Google
  Web www.gpspassion.com


GpsPasSion LIVE!
www.flickr.com
This is a Flickr badge showing public photos from GpsPasSion Live !. Make your own badge here.

www.NaviBlog.com



Versions

Links/Liens




Portal/Portail
Rechercher

- -

Polls/Sondages
Sondage
Pour vous guider sur la Route :
GPS Mobile (SEM)
GPS Intégré
Smartphone
Autre
Voter  -  Résultat des votes
Votes : 2167




Club GpsPasSion
Soutenez le site!

USA: (US$)
EUROPE: (€)
Guide Paypal


GpsPasSion Forums
Home | Profile | Register/Enregist. | Active Topics | Search/Recherche | FAQ
Username:
Password:
Save Password
Forgot your Password?

 All Forums
 GPS Programming
 "Beta" GPS software
 MTK Loggers - Configuration software - BT747
 New Topic  Reply/Répondre
 Printer Friendly
Previous Page | Next Page
Author Previous Topic Topic Next Topic
Page: of 50

mdeweerd

798 Posts

Posted - 18 nov. 2007 :  22:30:57  Show Profile  Reply with Quote
I think that on Mac, the debug information is sent to the java console (as on windows).

If the responsiveness is low without a device connection, then it looks like it is something related to the SuperWaba environment in combination with your java version. Personnaly, I would profile the execution to get a better idea, but that is probably too much for you.
Go to Top of Page

panyo

17 Posts

Posted - 18 nov. 2007 :  22:57:08  Show Profile  Reply with Quote
quote:
Originally posted by mdeweerd

On windows, java has a profiling option built in, but when activating it, the application is really slow.

Anyway, I profiled the application using eclipse (under Java 1.5) and when the GPS NMEA string decoding is active, 75% of the cpu goes to this activity. Deactivating it may therefore help improve performance on your machine.


Ok, I am new to Java development so I don't know the Linux development
tools yet. I think NMEA string decoding on the MacBook is not an issue.

I got librxtxSerial.so for PowerPC from ToyBox (http://rxtx.qbang.org/ToyBox/2.1-7-build1/Linux/glibc-2.3.5/powerpc-750-linux-gnu/ )
and the interface works well on Gentoo on my 450Mhz Cube. I will
see if I can bring the gps unit back to life here, but probably a better tool for the purpose would be a way to directly query and listen to the serial interface while looking up the codes from what you have posted. I think such a tool was mentioned earlier in this thread?

Palm T3, iTrek Z1, Gentoo Linux PowerPC

Edited by - panyo on 18 nov. 2007 23:09:50
Go to Top of Page

mdeweerd

798 Posts

Posted - 18 nov. 2007 :  23:29:47  Show Profile  Reply with Quote
On Linux, you should be able to 'cat' the device path to see what is received.
Go to Top of Page

panyo

17 Posts

Posted - 19 nov. 2007 :  08:10:53  Show Profile  Reply with Quote
quote:
Originally posted by mdeweerd

On Linux, you should be able to 'cat' the device path to see what is received.



tail -f /dev/ttyUSB0
Wow- human readable. That was easy. To send commands, just redirect to the port?

On Gentoo, I was able to set to 5Hz, and back to 1Hz, apparently no
problem.

I was not able to see debugging info in the console on gui selections on the Mac. I have seen output there before, but not for form selections, unless there was an error.

I took the gps out for a drive for the first time since the setback today. Set to 1Hz, 1/1000, it seems that what I have done is to re-introduce the 'static navigation' bug. This is going beyond your program, so I will post in detail in the MTK tweaking thread:
http://www.gpspassion.com/forumsen/topic.asp?TOPIC_ID=81990

Edit:
Ok, looks like the iTrek Z1 is fixed with unknown combinations of setting from Linux and repeatedly removing the battery. I have 5Hz, though it seems to be swamping roadnav and maybe gpsdrive. It might
be of interest that in Linux, tail shows me the USB port output, but
in MacOS, it does not. Furthermore on the Mac,
cat /dev/cu.iBT-GPS-SPPslave-1 shows human readable NMEA strings,
cat /dev/cu.SLAB_USBtoUART does show ascii characters.

Palm T3, iTrek Z1, Gentoo Linux PowerPC

Edited by - panyo on 20 nov. 2007 06:39:44
Go to Top of Page

mdeweerd

798 Posts

Posted - 21 nov. 2007 :  00:17:21  Show Profile  Reply with Quote
NOt having a Mac, it is difficult to experiment.
You could try to connect to the '...iBT...' connection by following the instructions explained in the README of the software, but I think that some people reporte not being able to write to that port.
I think one also needs to be carefull about the order in which ports are used. I read on http://www.hexten.net/wiki/index.php?title=Wintec_GPSBabel_Usage&printable=yes : "You will also have a device called /dev/tty.SLAB_USBtoUART. Do not use the tty device. If you do so the connection will hang forever. "
Go to Top of Page

mdeweerd

798 Posts

Posted - 25 nov. 2007 :  19:11:59  Show Profile  Reply with Quote
Another user (see sourceforge) says the applciaitonis slow for him too on MacOS.
I've investigated a bit further on the profiling capability of java and on windows, I came up with this:

java -agentlib:hprof=cpu=samples,interval=5,lineno=y,file=java.hprof.txt waba.applet.Applet /w 320 /h 320 /scale 1 /bpp 8 BT74

So if you add "-agentlib:hprof=cpu=samples,interval=5,lineno=y,file=java.hprof.txt" to the java invocation, then you should get a 'java.hprof.txt' that hopefully points us in the direction of what makes the application slow on MacOS.
Go to Top of Page

mdeweerd

798 Posts

Posted - 26 nov. 2007 :  19:12:28  Show Profile  Reply with Quote
On MacOS, unzipping the jar files resolves the speed problem for some mysterious reason.
For more info, look here: http://sourceforge.net/forum/message.php?msg_id=4641534 .
Go to Top of Page

mdeweerd

798 Posts

Posted - 26 nov. 2007 :  23:27:30  Show Profile  Reply with Quote
I've uploaded a new version 1.15. While trying to implement another method of reading the serial link, I kind of reproduced a slow interface on windows.
From this, combined with the 'fast' profile report, I made some changes that should improve responsiveness.
Go to Top of Page

panyo

17 Posts

Posted - 27 nov. 2007 :  06:33:20  Show Profile  Reply with Quote
I posted a log from a simple test on sourceforge. The user interface on my Mac has a lag that requires an intermediate action before the UI displays response to inputs.

Also, could the fact that the output from the USB device on the Mac is not ASCII be an issue?
Here is some example output, that is further distorted in
the post preview:
cat /dev/cu.SLAB_USBtoUART
G? ~qc~??NB?j?Bc?B?b
^#62398;SO?<??BF|r??G? ~qc~?c}??jR~B??b~?????S?sb|Yq}?S??G? ~qc~?F??}B?b?xr??_O{b?B#44932;b?h_?G? ~qc~?F??}B?b?xr??_O{b?B#44932;b?h_???B?B??C~?F??}B?b?xr??_O{r?B?~?b?h_???B?B??C~?F??}B?b?xr??_O{b?B?~?b?h_??| ~qc~?F??}B~b?hr???V?{?B>YB}b?R??| ~qc~?F??}B~r?xr???bz?#1727;Bb?h_?^C

Palm T3, iTrek Z1, Gentoo Linux PowerPC
Go to Top of Page

mdeweerd

798 Posts

Posted - 27 nov. 2007 :  19:55:29  Show Profile  Reply with Quote
I would expect readable output on the device - I think someone reported readable output on Mac.

Normally it should not have an effect on the user interface behaviour - especially when the application is not connected yet. For another Mac user, the change I made in 1.15 does make a difference.

Go to Top of Page

MMRizio

9 Posts

Posted - 28 nov. 2007 :  14:36:43  Show Profile  Reply with Quote
Hi mdeweerd,

I am trying to use your application on my PC to program the logging behaviour of my 747. According to my BT application the 747 uses ports 16 and 17 as incoming and an outgoing port. But at both ports my PC refuses to communicate with the 747.

I am using "run_rxtx.bat" after unzipping and the application successfully launches. Whichever port I choose in the application, I always get: "Info: trying to open COM0" followed by a "gnu.io.NoSuchPortException".

Given the large numbers of happy users there must be something small I have overseen, but I can't figure out what...

Maurice
Go to Top of Page

mdeweerd

798 Posts

Posted - 01 déc. 2007 :  23:37:49  Show Profile  Reply with Quote
Hello MMRizio

In windows, choose 'Connect to:' - 'Bluetooth' is to be used only on a portable device.
You can connect to port 16 or port 17.

The problem is however that the application currently works with only one port. Two separate ports (one for in- another for outgoing traffic) are not possible today.
Go to Top of Page

mdeweerd

798 Posts

Posted - 01 déc. 2007 :  23:45:23  Show Profile  Reply with Quote
V1.19 of my software is now online and one user confirms that it solved the lag problem (it introduces a flicker problem with popups, but those do not appear that often).

The SW also has extra functionality: you can force the device to log a position from within the SW (On the easy tab).

On the connection tab, you'll also find information about the flash in the logger. If the flash type is understood by the SW, it will show the manufacturer and possibly the size. Example: 'FlashInfo:C22015C2 (MX,2MB)' . If the flash was not known, only 'FlashInfo:C22015C2' would be displayed. => If you have a flash unknown by my SW, let me know (in this forum or by mail).

Also, besides the MTK FirmWare, the logger capability is another FirmWare with its own version. That is also shown now (e.g., 'Logger: V1.20').

Buttons were re-arranged a bit in order to be able to add this information.
Go to Top of Page

Joop

3 Posts

Posted - 01 janv. 2008 :  18:57:33  Show Profile  Reply with Quote
Happy new year all.

Hi mdeweerd,

I just received the 747 and are trying V1.23 of your software on a Ipaq3970 PPC.
The installation of Superwaba and your app went smooth.

When connecting to COM5 (BT port) I get the following info after switching back to the Con tab:
Lat:xxx Hght:xxx
Lon:xxx Time:xxx (updating every second, except when trying to get a log)
Geoid:xxx
But no firmware ID stuff below that.

After that I do not seem to be able to exchange any information with the GPS. Should it show the GPS log_enabled/disabled status on the log tab? Or is it only meant to set it?
With the Datalogger utility of Transystem I can enable or disable logging and see it persist over several connections. You app does not seem to show me the current status.

Also I cannot Set&Erase the format (waited 5 minutes), or get the log (waited 15 minutes without the progressbar changing from "0 b").
I did enable Debug, but I do not know what to expect in the log. Only the PMTK commands issued to the GPS, or also any PMTK responses? This is the debuglog from connect to getlog and cancelled after about 20 minutes or so:
011696 PMTK605
011737 PMTK182,2,9,9F
011781 PMTK182,2,12
011824 PMTK182,2,12
015333 PMTK182,2,7
018837 PMTK182,2,8
022338 PMTK182,2,10
025847 PMTK182,2,6
029354 PMTK182,2,2
032860 PMTK182,2,12
036360 PMTK182,2,7
039866 PMTK182,2,8
043368 PMTK182,2,10
046871 PMTK182,2,6
050378 PMTK604
053888 PMTK605
057398 PMTK182,2,9,9F
060899 PMTK182,2,12
064402 PMTK182,2,12
067912 PMTK182,2,7
071412 PMTK182,2,8
074920 PMTK182,2,10
078427 PMTK182,2,6
394928 ======= End =======
It is rather unsatisfying when one does not see any progress, error or other response.


Edited by - Joop on 01 janv. 2008 19:04:41
Go to Top of Page

mdeweerd

798 Posts

Posted - 01 janv. 2008 :  20:24:43  Show Profile  Reply with Quote
I observe that reception from your device is ok, but emission is not.

As you are using your PDA, you are likely using bluetooth.

To emit to the 747 over bluetooth, you need to hack your device with a 1k resistor. You'll find a link on the project wiki page.
Go to Top of Page
Page: of 50 Previous Topic Topic Next Topic  
Previous Page | Next Page
 New Topic  Reply/Répondre
 Printer Friendly
Jump To:
GpsPasSion Forums © 2002-2014_GpsPasSion/Manzanite Go To Top Of Page
This page was generated in 1,7 seconds. Powered By: Snitz Forums 2000 Version 3.4.05