Beta Test / Technical

[blubaju]

any daredevil Win7 pilots here?
Better then XP?

[dacust]

I haven't used Win7, yet. But I have a friend that moonlights as a PC guru, going to people's houses to do upgrades, recover PCs, clean viruses, and whatnot. His day job is programming .NET stuff.

As a registered Windows developer, he got Win7 several months ago. He is impressed. Says it is the cleanest Windows, yet. That is, it is as stable as most releases are many months after release. He has done both the 32 and 64 bit versions on multiple machines of his own, and many client machines. Says he has never had to go hunt for a driver (a common problem with new releases of previous Windows versions). He highly recommends it. He was MUCH less enthusiastic about Vista.

Based on his recommendation, next PC I get will be at minimum a dual core with 4 GB RAM and the 64bit Win 7. And I may break down and buy one this weekend.

-dan

[sidetrack]

Originally Posted by GPS-Marco

Thomas, I used your header: OK
Then I replaced into your header your PLYs(4 points) with my PLYs(8 points): Not OK.

I guess the problem is in handling non rectangular borders (on maps crossing 180 longitude).

Have a look.

Ciao, Marco.

P.S. about using MapCal, I had to make some manual corrections since it only allowed me to use East longitudes (175, 180, 185, 190) otherwise it complains of a coordinate swap. Than I changed 190 with -170 and so on by hand directly in the CHARTCAL.DIR... Ciao.

Hmmm, I experience something similar on this chart. With a rectangular border (actually I didn't set the border) of uncorrected coordinates my Chart NZ14605 won't display when centered West of longitude 180.

REF/1,7958,647,-7.500000000000,-167.000000000000
REF/2,3444,647,-7.500000000000,-180.000000000000
REF/3,7958,12221,-37.916666666667,-167.000000000000
REF/4,3444,12221,-37.916666666667,-180.000000000000
REF/5,517,12221,-37.923333333333,-188.428333333333
REF/6,517,647,-7.500000000000,-188.428333333333
REF/7,5181,5097,-19.166666666667,-175.000000000000
PLY/1,-5.85698835326766,-189.902847284021
PLY/2,-39.3462397402451,-189.902826117557
PLY/3,-39.3401363176733,-165.574555869907
PLY/4,-5.85699166414613,-165.574585397281

certainly not an OCPN bug as I've fed it improper coordinates, but definitely a similar effect. If you notice, my calibration points also are not corrected.

[Wester]

Originally Posted by blubaju

any daredevil Win7 pilots here?
Better then XP?

Win7 64Bit and after 6 day's on terrafirma no problems so far.
I will report if i find any, i allso run XP on a different PC

Leif

[Zoef]

Originally Posted by blubaju

any daredevil Win7 pilots here?
Better then XP?

Nothing daredevil about Win7 + OpenCPN; works like Vista. Even tried Tiny7 for about 10 days, no problems, just faster.

Zoef

[blubaju]

Originally Posted by Zoef

Tiny7 .. no problems, just faster. Zoef

Tiny7? thanks for that hint ;-)

New to me, how safe? 10 years XP, per day not so expensive, I will buy win7, less worry.

[bdbcat]

GPS-Marco / Cagney....

Confirmed bug. Code had a hard time finding the min/max longitude if the chart had a PLY shape other than rectangular.

So, yet another mole whacked. Should be OK for the next Beta.

Roadkill, anyone?

Thanks
Dave

[Even Keel]

Incomplete update of GPX file

I don't know if you already correct it and/or if it's already listed but I noticed a very small bug in the update of gpx files :
-name, departure point and arrival point aren't saved
- plan speed is updated.
Synoptic :
- I've saved a route in gpx file by export with name, departure, destination and planned speed .
- I've deleted this route
- I've closed OpenCPN ( to be certain it's not a memory problem).
- I've reloaded OpenCPN.
- I've loaded the saved GPX file by GPS In --> The route is shown.
- I've right click on the route then Route properties
- Route's name is "N.N", depart from and destination aren't filled in.
- Planned Speed is good. It looks like Plannedspeed is a memory parameter as when I change the planed speed for Route A, it changed it for ALL routes. I don't know if it's a fonctionnal choice or a bug. In my opinion, each route should have her planned speed but designer's decision may be different.

I hope it'll help to tune up this fabulous smart free software.

[jonasaberg]

I think this subject should probably go in this thread instead.

Dave,
Regarding my memory leak suspicion.
I dug up an *.nxb file I recorded when we sailed the 300+ nautical mile off-shore sailing race "Around Gotland" in in the baltic sea in 2006.

The nxb file can be used as input into NX2 Race as reported here .
I have then let it run on two computers during 24 hours. (The race took around 50-60 hours or so so I did not let it finish...)

Computer 1.
A fairly modern core2duo (Lenovo T61) dualcore. Same as in the earlier report about dualcore crashes. Using one cpu it ran fine but it consistantly report around 40-50% CPU usage for OpenCPN. It starts at around 30MB memory usage that increase up to a few hundred MB, then drops to maybe 30-50MB and then increase again. It goes in cycles...
I am not sure of this, but it seems the cpu figure never goes above 50% so I suspect 100% is both cores running at full throttle, one cpu can then never go above 50%. So, it seems it need one core at almost full throttle to run Opencpn.

I ran it using track/every ten seconds.

Computer 2:
An older Acer Aspire 1350. About 2004 or so, AMD Mobile CPU 512MB RAM shared with graphics.
This is the dedicated nav notebook we use in the yacht.

I ran NX2 race and OpenCPN using the GR2006.nxb file. It started at 20% CPU, 30M and increased to around 500MB over about 10 hours. The powersave features are set to turn off screen and harddisks after a few minutes, and when moving the mouse to start the screen up it up it seems to trigger some sort of memory clean up since the memory usage dropped to 100MB or so.
The system behaves more and more slowly after a few hours and it wouldn't be very practical to use it while sailing like this, cpu climbs up to around 90 %.

I think it might be the track file, number of track points that makes it slow. Maybe a restriction on how many track points are shown are needed or a different way of updating the screen?

edit:
The number of track points should be a user defined parameter.

Same track setting, every ten seconds.

I have not had time to do more than letting the computer run and taking occasional glances.

If it would be of help or if anyone else would like to simulate using my data I could upload the *.nxb file somewhere. (20MB in a zip file).

/Jonas

[Psyches]

Dave / Jonas,

Not sure if you're using Win or Linux, but if it's Win - here's a fairly new tool that could help diagnose CPU usage: ProcDump

I've been planning but haven't had the chance to try this one out yet, but the other sysinternals tools are great.

Mark

[modul8]

Ok. now I'm trying OpenCPN on a ubuntu distro, on a quad core PC. I am not so sure how to configure the USB bluetooth gps. pairing works fine, but I can't get the software to be aware of the gps' presence.
anyone know a good thread?

[manimaul]

Originally Posted by modul8

Ok. now I'm trying OpenCPN on a ubuntu distro, on a quad core PC. I am not so sure how to configure the USB bluetooth gps. pairing works fine, but I can't get the software to be aware of the gps' presence.
anyone know a good thread?

discover bluetooth gps address:

will@usb-linux:~$ sudo hcitool scan
Scanning ...
    00:1B:C1:02:6B:B5    HOLUX_M-1200

bind to addess

will@usb-linux:~$ sudo rfcomm bind /dev/rfcomm0 00:1B:C1:02:6B:B5

select /dev/rfcomm0 in opencpn toolbox:

[BernardOfCa]

I am not getting a lock on my GPS signal. Here is what I know:

1) Other navigation programs do work OK
2) OpenCPN opens the right port (it is locked for other applications)
3) I am using the NMEA output from a Raymarine autopilot connected through a serial to USB hub from Prolific
4) I have looked at the stream with Tera Term and it looks good

Here is a dump of data collected this weekend:

$IIGLL,3324.06,N,11821.96,W,065051,A*2C
$IIHDG,135.,,,13.,E*17
$IIBWC,065051,,,,,,T,,M,,N,*06
$IIHDM,133.,M*13
$IIHDT,146.,T*11
$IIVTG,000.,T,347.,M,00.0,N,,K*47
$IIGLL,3324.06,N,11821.96,W,065053,A*2E
$IIHDG,133.,,,13.,E*11
$IIBWC,065053,,,,,,T,,M,,N,*04
$IIHDM,130.,M*10
$IIHDT,143.,T*14
$IIVTG,000.,T,347.,M,00.0,N,,K*47

Looks like a bug somewhere. I may try to generate a VC project for the source. If someone has already done that I am interested.

[manimaul]

You might want to look at what you're outputing there. It seems like an odd combo of data to me.

Start with GGA, VTG,ZDA or RMC and only add what you need from there.

[BernardOfCa]

Originally Posted by manimaul

You might want to look at what you're outputing there. It seems like an odd combo of data to me.

I don't really have a choice as the Raymarine autopilot does not give me this flexibility. The software should be able to find the GLL sentence and deal with it. Interestingly, it seems that the heading is correctly decoded as the boat rotate around a fixed position.