Beta Test / Technical

[Ptizef]

Originally Posted by idpnd

I can't get it to zoom all the way in on certain chartsets/regions, I believe not all chartsets are complete?

It seams that there are problems in some areas with the new CM93ed2_2009 from CNF.SeaSoft
An example : around the position 47°34N 02°55W with my old CM93 dated 1997/1998 , I could zoom to scales 1/50000 (E) , 1/20000 (F) and 1/7500 (G)
With the new version I just loaded , I have only scale 1/50000
After time spent trying to understand , I replace the file 03901020/F/04121071.F by the one from my old version and I added the file 03901020/G/041210071.G which was missing . Now I have again the three scales ( off course the last two ones dated 1997)
That means that in this version there are files missing and others unreadable . It would be interesting if someone could be confirm by reading it with another soft (Maxsea ?)
I take the opportunity to suggest a modification the see in the maps area of opencpn the complete name of the maps .
instead of "CM93Chart F 1/20000" I would prefer
"CM93Chart 03901020/F/04121072.F 1/20000"

Also I can confirm that "max details here " never worked with CM93 nor "scale in scale out"

Jean Pierre

[bdbcat]

Manimaul...

The GRIB icon cpp file you posted will not compile. It complains about embedded URL. I see "inscape.org" in the .cpp and in the .png.

Don't know inkscape and the tools well enough to fix it. Will you have a look at it?

Thanks
Dave

[manimaul]

hmmm, i'm on it. in case you are interested, i export as png 32x32 from inkscape, run png through optipng, then throug png2wx.pl perl scipt.

optipng
png2wx.pl

i'll see if i can fix.

will

[manimaul]

for the grib png icon, i processed in windows. whereas the previous one were processed in linux... i'm not sure why there would be a difference. any ideas? i will process png on my linux box and repost.

[manimaul]

Try this one : grib.zip

if this compiles...will it be available in cvs for testing?

[Psyches]

Originally Posted by manimaul

Try this one : grib.zip

if this compiles...will it be available in cvs for testing?

Guys, the .cpp has embedded quotes in the string, that should be octal \042.

Here is one that compiles for me. (sorry, I seem to be forum-challenged and could directly upload a .zip, please rename)

[Psyches]

...not...could not upload a .zip. apparently I can't put a msg together either; manimaul, if you have hints, I'm eager to learn. [nevermind, i see you ref via url directly to matrixmariner.com]

re: the filename...sorry, I'm trying to get a grip...er, a grib.

[bdbcat]

Psyches....

Got it, compiles fine.

Thanks
Dave

[sinbad7]

Dave..

Updating delay of AIS targets in OpenCPN..

I have mentioned this problem in the early days,and it still exists. Seems like OpenCPN does not read or receive the second message received containing ships name,size and the rest of the static info.

I tested it out again today from the White Rose oil field off Newfoundland. As you see,the WebVTS browser displays all info immediately,whileOpenCPN,using the same feed remains in yellow state for a long time,sometimes never changing to green at all.

Would appreciate if you could check it out when you have a spare moment.

[Zoef]

Originally Posted by Ptizef

It seams that there are problems in some areas with the new CM93ed2_2009 from CNF.SeaSoft
An example : around the position 47°34N 02°55W with my old CM93 dated 1997/1998 , I could zoom to scales 1/50000 (E) , 1/20000 (F) and 1/7500 (G)

Jean-Pierre,

I got my 2009 maps from a Norwegian source, my old maps date from 2002/2003.
I can confirm your observations concerning opening/availability of detail maps.
When I tried both mapsets with CMAPECS I found the same as with OpenCPN. So, the problem is in the cm93 mapset of 2009. C'est dommage.

Klaas

[Nissenudel]

Also noticed some Updatingdelay of AIS targets in OpenCPN !

I also find that three of my parallell monitored WebVTS browsers displays all info immediately, whileOpenCPN with the same feed, for some targets, remains in yellow state for a long time, sometimes never changing to green at all.

Nisse/Sweden

[bdbcat]

Sinbad7 and others....

In opencpn, the target will be yellow until the MessageID(5) message containing the name/dimensions/etc. is received.

I believe that WebVTS uses a back-end cache to match MMSI numbers to ship data and name.

If you recall, we looked at the TCP stream carefully and sometimes found that for some targets, the MID(5) message never occurred in the data stream, and so the ship particulars shown on WebVTS must have come from a cache built into WebVTS.

We plan no equivalent global MMSI<->Ship Data caching function for opencpn. If opencpn can be used as a global AIS browser as designed, it is incidental.
Source and accuracy of data from a TCP/IP AIS data feed is unmeasurable, and its use in opencpn is not recommended.

Opencpn accurately monitors live AIS data from a physical radio receiver over useful navigational distances.

Thanks
Dave

[sinbad7]

Dave...

We've had this dialogue before and Siitech assured me they are not using any cache for their received AIS data,it is always in real-time.

When using the same direct feed with Fugawi ENC or SeaClear,all the ship targets display with full data after a very short period.

You may not be interested in the support of a global AIS monitoring but it should provide you with a substantial testing base. There are also those who appreciate a 'sneak peak' at vessel traffic outside their own range.

I have another query regarding the way OpenCPN displays c-map data.
Using the same c-map database with for example MaxSea,a very different display is presented. What could be the reason for this?

[cagney]

This site http://www.siitech.com/prod_as.htm contradicts that statement about Siitechs AIS service.

AIS Server

SiiTech AIS Server is a software solution that collects AIS and ARPA (Automatic Radar Plotting Aid) messages, redistributes them, and decodes and stores AIS and ARPA data in the database.

Thomas

[bdbcat]

Sinbad7:

Regarding the difference between opencpn and maxsea cm93 chart displays:

1. cm93 is equivalent to S57 ENC in the sense that it is simply a database of navigationally important objects. There is no information in cm93 nor S57 determining the actual chart display graphics. A "Presentation Library" is required in order to display a chart on the screen. Opencpn uses a version of the IHO S52 standard for chart display. Maxsea uses an internal home-grown Presentation Library, probably because when it was introduced the S52 standard was not well defined. So, the on-screen presentation is sure to differ in some details.

2. Are there useful features on the Maxsea display that are NOT being rendered by opencpn? If so, please point them out and I will track them down. We know that there are some objects/features/attributes in cm93 that are not quite S57 compliant, and these must be considered as special cases in opencpn. There could easily be some of these cases missed in the current code. I plan to fix these cases as they arise from user feedback.

Thanks
Dave