OpenCPN

[idpnd]

Dave - terribly sorry to bore you, the Swedish maps were actually 0 byte files from a faulty compressed file.. Many thanks for your attention regarding these bugs

[dacust]

The file that consistently causes OpenCPN to hang for me is 1R5MA001.000

The log and a screen print of the file location is at opencpndebug

Also, based on what you said, I added a warning on my web page that some of the Netherlands charts may cause problems. But I have been able to look at most of them with no problems.

Please let me know if warnings are needed on any of the rest of them. If we can narrow down the file names, I can post them as well.

I'm already running late to go to work. I'll add a list of the files for each area of charts as a reference.

[idpnd]

I actually get the crash when looking at one of my favourite inland locations in NL, shouwen duiveland (somewhere in the middle-west)

[dacust]

Originally Posted by idpnd

I actually get the crash when looking at one of my favourite inland locations in NL, shouwen duiveland (somewhere in the middle-west)

If you can tell me the chart number, I will put together a download with the problem charts omitted.

Or, Dave, is there a way programatically where you can find some or all of the problem ones?

If not, this evening, if I have time, I'll go through all of them and ferret them out.

-dan

[Kermit]

First, for my first message here, i would give an huge thanks to Dave for his work.

I founded an position error with some bsb chart (ver 1.3.0).
My boat is anchored on a buoy on a river in France near La Rochelle.
To explain it, I've uploaded 4 pictures:
This one show seaclear with the BSB chart:
http://img134.imageshack.us/img134/5585/image1wtn.jpg
The second show opencpn with the same chart:
http://img141.imageshack.us/img141/1349/image2lxd.jpg
The next show opencpn with cm93:
http://img229.imageshack.us/img229/484/image3dbe.jpg

It seem clear opencpn with the bsb chart make a error position. In fact this chart use the european 50 geodesic system and opencpn doesnt correct it.
The last one show seaclear who is be able to do the correction:
http://img151.imageshack.us/img151/3601/image4p.jpg
The informations on the map say:
Positions from GPS WGS84 must be corrected 0,06° to the North and 0,08° to East to be plot on this map.
I guess this informations are in the file.


please mr OpenCpn, do the correction. ;-)

[idpnd]

The course pointer line in Seaclear looks rather pretty as well

[bdbcat]

Re: Holland Charts

Got it. Another bug. This one relates to memory allocation. Dutch charts are extremely detailed in land and coastline features, which is very nice. However, some feature geometries were blowing a static memory buffer in opencpn.
Simple fix, in CVS, will be in 1.3.2.

Enough geek talk.

Thanks all for your direct feedback

Dave

[dacust]

Originally Posted by bdbcat

Re: Holland Charts

Got it. Another bug. This one relates to memory allocation. Dutch charts are extremely detailed in land and coastline features, which is very nice. However, some feature geometries were blowing a static memory buffer in opencpn.
Simple fix, in CVS, will be in 1.3.2.

Enough geek talk.

Thanks all for your direct feedback

Dave

Great!

And, this is a geek thread...

Thanks for all you are doing!

-dan

[bdbcat]

For dacust...

Are you still having trouble on Swiss ENC. Checking my notes....

Dave

[idpnd]

Testing CVS now..

10:29:52 AM: CPL Warning 1: Polygon assembly has failed for feature FIDN=2890,FIDS=1.

This error appears in the log file during initialisation of the NL charts.

10:30:39 AM: Initializing Chart /home/XXXXX/Desktop/charts-live/NL-ENC/Delta_Zuidwest/1R5GR011/1R5GR022.000
10:30:39 AM: Initializing Chart /home/XXXXX/Desktop/charts-live/NL-ENC/Delta_Zuidwest/1R5GR011/1R5GR022.000

The CVS still appears to hang on this file (same as before).

Got a segfault once:

10:41:28 AM: Building SENC file for /home/XXXXX/Desktop/charts-live/NL-ENC/Delta_Zuidwest/1R5MG001/1R5MG001.000 to /home/XXXXX/.opencpn/SENC/1R5MG001.S57
10:41:29 AM:  S57: Unable to find definition for OBJL=17008

10:41:29 AM:  S57: Unable to find definition for OBJL=17011

10:41:29 AM:  S57: Unable to find definition for OBJL=17016

10:41:29 AM:  S57: Unable to find definition for OBJL=17018

10:41:29 AM:  S57: Unable to find definition for OBJL=17021

10:41:29 AM:  S57: Unable to find definition for OBJL=17029

10:41:29 AM:  S57: Unable to find definition for OBJL=17050

10:41:29 AM:  S57: Unable to find definition for OBJL=17051

10:41:29 AM:  S57: Unable to find definition for OBJL=17055

Segmentation fault

Very pleased with the mouse wheel zoom, although I might add that it is highly sensitive and zooms in and out to extreme lengths in no time. Is anyone else experiencing this on X-windows?

[fulup]

I'm also monitoring daily CVS tree, and run on cheap Linux notebooks.

I agree with you, wheel zoom is a great feature. Nevertheless it is a little to fast, and if you have a little processing delay then it zooms from extreme to extreme far too quickly.

Dave (yet an other request ). If you could limit the number of zoom event coming from fifo wheel, it would make this great feature smoother .

For info: OpenCPN run perfectly fine on a small Linux note book. The advantage is that it limit power requirement to ~20W, as well hardware failure (no disk, no fan). I tried both EEEPC-901 and Dell Mini9 (199€ ==> French summer discount 1G+16GSSD), I have a small preference for Dell because of its keyboard and touch panel. EEEPC batterie last twice longer, it suport wifi "n", but on a boat who care .

[dacust]

Originally Posted by bdbcat

For dacust...

Are you still having trouble on Swiss ENC. Checking my notes....

Dave

Nope. I got all of them working. Only got half through the North part of the Netherlands charts, though. I saw your note about those.

[bdbcat]

All...

1. OK, the mouse gets novocaine, or maybe Ritalin. Slowed that baby down, anyway.

2. Dutch chart 1R5MG001.000 fixed. ENC contained light with no (apparent) color. S52 Spec says symbolize as magenta..... OK.

3. 1R5GR022.000 Cannot reproduce failure. Anyone else having trouble with this one?

Comments on mouse wheel performance still welcome. There are some tweak factors in the code....

CVS updated now.

Thanks all
Dave

[svEmerald]

Hi, great programme, I am looking forward to using it.

I'm trying to connect my Bluetooth GPS to OpenCPN. After reading this discussion I have edited the ini file to point to serial port 17, which is the virtual serial port created by the bluetooth programme. When I open OpenCPN, it says it can't open the serial port. I know the GPS etc is working, as I can get it in SeaClear. Any ideas on how to get them to talk?

(oh, it's an XP laptop)

Thanks!

[bdbcat]

For fionavar....

Current version 1.3.0 only looks at COM1-COM16 for available serial ports.
We do this for performance reasons. Enumerating and checking serial ports on Windows is slow, and affects program startup time.

Is there any way to get your created Bluetooth virtual port within the range of COM1-COM16?
How is that port determined in your system?

Maybe we need to look at more ports in 1.3.2. Comments, anyone?

Thanks for your interest....
Dave