Wrong coordinates on Caucasus map

[TurboHog]

Hey,

 

I've made a simple tool in Excel to aid in navigating DCS modules that have no advanced navigation systems such as the Huey and the Mig-21. You enter the coordinates of your waypoints in decimal degrees and the tool automatically calculates (1) distance between waypoints; (2) initial bearing; (3) final bearing; and (4) straight-to heading.

 

 

Distance is calculated using the haversine formula:

 

a = sin²(Δφ/2) + cos φ1*⋅ cos φ2*⋅ sin²(Δλ/2)

c = 2 ⋅ atan2( √a, √(1−a)*)

d = R ⋅ c

 

Initial bearing was calculated using the following formula:

 

θ = atan2( sin Δλ ⋅ cos φ2*, cos φ1*⋅ sin φ2*− sin φ1*⋅ cos φ2*⋅ cos Δλ )

 

wherein,

 

φ = Latitude

λ = Longitude

R = Earth's Radius

 

 

I noticed that there is a large discrepancy between calculated bearing and bearing as indicated in the DCS mission planner or on the F10 map. Distances also show a significant difference with DCS calculated distance, whereas the haversine formula should perform well even at meter-scale distances.

 

After ruling out the possibility that my calculations were wrong, I simply took the coordinates of a random point on the DCSW map (Novorossiysk airport) and entered the coordinates into google maps. Guess what? In reality, this point is located 20nm to the southwest.

 

So for your information: refrain from using real world navigational tools in DCS (or tools that are based on real world navigation, such as my tool). And for ED: please correct this.

[wolf5]

Try to subtract 6°

There is a difference between heading in map and magnetic heading

[TurboHog]

Try to subtract 6°

There is a difference between heading in map and magnetic heading

 

No this has nothing to do with Magnetical variance. I'm taking map coordinates and compare them. Everything is in reference to true North.

 

Besides differences are irregular and more significant than that, especially at shorter distances.

 

And please read the part about the 20nm offset between DCS and real world coordinates. That alone rules out anything that has to do with the Magnetical North.

[Nirvi]

No this has nothing to do with Magnetical variance. I'm taking map coordinates and compare them. Everything is in reference to true North.

 

Besides differences are irregular and more significant than that, especially at shorter distances.

 

And please read the part about the 20nm offset between DCS and real world coordinates. That alone rules out anything that has to do with the Magnetical North.

 

The black sea map in dcs is completely flat, not round like the actual world. So everything is stretched.

[TurboHog]

The black sea map in dcs is completely flat, not round like the actual world. So everything is stretched.

 

Yes that could explain the difference! I didn't know that the Caucasus map in DCS does not follow an Earth ellipsoid model. I always assumed it did, maybe because you can select an ellipsoid model in the Ka-50's ABRIS. That must have been purely cosmetic. It is a real shame that real world navigation aids don't work. I hope this will be different for future maps!

[wolf5]

My bad, reading too fast :doh:

[Grimes]

Its a weird system overall. The map origin, which is somewhere in the Crimea, is the only part of the map that points to true north if you try to get the heading of a unit based on its unit vector. As you move east and west from the origin the deviation becomes greater.

 

Using coordinates grabbed from DCS match the real world fairly well, though its not always exact. I believe they don't match up precisely due to the map not being a 1:1 copy of the real area, but the variation is close enough that it isn't a huge problem. I've attached a screenshot where-in I had used the scripting engine to search for all map objects on the Caucuses map, converted the coordinates using coord.LOtoLL, and then write the data in a kml format for google earth. Pictured are the 4 bridge objects that make up the bridge crossing the Inguri River in DCS.

 

I don't know what the actual conversion from LOtoLL entails, but it (or the inverse) must be done if you intend on switching between the two coordinate systems. If you figure it out, there are other people I know who would like to know the answer.

[QuiGon]

What about NTTR? Is it based on an ellipsoid model with correct coordinates?

[Reksi-Erkki]

After ruling out the possibility that my calculations were wrong, I simply took the coordinates of a random point on the DCSW map (Novorossiysk airport) and entered the coordinates into google maps. Guess what? In reality, this point is located 20nm to the southwest.

 

That's Gelendzhik... the Novorossiysk airport doesn't exist anymore, at least not according to Google maps :)

 

I checked a few airfield coordinates (from the F10 map, using the airfield icon as reference) and Google maps places them either on the runway or very close.

 

BTW, I just completed my Excel sheet for exactly the same purpose... the calculated distances are within 4% of the measured ones, and some of that is probably caused by my inaccurate measurements - I didn't bother zooming in so that I could have used the exact runway centerpoints. It looks like the acos formula would give better results, though.

 

I calculated the average bearings between the initial and final ones. They seem to contain an error of five to seven degrees - suspiciously close to the magnetic deviation in the area! I wonder if the F10 map distance/heading tool actually shows magnetic headings?

[Wolf Rider]

also got to keep in mind though, that " current real world data" will offer some disparage between real world and game world because of the ever

shifting (non conspiracy theory, but fact) magnetic poles...