Charts II: NGA - 2700 Charts

[HappySeagull]

Originally Posted by wdkester

Happy,
I'm confused by "GE can't get the Greenwich meridian right (it's .249 min west)" Couldyou say a little more?

...oh,I was looking up the accuracy of GE with an eye as to quick-check our handiwork....
The above ".249" is just an illustrative gem in all that...if you go to Greenwich in GE,the observatory is about that westward of GE's 0 longitude.There's all kinds of reasons touted for that, but I think they'd all be news to the painstaking folks that drew our charts.

[wdkester]

From the fount of all knowledge "Wikipedia-
"The shift from several spheroids to one worldwide spheroid caused all geographical coordinates to shift by many metres, sometimes as much as several hundred metres. The Prime Meridian of these modern reference systems is 102.5 metres east of the Greenwich astronomical meridian represented by the stainless steel strip. Thus the strip is now 5.31 arcseconds West. The modern location of the Airy Transit is 51°28′40.1247″N 0°0′5.3101″W / 51.477812417°N 0.001475028°W / 51.477812417; -0.001475028 (Airy Transit)"

This seems to be in the car park to the north of the observatoy.

[HappySeagull]

I take away from that that they couldn't get Greenwich right,so there's a "modern" location....but GE will do to spot a 7 mile error for sure.

[r.fairman]

On my version of GE the prime meridian is correctly marked 104 meters to the east of the old meridian strip If you look carefully it is now marked by a line of Sea Gull droppings

[HappySeagull]

..anyways,I 'm toying with the idea to adjust the DTM in the header for a "GE-agree" version of some of these NGA charts (just for personal interest) but I'm wondering if it's a reasonable pursuit.
a DTM would be a compact and fast way of communicating chart shifts around...and it'd be easy to revert the chart too.

[wdkester]

Happy,
I think you are talking about editing the header of a downloaded NGA/KAP with a hex editor to create a personal copy fo the chart. Great. Tell the rest of us how to do it.

I realize that I may be a single voice; but I'm against altering the header data using the "adjustment" commands at "opencpn.info/en/nga-charts-upload-files" as this alters the archive copy. I would like to see any suggested datum changes so that I might apply them at my option. I feel that any two people are likely to disagree on the adjustments depending on how they arrive at those adjustments: e.g. using GE, a single WP, multiple WP a single track, another chat etc.

I think we need a repositoryfor information like this.

[HappySeagull]

Hi,WD...yes Tiabui and recri and well, we we all are thinking of a revisions repository..but there's still Plans to do...

To edit the header,just open A COPY of the kap in "Notepad++"(portable is best for me) or anything that's half-decent....Not Windows' notepad- it's a little difficult-It adds characters to the beginning... and ,well,don't bother is my advice.

Then,save.Then drag-and-drop onto bsbfix.exe.It's a utility from the libbsb folder.PM me if you can't find it online.Opensource....if you don't have this,you should.bsbfix can be anywhere on your computer.Dragging the file onto a desktop shortcut to it should work too.
If bsbfix spews a ream of gobbledeegook.You failed.Try again.you just need to pay attention to how things are done in there.Don't add spaces unless it's in the name ...stuff like that.

[cagney]

@wdkester

I agree with your views.

To change the DTM is not difficult.
Let's take 83495 Wallis Island as an example. It is way off compared to CM93 which is said to be WGS 84. I applied

DTM/-5.0,-131.0

To bring it into reasonable agreement.

As you said open a hex editor (work with some text editors as well), edit the DTM line. Save. Then, as a general recommendation, run "bsbfix" from libbsb on the changed file.
Sometimes you can get away with changes without doing this, and some recommend aginst using "bsbfix".

Then go to the chart tab and run "Scan charts, and update database" in 2.6 or "Force full Database rebuild" in earlier versions.

The correction should now be applied.

I have a longstanding feature request aimed at making this process easier. Check FS#309 .

Thomas

[Viking Sailor]

Happy,

Be careful with using GE as a reference. I have seen a couple of spots where two over lapping photos did not align correctly. I wish I could remember where these spots are so that you could see them for yourself but I'm sorry I can't.

[HappySeagull]

HxD is a freebie hex editor...It's a little tricky,to edit with it,but you do get the feeling you are doing serious things in there!...Surveyor's recommend of Notepad++ has been good though.It's a lot less "crucial-feeling".
Thomas is right,you can just try without bsbfix

For windozers,A good idea is to just quickly check your new kap (or any kap)in Thomas' "Kapview" which won't often load it if the kap is no good...This can save firing up Opencpn and rebuilding database and all that.It also spits out a handy tif!
I finally made a dragndrop script for irfanview or xnview if anbody wants to dragndrop a kap to View ,edit and rebuild with mapcal +imgkap....

[HappySeagull]

Originally Posted by Viking Sailor

Happy,

Be careful with using GE as a reference. I have seen a couple of spots where two over lapping photos did not align correctly. I wish I could remember where these spots are so that you could see them for yourself but I'm sorry I can't.

Thanks Viking..yes,I think you're right.I'd end up with 2 bad charts!Trusting neither.No substitute for somebody there with local knowledge.But it's still something to keep in mind.People upthread have said the oldest of these charts are just plain bad in the Philippines ,and other places.Checking them out in GE is at least a start to see if they are within seeing distance!

[wdkester]

Thomas,

Just a little more help and I'll have it
1. What are the units for DTM? deg or min or sec or ?
2. Which is entered first N/S or E/W?
3. Which way do positive DTM values move the chart?

Thanks

[wdkester]

Viking,
The only GE problem that I've seen is a beaut. Look at Eluthera in the Bahamas about the N/S mid-point. It may have been reported and fixed by now tho.

[cagney]

Originally Posted by wdkester

Thomas,

Just a little more help and I'll have it
1. What are the units for DTM? deg or min or sec or ?
2. Which is entered first N/S or E/W?
3. Which way do positive DTM values move the chart?

Thanks

David

1 Seconds of arc.
2 First lat then long corrections.
3 N and E are positive and moves the chart in just those directions. So in the example I used, the chart is moved 5" to the South and 131" to the west.

Thomas

[manimaul]

Wow, I guess I've been out of the loop for a while. I didn't know this was going on. You guys have been busy .I created a similar scraper a while back to pull chart images from NGA: ngaRasterScraper.py

Not sure if it helps... but feel free to use this script. Note: requires python (obviously) and imagemagick

Grabs jpg files and pieces them together into a png chart image. By default it grabs zoomify level 6. Edit zoom = "6" to grab whatever zoom you want.

Also, this script will grab every available chart from NGA: so it may take a few days to run .

PS ~ Is there a central repository for user calibrated KAP header files?

Will