98 beta D3D.

Problem is occasionaly *poof* during downloading mode. Fire them program back up, and it usually gets it the second time. Once it did take a few tries, three I think.

But either way, I have the logs, but they both start the same way, then once just abruptly ends in the middle of the download.... and I have a bunch of those "Bogus cadence or HR point....." iun both logs.

If you have a debug version with full logging ya want me to try, let me know.

Other then that, like the new D3D.

Thanks guys,
TC