OpenCPN Version 4.8.0 Released

[verkerkbr]

Originally Posted by Gilletarom

Hello Dave,

Last night, it took me at least 20 clicks right for the tide window to open.

This morning, new test concerning the display of the tides. In one case out of two, with right clicc, it is the general menu that opens, in one case out of two, it is the window of the tide that opens.

And this is a phenomenon that I had already mentioned in an earlier version.

Gilletarom

On my Raspberry Pi 3 I just installed the latest Debian version called Stretch. The problem with the tidal graph is now gone. I can click on the tidal bar and the graphic image appears without problems.

Although it is not recommanded to update the present Jessie, it seems to worked well.

I don't know if the fingerprint for Oesenc has survived the update.

Cannot try this. Other RPI with the fingerprint is on the boat.

Regards,

Bram

[brucechun]

Hi Dave,

I am working with 4.8 on a HP608 high definition Windows Tablet running Windows 10. The vector of AIS targets under Night Navigation Color Scheme became three time thicker than those of previous OpenCPN version. Attached is a screenshot.



Thanks, best regards,

Bruce

[bcn]

Bruce,

can you check if the detected screen size is the correct one?
Options|Display|advanced

Hubert

[petter5]

Same behaviour under xubuntu 16.04-4 opengl on.

When that is said,this behavior dosen't make any big problem.

Originally Posted by brucechun

Hi Dave,

I am working with 4.8 on a HP608 high definition Windows Tablet running Windows 10. The vector of AIS targets under Night Navigation Color Scheme became three time thicker than those of previous OpenCPN version. Attached is a screenshot.

Attachment 154302

Thanks, best regards,

Bruce

[brucechun]

Originally Posted by bcn

Bruce,

can you check if the detected screen size is the correct one?
Options|Display|advanced

Hubert

Hubert,

OptionDisplayAdvanced is set to Auto...
I am not sure if it is detected correctly.


Referring to Petter5,

It does not related to OpenGL in my case under Windows 10. The behaviors are the same regardless if opengl is on or off.

Bruce

[petter5]

I have a dual display set up with different sizes of each display - same behaviour on both displays. The largest display is detected correctly, but when dragging opencpn from the one to the another display, the behaviour is the same.

Originally Posted by brucechun

Hubert,

OptionDisplayAdvanced is set to Auto...
I am not sure if it is detected correctly.


Referring to Petter5,

It does not related to OpenGL in my case under Windows 10. The behaviors are the same regardless if opengl is on or off.

Bruce

[brucechun]

The behavior could have created by the way that an AIS target is drawn on screen. Apparently, it is drawn by black line surrounded by color lines, regardless if it is daylight or night mode. So, lines marking the vector and the symbol of a target are thick. However, it is not obvious under daylight/default mode.

While changing to night navigation modes, the color scheme changed/inverted the colors of background and other things. The colors lines marking AIS targets are not actually changed/inverted. Thus, making it look very thick as the focus is now drawn on to the lighter lines surrounding the target rather than the black line in the middle.

Regards,

Bruce

[cagney]

Play around with this setting in the opencpn.conf (ini) file

AISCOGPredictorWidth=3

Remember to

Always exit OpenCPN before editing the configuration file.
Edit the file with a text editor, and save your changes.
Restart OpenCPN.

Thomas

[brucechun]

Originally Posted by cagney

Play around with this setting in the opencpn.conf (ini) file

AISCOGPredictorWidth=3

Remember to

Always exit OpenCPN before editing the configuration file.
Edit the file with a text editor, and save your changes.
Restart OpenCPN.

Thomas

Hi Thomas,

No, it does not change anything. It is the same. As mentioned, it seems to be caused by the way that any AIS target, the whole target, was drawn. It is not restricted to COG Predictor. Actually, all lines, including target size outline, related to AIS target were drawn thicker than previous version.

Bruce

[mcric]

having OpenCPN 4.8.0 running native on Mac OSX 10.11.6 (El Capitan) AND on Navigatrix-Linux (3.19.0-75-generic #83~14.04.1-Ubuntu) as Virtual Maschine under VMWare Fusion.
Also having OpenCPN 4.6.1 on Navigatrix-Linux (3.19.0-75-generic #83~14.04.1-Ubuntu ) as Virtual Maschine under VMWare Fusion

All the Systems running paralell on my MacBook.
the 4.6.1 is my Backup-Slave System, get's the gps&Ais via Network from my Master-System (virtual VM Navigatrix-Ubuntu) .
The GPS & AIS - Receiver plugged in as USB-Stick in the virtual Master- Maschine.

On OSX, the OpenCPN 4.8.0 receives it's GPS & AIS data from my Master-Linux OpenCPN 4.8.0 (virtual Maschine)

Both 4.8.0 crashes aftet a couple of seconds after starting.

The last message i got from the logfile:
VMWare-Navigatrix-OpenCPN 4.8.0:
10:45:50 CEST: Datastream connection lost: TCP:0.0.0.0:10110
10:45:55 CEST: Datastream connection lost: TCP:0.0.0.0:2947
10:46:00 CEST: Datastream connection lost: TCP:0.0.0.0:10110
10:46:00 CEST: Datastream connection lost: TCP:0.0.0.0:10110
OSX native OpenCPN 4.8.0
10:39:47 CEST: Datastream connection lost: TCP:10.1.1.12:2947
10:45:46 CEST: TCP Datastream connection established: TCP:10.1.1.12:2947
10:45:46 CEST: TCP Datastream connection established: TCP:10.1.1.12:10110
10:45:50 CEST: Datastream connection lost: TCP:10.1.1.12:10110
10:45:55 CEST: TCP Datastream connection established: TCP:10.1.1.12:10110
10:45:55 CEST: TCP Datastream watchdog timeout: TCP:10.1.1.12:2947
10:46:00 CEST: TCP Datastream watchdog timeout: TCP:10.1.1.12:10110


The OpenCPN 3.6.1 on my backup-Linux never crashes.

Before I updated to 4.8.0 nothing crashed...

I have no idea where the point of error can be ?

[bdbcat]

mcric....

There is way too much complexity in this configuration to analyze from a simple log trace.

Please try to simplify the system configuration, and see if you can isolate the critical point of complexity.

Thanks
Dave

[transmitterdan]

Originally Posted by brucechun

Hi Thomas,

No, it does not change anything. It is the same. As mentioned, it seems to be caused by the way that any AIS target, the whole target, was drawn. It is not restricted to COG Predictor. Actually, all lines, including target size outline, related to AIS target were drawn thicker than previous version.

Bruce

Bruce,

In the opencpn.log file there may be a line like this:

Metrics: m_display_size_mm: 344 wxDisplaySize: 1920:1080

What does this line show in your log file?

[brucechun]

The line showd

Metrics: m_display_size_mm: 120 wxDisplaySize: 1536:2048

The resolution right, but the display size seems wrong. It is a eight inches display.

Bruce

[bcn]

You can put the correct size manually into the field.
Remember: horizontal width in mm (not diagonal)

[brucechun]

If it is the case, then the display size is detected correctly.