OpenCPN

[sinbad7]

LeaseOnLife..

Sindbad, wonderful boats, but please take this to another thread!

Ooops!! Sorry for the intrusion.. a little off topic,but I WAS hoping perhaps to put OpenCPN to follow the race by AIS,which,after all should be of some interest here. Judging by previous 'off-topic' messages,I am not the only one.

As some of our techno guys seem to be under some stress,perhaps we should set up a fresh thread for us mere mortals? On the other hand,as the saying goes: "Large streams from little fountains flow!"

[bhcva]

Scotte....I have succesfully used SeaClear outputting nmea onto the seatalk network thru the convertor (Raymarine product)....works just fine. I haven't found a way to make Opencpn send the data. Bruce

[Ptizef]

Dave,
I test for a few days 133 903 then 911 and I use Vista.
Two problems was detected :
1)The "Send to GPS" function doesn't work any more .When clicking on it , it happens nothing . It was OK in 132 .
2)The keys F10 and F11 don't works well . Sometimes , when pushing on one , it's happen nothing and I have to push many times . It was the same in 903
Also I would like to suggest some improvements :
1) Active routes :
Possibility to see in a small pop up window the bearing and length of a leg when pointing on it ( instead of opening the properties window to see these informations )
2) Route and WPs activation
When activate a route , the first route's WP is always activated without consideration of the boat position . I suggest the activation of a route activate always the next point AFTER the boat in the route (as it is the case in my Garmin)
3) route/WP management
Add the possibility to modify all WPs directly in the route properties window
4) The functions "delete all WP" and "delete all routes" should be accessed from the general menu ("Max details here ..............Object query") and not from the WP and routes menus
5)Could it be possible to have actual mean speed and times in the tracks properties window instead of the "planned speed" and times calculated with it which is relevant for a planned route but not for tracks .

Many thanks

Jean Pierre

[LeaseOnLife]

Dave, Jean Pierre

I am glad Jean Pierre mentioned it, I thought it was me, fat-fingering..

F10 & F11 do have a problem.

Here is what I can make reproducible, somewhat:

• F10 is being ignored every second time.
• When using F11 after an odd number of F10, it is being ignored, any amounts of F11s don't do anything. hitting F10 one more time or clicking the mouse on the chart area seems to make F11 work.
• When using F11 after an even number of F10s, it works.
• Basically, when using F11 after F10 has been ignored, then F11 works.
• Once F11 works, it works every time until one hits F10.
• The problem does not occur when using the on screen buttons

Weird, but hopefully this helps

Dirk

PS: I am just getting started with routing on opencpn. Jean Pierre's suggestion get a +1 from me, every single one!

[GPS-Marco]

Well, second post about datum conversion.

Maybe it's simpler that I thought. The KAP file header has a row that specifies how many seconds of latitude/longitude the chart has to be shifted to convert the chart map datum into WGS84. The row is identified with the code DTM/ (Datum Shift Record) followed by 2 comma separated floats specifying the lat/lon seconds towards North/East the chart has to be shifted (negative values for south/west shifts).

Refer to http://www.iho.shom.fr/publicat/free...ns_for_RNC.pdf for (little) more info.

Example: this is an extract of the .KAP file of a chart that needs to be shifted 0.06 minutes South and 0.07 minutes West (those shift values are printed on the paper chart):

....
PLY/2,42.9825886997,9.3332371953
PLY/3,42.9825757964,9.3534064474
PLY/4,42.9582750305,9.3534199359
DTM/-3.6,-4.2
CPH/0
WPX/1,-5.0727717924e+005,5.4042614513e+004,7.1159982812e+ 001
WPY/1,3.1786827369e+006,2.2571069706e+000,-7.3948263675e+004
....

I hope it will be easy for you to implement such shifts in OCPN. It would be great for non-WGS84/NAD83 charts.

Reards, Marco.

[bdbcat]

GPS-Marco....

Thanks for the info on the DTM record. Didn't know that.

Opencpn implements datum shift for non-WGS84 charts already. We use the datum specified in the KNP field, such as this:

KNP/SC=47600,GD=European 1950
PR=MERCATOR,PP=0.000000,PI=UNKNOWN
SP=UNKNOWN,SK=0,UN=METERS,SD=LOWEST LOW WATER,DX=4.76,DY=4.76

We calculate offsets based upon a transform depending on the "GD=" string value. This transform maps, for example, from European 1950 datum to WGS84 in a very precise way.

I tested this for two Euro 1950 charts which I have. The values calculated by opencpn for the shift offsets in seconds almost exactly matches the value specified by the DTM record in the header.

Also, some chart features like lighthouses match the corresponding CM93 chart for that region within navigational accuracy.

So I think this works correctly now.

Do you have a specific chart/location with a problem? Perhaps you could post the BSB header, including the KNP and DTM records for study.


Thanks again
Dave

[GPS-Marco]

Originally Posted by bdbcat

GPS-Marco....

Thanks for the info on the DTM record. Didn't know that.

Opencpn implements datum shift for non-WGS84 charts already. We use the datum specified in the KNP field, such as this:

KNP/SC=47600,GD=European 1950
PR=MERCATOR,PP=0.000000,PI=UNKNOWN
SP=UNKNOWN,SK=0,UN=METERS,SD=LOWEST LOW WATER,DX=4.76,DY=4.76

We calculate offsets based upon a transform depending on the "GD=" string value. This transform maps, for example, from European 1950 datum to WGS84 in a very precise way.

I tested this for two Euro 1950 charts which I have. The values calculated by opencpn for the shift offsets in seconds almost exactly matches the value specified by the DTM record in the header.

Also, some chart features like lighthouses match the corresponding CM93 chart for that region within navigational accuracy.

So I think this works correctly now.

Do you have a specific chart/location with a problem? Perhaps you could post the BSB header, including the KNP and DTM records for study.


Thanks again
Dave

Hi Dave, then the issue is really simple to solve. I think to remember the map header specifies GD=European Datum 1950 and not GD=European 1950. So the solution should be just to add a new string to check in the KAP header parser to recognize this GD string as ED50.

Moreover it would be great, if the parser does not recognize the map datum string, to use the DTM data record, if present: just in case some "strange" string is used to describe the datum

I'll be more precise on the exact GD string on the maps I have later on when I'll be back at home.

Let me thank you for you effort.

Ciao, Marco.

[GPS-Marco]

Quick update about map datum strings in BSB charts. Theese are the GD strings I've found (case sensitive):

GD=EUROPEAN DATUM 1950
GD=ROMA DATUM 1940
GD=ROMA 1940
GD=HERMANSKOGEL DATUM
GD=WGS84
GD=LOCAL DATUM UNKNOWN
GD=Local Datum

Almost all charts have the DTM field set (very useful for charts with GD=local/unknown ...)

Hope this help.

Ciao, Marco.

[idpnd]

Having finally had a chance to upgrade to the latest CVS, I must say I do concur wiht above remarks, namely:

- The current ship icon looks far too big in all but the most magnified resolutions, and should be reverted to the red circles/cross-hair when zoomed out

- The icon does not seem to render very well, and looks far less attractive than the icons on the menu bar (I believe there is already a corrected version out?)

[LeaseOnLife]

Happens every time after a opencpn restart:

AIS target query window shows up partially off the screen. When moved, opencpn remembers the position for all next AIS target queries, until next program start, then the first AIS target query window is back to the initial off screen position

System: XP
Screen: 1024x768

[dacust]

Another quirk.

After I changed the path for the SENC directory to G:\\charts\\SENC, the window size shrunk for S57 create and load.

The window is very narrow and chops off a lot of the information. Looks like the window width is based on the length of the path.

Very, very minor bug.

-dan

[manimaul]

Originally Posted by idpnd

Having finally had a chance to upgrade to the latest CVS, I must say I do concur wiht above remarks, namely:

- The current ship icon looks far too big in all but the most magnified resolutions, and should be reverted to the red circles/cross-hair when zoomed out

- The icon does not seem to render very well, and looks far less attractive than the icons on the menu bar (I believe there is already a corrected version out?)

The corrections should be in the next cvs upload. See Post #1250

... the original looked much worse in gtk then msw.

Will

[chris blair]

I'm using XP and the stretched icon seems to happen when I zoom in with raster and vector charts. Also, my $.02 is that the icon is way too big. Something 1/4 the size would be adequate for me.

Thanks.

Regards
Chris
SV Toucan Dream

[LeaseOnLife]

Originally Posted by LeaseOnLife

Happens every time after a opencpn restart:

AIS target query window shows up partially off the screen. When moved, opencpn remembers the position for all next AIS target queries, until next program start, then the first AIS target query window is back to the initial off screen position

System: XP
Screen: 1024x768

Adding more findings:
opencpn only remembers the position of the window after pressing OK. If one closes the window with the close button <x> or with Alt-F4, opencpn seems to loose the position and the next time it opens the window again at the default-partially-offscreen position.

[sinbad7]

Manimaul..

Your icons are great and a far cry from the original 'tank ship', BUT, why not make it the same size as the AIS icon? That seems to be just about right in most scales. See attached screendump of AIS ship icons. And just give it a 'plus' sign to indicate it's geo pos. I would further suggest that the yellow and red start point of the routes use the same boat icon. There is no such thing as an automatically adjustable boat icon according to chart scale I suppose?