Subscribe to this thread
Home - General / All posts - manifold 8 _ version 8.0.29
yves61
438 post(s)
#18-Oct-18 23:06

Why do I need a false easting and false northing of approx 43 m to get Virtual Earth satellite imageserver correspond with the drawings in my map ?

The projection settings of the drawings ought to be right.

Can someone guide me to help understand ?

Attachments:
VE_FE_FN_43m.map

tjhb
10,094 post(s)
#19-Oct-18 00:04

It's difficult to say without the image.

The [Virtual Earth Satellite Image] in the project is linked externally, so not available in the downloaded project.

But why do your drawings [Adp40000 Drawing] and [Gbg40000 Drawing] have false easting and northing of 150000.01256 and 5400088.4378000004? Can those parameters be right?

Their other projection parameters also look strange (everything except Center Latitude).

That is not a standard Manifold 8 projection. What standard are you using?

yves61
438 post(s)
#19-Oct-18 07:43

The Drawings have the very common projection for Belgium.

Above projection settings are almost equal to the National Grid Belge 1972 for a few figures at position 1/1000 and 1/10.000 after the decimal point.

Attached also the original .prj files that come with the shapefiles

Also attached a screen print with the projection for Virtual earth, with local offsets to accomodate for the restricted window.

It seems I need to use the same false eastings and northings for Virtual Earth if I use the global settings

Attachments:
Adp40000.prj
Gbg40000.prj
VirtualEarth_projection.PNG

tjhb
10,094 post(s)
#19-Oct-18 07:57

I don't know this projection, but it looks as if the value for the center longitude (central meridian) is significantly wrong, and the value for the second standard parallel is slightly wrong, in those .prj files.

Never trust a .prj file--even if you wrote it yourself!

Instead, assign the correct standard projection after importing the drawings.

I am looking at https://epsg.io/31300.

yves61
438 post(s)
#19-Oct-18 08:28

I meant ...

Above projection settings for the drawings are almost equal to the National Grid Belge 1972 for a few figures at position 1/1000 and 1/10.000 after the decimal point

... for the false easting and false northing settings.

The drawings come from trusted source : the regional authorities in Flanders (Belgium) (AGIV) .

I thought it to be epsg:31370.

but will check with your suggestion epsg:31300

Could someone check if the same correction (43 m) for false easting and false northing are needed with other GIS software like with QGIS to exclude a possible bug in Manifold 8 ?

yves61
438 post(s)
#19-Oct-18 08:51

Allright, seems the .prj files that come with the original drawings are a mix of epsg 31300 and epsg 31370 settings :

the .prj file:

- the false easting and false northing figures equal false easting and false northing from epsg 31300

- the central meridian figure equals the central meridian figure from epsg 31370

- the standard parallel 1 figure equals the standard parallel 2 figure from epsg 31370 (SP1 and SP 2 are inverted)

- the standard parallel 2 figure equals the standard parallel 1 figure from epsg 31370 (SP1 and SP 2 are inverted)

yves61
438 post(s)
#19-Oct-18 10:20

I tested Manifold 8 versus QGIS (2.14.9 Essen) ( I know already outdated but still).

With QGIS

- I have Bing Aerial as a background

- I created a new vector layer, and imported my original drawings, having made no changes to the original .prj files !

- Both the automatic spatial reference made by QGIS or setting manually the spatial reference to EPSG 31370 show my drawings well aligned with Bing Aerial

- If I set the spatial reference of my imported drawings to EPSG:31300 then QGIS is showing my drawings totally off position compared with EPSG:31370. EPSG:31370 seems right for QGIS

So, what goes wrong within Manifold 8 ?

Virtual Earth versus my drawings : is manifold making wrong assumptions, applying wrong projections or making wrong transforms in order to align my drawings with VE ? Is this a bug ?

Attachments:
QGIS_aligns well.PNG

yves61
438 post(s)
#19-Oct-18 10:50

Just to compare.

Attached image with Manifold aligning wrongly.

Can this be reproduced by you fellows ? May be even in manifold Viewer 9 ?

Attachments:
Manifold_aligns wrong.PNG

Dimitri


7,413 post(s)
#19-Oct-18 11:52

Can this be reproduced by you fellows ?

No one here can reproduce what you are doing without your data. Please post a link to your original drawing in the original format.

May be even in manifold Viewer 9 ?

What happens when you try in Viewer?

yves61
438 post(s)
#19-Oct-18 12:06

Check the .map attached on top of this page.

Does this help ?

Dimitri


7,413 post(s)
#19-Oct-18 14:52

Does this help ?

No, because it is not the original drawing in the original format. It only shows what were the results after you did whatever you did to import the drawing.

When there is a problem like you describe, it is almost always one of three errors:

1. The starting data has an error in it.

2. There was some sort of error made during import or during assignment of the initial coordinate system.

3. There is an error in Manifold.

To find out which of the above three, we need to start with the original drawing in the original format.

yves61
438 post(s)
#19-Oct-18 14:10

Please find attached a new manifold 8 .map file with the drawings to check on.

Attachments:
VE_FE_FN_43m_v02.map

yves61
438 post(s)
#19-Oct-18 14:35

I have now some additional comments in the .map file on the projections.

Check this new version of this manifold 8 .map file attached;

Hope someone may shed some light on this issue. Have this problem since years with all shapefiles downloaded from the same source.

I remember that also for google satellite imageserver also offset problems occured although the false easting and northing distance need be corrected with about 27 m with Google instead of 43 m with VE.

Attachments:
VE_FE_FN_43m_v03.map

yves61
438 post(s)
#19-Oct-18 14:48

Dimitri,

You asked what happens when using Viewer 9. -->problem remains.

See attached image.

Attachments:
Manifold Viewer 9_ same problem.PNG

Dimitri


7,413 post(s)
#19-Oct-18 14:58

Have this problem since years with all shapefiles downloaded from the same source.

It is really stupid to speculate without seeing the original drawing in the original format, but what the heck... I'll speculate anyway:

Long experience indicates that when there is a recurring problem with shapefiles downloaded from the same source, but not from other sources, the problem is usually some systematic error in the files that source provides.

But that's all one can say with taking a look at the data in original form.

yves61
438 post(s)
#19-Oct-18 16:00

Sorry, I thought the .map file would be original enough, but not so;

No harm, I will fetch a fresh set of files. They encompass a larger area though. I cropped the area somehow to have less heavy files to upload.

I will upload the original files as soon as I have them. They may be under license terms though.

Dimitri


7,413 post(s)
#19-Oct-18 16:22

I will upload the original files as soon as I have them. They may be under license terms though.

The forum does not allow upload of larger files. You can always post a link here to a cloud account or some other account from which the original files can be downloaded. However, if files are subject to licensing you might not be allowed to post a link to them in a forum that the entire world can read. :-)

What you could always do is to ask for tech support. If you are convinced the problem is a bug in Manifold you could always file a bug report. In either case, you could provide files of any size by uploading them to Tech Support (they'll provide instructions). Tech also works with confidential data all the time.

If you need to get this resolved the right path is to launch a tech support incident. See the discussion in the a bug reportpage.

yves61
438 post(s)
#19-Oct-18 16:15

(deleted previous reply.)

I will send the zipped files to tech for bug report.

Dimitri


7,413 post(s)
#19-Oct-18 16:44

Great!

Manifold User Community Use Agreement Copyright (C) 2007-2021 Manifold Software Limited. All rights reserved.