Charts II: NGA - 2700 Charts

[P_Dub]

Could you give me the exact ftp site for your charts?
Bye
P_Dub

[sinbad7]

The usual: NGA Charts Status Region Miscellaneous at OpenCPN.info (English)

Tore

[recri]

Originally Posted by bcn

Thomas, Pavel,

- the question of the hierarchy of chart types is still open.
If there is a 90 deg skewed chart which insets or corners out of range, how shall we categorize that chart: 90 deg skewed?

- rectangular charts with insets "on the water" are treated in non uniform way:
as "doable with insets" and as "special PLYs required"
Your idea?

Hubert

Tangent to your question, but related, here are some wrongly categorized charts I just fixed:

14202 was listed as requiring a special PLY, and that's true, but its real problem is that it is non-rectangular. The NW and NE corners aren't at the same latitude, the corner ref points of the chart do not form a rectangle.

37162 was listed as non-rectangular, because it has a small bump on the east edge. But the chart body, excluding the bump, is perfectly rectangular, the REF points came out perfectly. In this case, it should have been listed as requiring a special PLY to accommodate the bump.

53180 was also listed as non-rectangular, but it is another rectangular chart with a couple of bumps on the edges. It requires a special PLY, eventually, but its corner coordinates, REF points, and everything else can be entered now.

It should be noted that "Rectangular, not doable in Phase 1 because of special PLYs needed" charts will get converted to KAP's if you enter the necessary data. So you can check the result in OpenCPN even though the chart will still need some additional polishing to be finished.

[bcn]

Recri,

the case of the 53180 would result in chart where the "extrusions" are cut off, I suppose.
Usable chart but not "what the creator of the chart pretended"

What is our policy for a case like that?

Hubert

[recri]

I think we should rename "Rectangular, not doable in Phase 1 because of special PLYs needed" to "Rectangular, partly doable in Phase 1 - special PLYs needed". A lot of these charts are perfect except that they need the edge bumps PLY'ed into the quilting or chart junk masked out on the water. But people aren't entering the data because they interpret the category name to mean the data won't be used.

On the other hand, I think we need "Rectangular, partly doable in Phase 1 - some or all corner coordinates not supplied" as a new category because we should give estimated corner coordinates special attention. These are turning up in ocean island charts.

[recri]

Originally Posted by bcn

Recri,

the case of the 53180 would result in chart where the "extrusions" are cut off, I suppose.
Usable chart but not "what the creator of the chart pretended"

What is our policy for a case like that?

Hubert

The whole chart, extrusions included, is visible in opencpn if you turn off quilting. The quilt neighbors may cover the extrusions, too, if they are important enough. And we intend to implement the PLY's at some point.

So it is the same, I think, as the rest of our "partly doable" categories -- the chart is partly done, use what you can of it for the moment, expect the rest to become more useful as our processing improves. No?

[HappySeagull]

I don't want 'em to get lost in the pile(the edge-bumps)...so I've been marking them "special plys needed" but it's ok to calibrate em for sure...they could easily go into a new "partly doable -special plys" needed category if nohal and netsurfer think it's worth the bother to mess with ...
anybody that wants these right away could anyways of course pop-in,do them as doable and get their kap,but again,I don't like the idea of that.There's no big hurry,is there?

[bcn]

Happy,

that was the question:
how do we want to handle this - what is the idea, the way to go?

Done the NGAs - some day/to some degree - there might be a bunch of other charts.
And we should have the principles/categories right

BTW I was a bit surprised that the whole bitmap chart will show up when quilting disabled - case 53180 - learning I suppose...


Hubert

[recri]

A new status report. I'm looking at more categories of charts, refined the missing coordinate test, and changed the order of the tests so that projection issues will only turn up if a full set of coordinates is available.
-----
processed: 801 charts
accepted: 773 charts, mean ratio 1.00481 stddev 0.00191

problem: chart 26224 ratio 0.00020 - 1d error in corner coordinate
problem: chart 27211 ratio 0.67606 - 10' error in corner coordinate
problem: chart 54418 ratio -3.74932 - 2d error in corner coordinate
problem: chart 62250 ratio 1.38783 - 1d error in corner coordinate
problem: chart 62271 ratio 0.76884 - 17' error in corner coordinate
problem: chart 81511 ratio 1.01297 - looks right
problem: chart 83495 ratio 1.02724 - looks right

missingcorner: 37234 37238 37244 37246 37264 52200 61611 81251
----
54418 is marked as checked and locked as of 3-Sep, but the N edge is south of the S edge.

83495 looks right, but there aren't any corner coordinates printed on the map and the ones entered all end with 0" or 30", so they may be rough estimates.

The missing corner charts lack corner coordinates. Most of them just have a category chosen and no other data entered. It would be good to check if the REF points can be entered before you decide a chart is doable.

[wdkester]

Originally Posted by bcn

For the manual try:

ftp://ftp.rr-consult.com.es

user: ftpuser
Pw: ftpuser

The new manual is a great help!

I'm guilty of having labled doz's of charts International for what I thot were good reasons, what with they're being non US charts issued by a US org.. I began to realize that almost all NGA charts are nonUS. Is this a useless category? Shouldn't we just stick with the General, Coastal, harbor tags? Let's decide before I go back through the lot.

I would like to hear more about the "WGS datum without the year issue". I been calling these "other" with a note "WGS no suffix". My reasoning is that if they were held to 72 or 84 standards they would have said so. Am I wrong? I don't think the manual is clear on this.

[recri]

Originally Posted by wdkester

I would like to hear more about the "WGS datum without the year issue". I been calling these "other" with a note "WGS no suffix". My reasoning is that if they were held to 72 or 84 standards they would have said so. Am I wrong? I don't think the manual is clear on this.

Yes, I've been wondering about that, too. There are several "other" variants in the database labeling those charts. But most of them seem to be labeled "WGS84" in the data base, as if that was an approved synonym even though the charts look to predate "WGS72".

It seems that the right way to label them would be a new menu drop down that says "just WGS".

[HappySeagull]

Hey,recri-that works really great!I haven't even had time to download_and- check and you're "on my case" -why bother,eh?it's much easier to just wait for you!

[dacust]

Originally Posted by bcn

An updated version of the manual is available:

ftp://ftp.rr-consult.com.es

user: ftpuser
Pw: ftpuser

Thanks everybody for the suggestions!

New manual now at:

==> NGA Chart Calibration


-dan

[P_Dub]

What is the full ftp site address for the finish product .kap file charts?
Regards
P_Dub

[HappySeagull]

WGS72--->WGS84
X=0 Y=0 Z=4.5meters I'm not sure I believe Z as overall.
but I guess the bumpiness of the model has changed.OpenCPN is going to open every chart as wgs84 anyways.

But google up something better if you like.