Charts II: NGA - 2700 Charts

[recri]

Originally Posted by phiggins

Chart 93022 should have a DTM/46.140,5.940 . Even though the chart doesn't indicate this its off by this amount. A lot of the charts in that area are off quite a bit and need to have DTMs.

Are you saying that we should compute a datum correction by inspection and enter it as if it were printed on the chart?

[phiggins]

Are you saying that we should compute a datum correction by inspection and enter it as if it were printed on the chart?

Yes, if we want accurate charts. I am using my latest version of GE2KAP to check the charts mentioned and it indicates they are off according to Google earth. A lot of charts in that region are off. Another user of GE2KAP has recreated many charts for the Phillipines and says the adjustments are accurate (Valhalla any comment?).

[bcn]

The idea was to _reproduce_ the charts as they were intended.

Not to make interpretations or improvements - in this phase of the project.
If a chart is not usable actually we should mark it as such.

And later on we might decide to tweak it.
That would require a common approach how to make any type of "repairs".

Precision of Google Earth might be another point to discuss in that context perhaps.

Hubert

[cagney]

Originally Posted by z.stadler

I'm getting this error:

'Corrects Chart Datum to' is set to 'other' - please enter the datum the chart is corrected to (required)."

although I've entered the "other Datum".

Chart 93047 has this note:

TO THE NAVIGATOR
Operational Datum Adjustments
To position Luzon coordinates on this chart:
Subtract 04.06 seconds from Latitude.
Add 04.45 seconds to Longitude.

Which I've tried to capture like this:

A similar note exists on chart 93044.

Am I doing something wrong or is there an issue with the validation checks on the server?

P.S., in the mean time, I've saved it with a "No DATUM NOTE given on chart"

The charts are WGS72, which is fine for our purpose. You are trying to shift them to a Luzon date, which is a no no.
We only bother if there are corrections given to plot WGS coordinates on an "other Datum" chart.
Apart from that.....The fields for the datum shift are in minutes but you have entered seconds.

Thomas

[HappySeagull]

Ze'ev...you don't want to correct the chart TO Luzon datum....It's for people transferring fixes from a Luzon Datum chart to this one,which is not...similar has been seen before up - thread....Found it!
http://www.cruisersforum.com/forums/...tml#post768658

[recri]

Another corner for the gallery:

[z.stadler]

Originally Posted by HappySeagull

Ze'ev...you don't want to correct the chart TO Luzon datum....It's for people transferring fixes from a Luzon Datum chart to this one,which is not...similar has been seen before up - thread....Found it!
http://www.cruisersforum.com/forums/...tml#post768658

At this stage, I believe, our goal is to capture information from the chart "as-is". Corrections to "other" datums could be ignored by KAP creation process.

Currently there is no way to enter an "other" datum for the correction and there is an inconsistency between the error message and what I've entered.

I'm hoping to get a response from the people who have coded the validation rules on the server.

Ze'ev

[HappySeagull]

Ze'ev..yes,you are right.I tried to get a chart to accept "other" and it won't.Only allows correction to wgs72 or wgs84...maybe Nohal and Netsurfer just pasted the same dialogue code from above ... but there's no point in us digitally calibrating charts from wgs to a different datum is there?Not yet anyways...and the charts that are wrong(as PHiggins pointed out) should stay that way here given that all possible info from the chart has been incorporated towards being gps compatible...I'm sure that others have found charts they know to be "off"
There could/should probably be another directory for people to correct these charts and offer revisions....maybe that would be a dialogue to add..."Save to NGARevisions"-it could be an option that remains open after the chart has been viewed and locked.

[recri]

NGA 2700 chart summary for Sun Oct 02 17:19:25 GMT 2011
processed: 1896 charts
evaluated: 1887 charts
pxhoriz: mean 1.00026 stddev 0.00207
>10*sd: 43081 (1.02339) 43083 (1.02302) 43062 (1.02300) 43101* (1.02285) 43058 (1.02283) 43082 (1.02214) 43063 (1.02201) 43064* (1.02096)
>7*sd: 43041 (1.01666) 43084 (1.01616) 43122 (1.01615) 43123 (1.01572) 43121 (1.01553) 43141 (1.01514) 43142 (1.01501) 43265 (1.01482)
>6*sd: 43263* (1.01462) 43161* (1.01381) 43202 (1.01313) 43126 (1.01307) 43371 (1.01304) 43204 (1.01293)
>5*sd: 43225 43264 43343* 43344*
pxverti: mean 1.00001 stddev 0.00057
>25*sd: 43143 (1.01974)
pxsquare: mean 1.00001 stddev 0.00048
>25*sd: 43143 (1.01704)
ratio: mean 1.01685 stddev 0.48049
>25*sd: 43264 (21.77725)
>4*sd: 93030
notmercator: skipped 9 charts: 17421 18419 43186 43249 43252 43262 43285 56191 76181
untouched: 93110 93301 93320 93340 93342 93404 93405 93460 93469 93478 93500 93540 93580 93626 93629 93647 93650 93652 94033 94082 94120 94123 94160 94164 94165 94180 94187 94201 94220 94260 94280 94290 94322 94360 94420 94423 94440 96028 96763 96764 96780 96800 96820 96882 96883 96901 96902 96904 96906 96910 96939 96941 96943 96945 96947 96948 96949 96960 96962

[wdkester]

I would like to hear from those who are concerned about Google Earth accuracy, especially any areas that are mis-plotted by GE. When I was debating (with myself) about using GE,I did find that they sometimes stiche the sat photos together badly. Someone complains and asks them to correct it. Sometimes they do. I wouldn't be surprised to find that there were problems in the nether regions.

I have used Google Earth in East coast Mexico, Belize, and the Rio Dulce, especially for planning. Puerto Stuck, Belize is a good example. There is no large scale chart, the Rausher WP are somewhat west of the channel visable in GE. Using GE I found better water. Luck? Maybe.

I just used paul's GE2KAP to inspect "Harbor" scale charts in the Caribbean. Out of 50 some charts I found something to note on 20. The differences include: mangrove growth, shoreline differences, points of land that have grown, new features like bridges and piers and just plain bad datums like Puerto Morales.

[recri]

Chart 94220 is giving me fits, so I finally unpacked the KAP, downloaded the JPG, and discovered that the web interface is measuring the south edge of the chart 400 pixels too far north.

The y coordinate listed for S when you position the corners in the web interface is 12448. The actual position of the south edge measured on the JPG is 12848.

The chart is perfectly rectangular and oriented straight up, so there are only 3 other pixel coordinates for E, W, and N. I get the same values for them whether measured on the web interface or directly on the JPG.

[HappySeagull]

Originally Posted by recri

Chart 94220 is giving me fits, so I finally unpacked the KAP, downloaded the JPG, and discovered that the web interface is measuring the south edge of the chart 400 pixels too far north.

The y coordinate listed for S when you position the corners in the web interface is 12448. The actual position of the south edge measured on the JPG is 12848.

etc...

recri,go try it now.I used your pixel count and the kap looks pretty good in O.I don't know how it'll wash in the end,but there it is...

Gee!we've seen em all ! 3 cheers!

[recri]

Originally Posted by HappySeagull

recri,go try it now.I used your pixel count and the kap looks pretty good in O.I don't know how it'll wash in the end,but there it is...

Gee!we've seen em all ! 3 cheers!

Happy --

Sure enough, looks fine. Thanks! I thought those text fields were protected against direct text entry.

We've seen em all, but I have a few still need to be checked.

[HappySeagull]

recri,no,no.I just moved the crosses down"y"ise to where you said,pixel-wise...the little arrow control tabs(bad description!) were necessary for this -no way my hand is that steady!...I haven't used em often but they are handy when needed!

...I think the problem is just some glitch that's only in the calibrate-corner-display,not the jpeg, or my tinker couldn't have worked... we should call it doable-rectangular before it gets messed with...what do you think?

[recri]

Another summary for anyone looking for something to double-check.

These are the tails of the distributions. Charts marked with a * have already been rechecked and judged "okay".

I should probably pull the Norwegian 43xxx charts out so we can see the rest of the distribution tail.

NGA 2700 chart summary for Tue Oct 04 00:44:19 GMT 2011
processed: 1951 charts
evaluated: 1942 charts
pxhoriz: mean 1.00026 stddev 0.00207
>11*sd: (0.02274) 43081 43083 43062
>10*sd: (0.02067) 43101* 43058 43082 43063 43064*
>7*sd: (0.01447) 43041 43084 43122 43143 43123 43121 43141 43142 43265
>6*sd: (0.01240) 43263* 43161* 43202 43126 43371 43204
>5*sd: (0.01034) 43225 43264 43343* 43344*
>4*sd: (0.00827) 43287
>3*sd: (0.00620) 43079 54368 54362
pxverti: mean 1.00000 stddev 0.00035
>15*sd: (0.00522) 92356
>12*sd: (0.00417) 37403
>9*sd: (0.00313) 81563*
>7*sd: (0.00243) 83052* 38502
>6*sd: (0.00209) 22404* 73032 526*
>5*sd: (0.00174) 22542* 43343* 73271 72150* 24511* 91314* 76201* 96036*
>4*sd: (0.00139) 38583 73470 623* 24152* 22549*
>3*sd: (0.00104) 43060 81303* 91407* 83218* 81576* 73034* 24154* 71012* 29325* 43064* 91326* 521* 83039* 91030* 24159* 81729* 23051* 82025* 61003*
pxsquare: mean 1.00000 stddev 0.00027
>15*sd: (0.00400) 92356
>11*sd: (0.00293) 37403
>10*sd: (0.00266) 73291
>5*sd: (0.00133) 92370 92140* 83218*
>4*sd: (0.00107) 43204 43062 14221 38502 73653 92025* 82384* 91326* 707*
>3*sd: (0.00080) 83555* 91096* 81565* 71140* 61003* 83392* 21161* 91314* 92260 81576*
ratio: mean 1.00471 stddev 0.00277
>25*sd: (0.06935) 26125*
>3*sd: (0.00832) 71400*
notmercator: skipped 9 charts: 17421 18419 43186 43249 43252 43262 43285 56191 76181