OpenCPN Version 2.1.0 Beta

[Ptizef]

Hi Dave

The strange behavior I described in my #313 (post 3) can be easily reproduced : It happens when all eyes available are off .
See shot attached : even if a cm93 chart is displayed , the two eyes are off . Try to zoom in this condition and the cursor become crazy and it happens what you can see .

best regards
jean pierre

[cagney]

Originally Posted by bdbcat

Will....


Anybody else got any input on this design issue? I'd rather not code to moving target.

Thanks
Dave

I think the present design is pretty good. I suggest that we stay with it, at least for now. Users need more time to test this while actually under way.
Possible changes needs to come as a result of more experience with the present design.
I also think that it's OK to not quilt tmercs at all, for the next stable release. Very few places, except for Sweden(?), has larger areas contiguously covered with tmercs.

Regarding Ptizef's post #313 & #316.
I cannot repeat #2 and #3.

Thomas

[oem]

I still (with build 426) have occasional (too frequent) crashes. I had a crash with this log-file:

23:22:17: 2010-04-30
23:22:17: -------Starting opencpn-------
23:22:17: Version 2.1.0 Build 426
23:22:17: SData_Locn is C:\Program Files\OpenCPN\
23:22:17: Using existing Config_File: C:\Users\oem\AppData\Roaming\opencpn\opencpn.ini
23:22:17: Setting Viewpoint Lat/Lon 55.721, 9.5538
23:22:17: Setting Ownship Lat/Lon 55.7209, 9.5538
23:22:17: System default Language: da_DK
23:22:17: Opencpn language set to: da_DK
23:22:17: Using s57data in C:\Program Files\OpenCPN\s57data
23:22:17: Using Tide/Current data from: C:\Program Files\OpenCPN\tcdata\
23:22:17: Setting Viewpoint Lat/Lon 55.721, 9.5538
23:22:17: Setting Ownship Lat/Lon 55.7209, 9.5538
23:22:17: NMEA Data Source is....Serial:COM3
23:22:18: AIS Data Source is....None
23:22:18: Using WVSChart datafile: C:\Program Files\OpenCPN\wvsdata\wvs43.dat
23:22:18: NMEA AutoPilot Port is....None
23:22:18: Loading chart db version: V016
23:22:18: Chart directory D:\data\Drabant 22\søkort\OpenCPN\kms
23:22:18: Chart directory D:\data\Drabant 22\søkort\OpenCPN\ENC_ROOT
23:22:18: GPS Watchdog Timeout is: 6 sec.
23:22:18: Initializing Chart D:\data\Drabant 22\søkort\OpenCPN\kms\S_114_75000.kap
23:22:18: Initializing Chart D:\data\Drabant 22\søkort\OpenCPN\kms\S_F_850000.kap
23:22:18: Initializing Chart D:\data\Drabant 22\søkort\OpenCPN\kms\S_103_200000.kap
23:22:18: OpenPD: 324, new_count = 1 for COM3
23:22:27: Initializing Chart D:\data\Drabant 22\søkort\OpenCPN\kms\S_151_75000.kap
23:22:28: Initializing Chart D:\data\Drabant 22\søkort\OpenCPN\kms\S_102_200000.kap
23:23:09: Initializing Chart D:\data\Drabant 22\søkort\OpenCPN\kms\S_157_25000.kap
23:23:11: Initializing Chart D:\data\Drabant 22\søkort\OpenCPN\kms\SK_BRJNG.kap
23:23:11: VP Final Check: Georeference Chart_Error_Factor on chart D:\data\Drabant 22\søkort\OpenCPN\kms\SK_BRJNG.kap is 1.00088

So it seems that the crash has something to do with quilting (since I'm using that) and with the error in the log-file, since also OpenCPN never showed the SK_BRJNG.kap - chart.

All my charts were created with the lib2bsb-routines as documented in this forum. KAP-file attached. Running Vista

Any help?

[cagney]

oem

Nothing in your logfile can explain your crashes. I downloaded your chart and it displays and quilt quite normal.
What you experience is probably one of the crash causes already reported. To pinpoint this you really have to build your own OpenCPN from source code (normally including debug symbols), and run your development environment in debug mode. Check for example this post by Ptizef.
This helps Dave and the other developers to pinpoint the cause of the crash. We definitely need more windows user with the ability to produce this kind of screen dumps.
In Linux... it's easier to set up,.....and it's free.

Thomas

[oem]

Originally Posted by cagney

oem

Nothing in your logfile can explain your crashes. I downloaded your chart and it displays and quilt quite normal.
What you experience is probably one of the crash causes already reported. To pinpoint this you really have to build your own OpenCPN from source code (normally including debug symbols), and run your development environment in debug mode. Check for example this post by Ptizef.
This helps Dave and the other developers to pinpoint the cause of the crash. We definitely need more windows user with the ability to produce this kind of screen dumps.
In Linux... it's easier to set up,.....and it's free.

Thomas

Thomas, thanks indeed for your answer. Without quilt everything is ok as far as I have experienced. I thought it might have something to do with the chart error message in the log-file... I will continue to investigate/troubleshoot, it will take me some time to establish sufficient debugging knowledge and VC knowledge. I might try sometime.

It has something to do with the event of loading new charts into the quilting space, - I'm sure of that. But I agree, that under other circumstances the above chart might come up correctly. Which also means that it can have something to do with the charts already in the quilting space.

[cagney]

I was reading a bit in John P. Snyders book "Map Projections - A Working Manual", and came across the piece in the picture below.
This raises the question- Is it even possible to successfully quilt tmercs without re-projecting them, either to a common central meridian or, in our case, to normal merc charts ?
Thomas

[Ptizef]

Hi Oem,Cagney,Dave ,

I tried to use the chart in debug mode in V++/vista

1) The chart was displayed in both quilting and no quilting conditions with and without cm93 . But when there is another bsb with a very smaller scale , impossible to quilt it : see the screen shot test1 : there a bsb 1/15000000 , cm93 and bsb big scale SK_BRJN
I can't zoom + more and impossible to see the big scale . May be it is not a problem coming from this chart , but eventually a limit in the code ?
2) Playing with , rebuilding the base , using F9 ... I got many crashes . See screen shots crash3,crash4,crash5,crash6 are exemples

I hope this will help
regards
Jean Pierre
regards

[Ptizef]

Hi Dave

I got a crash witch can be reproduced : open opencpn , if not , enable quilting , re-build the chart base then press F9 = crash
If after re-building , you exit en re-open , no crash
see shot attached

regards
Jean Pierre

[oem]

Originally Posted by Ptizef

Hi Dave

I got a crash witch can be reproduced : open opencpn , if not , enable quilting , re-build the chart base then press F9 = crash
If after re-building , you exit en re-open , no crash
see shot attached

regards
Jean Pierre

Thanks for the help. I have isolated 4 charts from the area:

http://dl.dropbox.com/u/5863278/S_103_200000.kap
http://dl.dropbox.com/u/5863278/SK_BRJNG.kap
http://dl.dropbox.com/u/5863278/S_157_25000.kap
http://dl.dropbox.com/u/5863278/S_114_75000.kap

And then I tried to reproduce the error with only these 4 charts in different scales. It's not consistent, but scaling up and down in both quilt and non quilt mode can and will make the program crash after rather few attempts.

All charts built with tif2bsb-routines.

Log-file does not say more than in my post above.

[gaea]

I've installed the beta package in XP and Hungarian Danube charts because I'm in Hungary right now.
I got the charts from:
http://www.pannonris.hu/modules/ecdi...es.php?lang=en
I opened these charts in both Opencpn and SeeMyEnc so I could compare.
And there is a lot of the symbols missing like kilometer marks, bouys, signs on bridges and banks and so on.
Specially the kilometer marks are important; one can anounce himself on the radio like: ' mv Lyaemer sailing upstream in Mohacs kilometer 1447...'
Is often done, and always in foggy or nightly conditions.
Red and green lights are shown as circles.
I hope this helps a bit.

[BluegillBoy]

LOVE the quilting.
i also get a crash everytime i force a database rebuild and then change scale, either larger or smaller. sorry i don't know how to run debug to help any further.
another issue - when i first start up, the chart doesn't click into hi-res. stays hard to read. zoom in and it does the usual lo-res then hi-res after a few seconds and works fine after that. it's just upon initial star.
regardless, this is great! thanks.

[nohal]

Originally Posted by gaea

I've installed the beta package in XP and Hungarian Danube charts because I'm in Hungary right now.
I got the charts from:
http://www.pannonris.hu/modules/ecdi...es.php?lang=en
I opened these charts in both Opencpn and SeeMyEnc so I could compare.
And there is a lot of the symbols missing like kilometer marks, bouys, signs on bridges and banks and so on.
Specially the kilometer marks are important; one can anounce himself on the radio like: ' mv Lyaemer sailing upstream in Mohacs kilometer 1447...'
Is often done, and always in foggy or nightly conditions.
Red and green lights are shown as circles.
I hope this helps a bit.

Hi Gerard,
just checked Danube km 1447 and it looks quite as expected - some symbology rules missing (so we have question marks instead of the red-green buoys between km 1446 and 1447 etc. - obvious, known and easy to fix), but as far as I can tell, we have all the data I'm aware of there - if you want to see the kilometers, you have to turn the ENC text on ("A" icon with the plus sign next to the wrench icon selected + zoom to the level where the numbers are displayed according to the IENC standard) For the rest I need more information and, if possible, exact coordinates of the problem, as then I can have a look and investigate myself if it's a chart available publicly for download. Right now just don't waste time with reports like "There is a small blue square instead of sign XY" as not all the signs are in the ruleset now and I know about it - there will be time for completion of the ruleset later. What's important is stuff that's really missing completely (the bridge signs are of utmost importance here, but there is no bridge on this chart so I couldn't investigate what are you talking about here). The more precisely you specify the problem, the more likely it will be fixed soon.
And don't be too fixed to the look of SeeMyEnc, it's for sure more advanced than OpenCPN right now and looks cute, but is quite far from the Ecdis standard and has it's own errors at the same time...

Cheers

Pavel

[bdbcat]

Hello there...

OpenCPN Version 2.1.0 Beta build 501 is ready for download and test.

Changelog:

83. Correct quilt reference scale calculations for some S57 ENCs, e.g Chicago area....
84. Correct logic error for NMEA $xxGLL sentences with no checksum appended.
85. Correct for crash after database rebuild in quilt mode.
86. Limit Zoom-in range to sensible value.
87. Correct Windows RouteManagerDialog display and sorting method.
88. Correct Windows ChartInfo Popup window text formatting.
89 Implement new quilting control algorithm
90. Correct ship's bell .wav files for linux application.



Notes:
89. New simplified hybrid quilting UI. It works like this:
a. Chart selector bar at bottom is familiar model, representing all charts in the database containing the center point (lat/lon) of the screen.
b. Choosing a chart on the bar (by clicking) builds a quilt with following properties:
The chart chosen, plus all charts of smaller scale of the same type, projection, and skew are shown in the quilt.
c. Right click menu on the bar allows excluding/including a particular chart from the quilt. Not currently saved in config...
d. Charts appearing in the quilt show brighter color bars, as with single chart mode, except there are several shown at once.
e. Note that a small scale chart in the bar may be available for the quilt, but be eclipsed by larger scale charts. In this case the brighter bar color may be confusing, since the chart is really not visible. We may need a special icon or color pattern to indicate this case.
f. As a consequence of this new algorithm, it is not possible to combine chart types (Raster/Vector/Cm93) on one quilt, with one useful exception.
g. Exception to (f). If the quilt does not cover all of the screen as composed, and cm93 charts are available, then cm93 will be used to fill out the uncovered region of the screen.
h. We have eliminated the Toolbox quilt chart type options and the viz/inviz "eye" icon.

One problem I have not quite worked out is this: US NOAA ENC's have the unfortunate property that small scale charts contain "holes" where larger scale charts are available. Have a look at Chicago in SCM. Problem: If the ViewPort center is not exactly within a particular large scale cell, then that cell is excluded from the quilt, leaving a hole in the smaller scale chart containing it. Not what we want. So, I coded a special case for S57ENC. I add to quilt all ENC cells which overlap the ViewPort in any way. I generally don't like special cases. They come back and bite later...

So, there we have it. Another UI concept to debate. I'm looking for something that is really new here, not a simple clone of other existing commercial products. Do try to keep an open mind on this, and feedback is very welcome here.

This is Version 2, after all....


Other notes:
TMerc Quilting: Still not right, but I have not given up yet.....
i18n: I have limited bandwidth, so am not incorporating incremental Beta Build .po files. Lets wait until we are about ready to release and update the .pot and .po/.mo all at once.

Release schedule: Target is June 1 for Version 2.1.0 release.

CVS Updated.

Thanks, and Good Luck
Dave

[MarkJ]

OK starting on a 25 mile run in 1 1/2 hours.

I'll tell you if it keeps us in the water or runs us over terra firma

[oem]

Originally Posted by bdbcat

Hello there...

OpenCPN Version 2.1.0 Beta build 501 is ready for download and test.

Dave

Thanks..

Of course immediately I tried to reproduce my error in post http://www.cruisersforum.com/forums/...tml#post444723. I can still easly make the program crash with the four charts scaling up/down. Still not consistent, but still rather easy to provoke.

I have tried several times to delete the ini-file, and also tried to rebuild the database several times. This seems to help, but after some time the program crashes again. Could it be a problem with some data in the chart database?

It can crash with and without quilt.

Testing around with the four charts can corrupt the program so much that even the world map does not display correctly, but displays a number of horisontal blue lines and the 'Viewpoint' is changed to eastcoat of America.

I don't know if it would help uninstalling the program completely and then re-installing? I'm running the 501 build for sure.

I tried to unplug my ND-100 gps, suspecting data from that to be the cause, but I don't think so. Seemingly correct (=ascii readable) NMEA-data is transferred and the unit works fine in Polarview and GE.

The program can be corrupted so much that it crashed (=freezes) before any button is hit. But then sometimes I'm fast enough to hit the 'Toolbox'-button and then typically rebuild the database, quit and then successfully restart for some time.

I'm sorry to give up right now, - I'll try later again

It's an incredible work, though, the progress and the improvements in the program. It's still like opening a cristmas present