Jump to content

dcsmap.a10c.de -- another web-based map


FSFIan

Recommended Posts

For some ideas I am toying with, I needed a decent map of DCS: World in a browser. Here's the result:

 

dcsmap.a10c.de

 

  • Drag to pan around
  • Mousewheel or Shift+Drag to zoom
  • Alt+Shift+Drag to rotate

 

The approximate scale is given in the bottom-left corner. In the top-right corner, the position of the mouse pointer is displayed both with decimal degrees and in degrees/minutes/seconds format.

 

The map shows roads, rivers, airports and buildings exported from DCS: World on top of real-world terrain data. This could be useful for VFR navigation (navigating by real-world maps can be difficult if you don't know whether that landmark you want to follow is represented in DCS or not). Power lines can be identified by lines of small buildings (the actual power poles).

 

Future Development

 

Due to the awesome OpenLayers 3 JavaScript library, it is relatively easy to add additional layers to the map, for example from .kml files.

This could be used to display additional information on the map, such as airport frequencies, locations of navaids, etc.

 

If you have any data sources or ideas, please post them here.

 

Right now, this is a side project that I don't have much time to work on, but I will try to make small improvements from time to time. If you are a web developer and are interested to help out, please contact me.

Link to comment
Share on other sites

Ian, I am not a web developer ... BUT if you are going to keep doing browser based stuff ... I might need to have a look at javascript.

 

About the map.

 

The first thing I wondered when you mentioned using real-world terrain data was if the terrain matches up to DCS map. I noticed recently during some testing with Tacview that the map data for the DCS Georgian map does not match up to real-world SRTM data. Features are off by about 100-200m.

 

Vyrt from Tacview says it is a DCS issue and I have not done any testing to prove or disprove this.

 

If it is a DCS issue maybe the original map data for Georgia came coming a Russian source with a slightly different standard and ED themselves did not do the conversion properly when they added lat lon reference points to their own map. It could mean that if we use real world maps and other real world kml references they won't actually line up with in game features. It won't matter for some usage but for other uses it will matter.

 

What Vyrt has done in Tacview is to use the DCS map to generate a custom DEM model only for DCS recordings. Tacview actually uses real world SRTM and OTOPO sources for Falcon BMS and GPS playbacks.

 

It actually would be interesting to know what the original data source was for the Georgian map. With the right transform we might be able to take real terrain data and conform it to DCS topology. That would be a solid solution.

 

I hope that NTTR and the Hormuz map won't have this issue.

 

===

 

On the topic of what you have done, it's really great for a raster solution.I feel that vectors are better for many reasons and it makes me wonder if they would really be that hard to generate?

 

Vectors for buildings - possible. Vectors for powerlines - possible. Vectors for roads, railway lines and rivers - is much harder. Hack the rn3 files? or do some fancy processing on sampled data?

 

Well out of the these visual elements - powerlines would be the easiest to use as a test case for vector output. Export the positions of the poles from DCS using LUA, connect them up pole-cord-pole-cord and terminate them at transformer blocks.

 

BUT issue number one will still be unsolved. Real-world terrain data still won't match up with DCS terrain. Flyboys probably won't care but ... I'm focusing on CA and on the ground 100m-200m is too much error to have.

Link to comment
Share on other sites

The first thing I wondered when you mentioned using real-world terrain data was if the terrain matches up to DCS map. I noticed recently during some testing with Tacview that the map data for the DCS Georgian map does not match up to real-world SRTM data. Features are off by about 100-200m.

 

The streets are definitely off. To see this, go to dcsmap.a10c.de, open the JavaScript console (F12 should work in all major browsers), paste in the following and press enter:

map.addLayer(new ol.layer.Tile({ source: new ol.source.OSM(), opacity: 0.5 }))

This adds a half-transparent OpenStreetMap layer on top of the map. I verified that it's not a problem with the way I export data -- a Ka-50 landed on an intersection will display the same coordinates that the map shows. It looks like the terrain data (at least the coastlines) suffer from the same problem.

 

If it is a DCS issue maybe the original map data for Georgia came coming a Russian source with a slightly different standard and ED themselves did not do the conversion properly when they added lat lon reference points to their own map. It could mean that if we use real world maps and other real world kml references they won't actually line up with in game features. It won't matter for some usage but for other uses it will matter.

I agree.

 

It actually would be interesting to know what the original data source was for the Georgian map. With the right transform we might be able to take real terrain data and conform it to DCS topology. That would be a solid solution.

That would be cool, but I feel like I am already in over my head with all this GIS stuff (it's fun to learn though).

 

On the topic of what you have done, it's really great for a raster solution.I feel that vectors are better for many reasons and it makes me wonder if they would really be that hard to generate?

 

Vectors for buildings - possible. Vectors for powerlines - possible. Vectors for roads, railway lines and rivers - is much harder. Hack the rn3 files? or do some fancy processing on sampled data?

There is one other problem to solve. There are about 800,000 buildings in the current DCS: World map. Asking a browser to render them all does not work (that's why the building layer is raster data right now). I had to split the data up into several GeoJSON files to avoid crashing TileMill (maybe I should have set up a proper PostGIS database). I think this can be solved with vector tiles, I'll have to look into that.

 

I would really like to know more about the format of those .rn3 files. I tried to reverse-engineer them once but did not get very far. (All I managed to do is change the color of the roads on the zoomed-out DCS Mission Editor overview map.)

 

BUT issue number one will still be unsolved. Real-world terrain data still won't match up with DCS terrain. Flyboys probably won't care but ... I'm focusing on CA and on the ground 100m-200m is too much error to have.

For this map, we could do the same thing that I assume was done for TacView and export it pixel by pixel using land.getHeight(). That would take some work on my export code and a few hours or days of computation, so I have not bothered with that so far.

 

Finding a way to transform between DCS: World lat/lon and real-world lat/lon would be a much cleaner solution.

Link to comment
Share on other sites

I like where this is going. :)

Windows 7 64 Home Premium, i5 3570K (3.4 @ 4.4GHz), Asus P8Z77-V LX, 16GB dual channel 1600 ram, EVGA Nvidia GTX980ti, 240 GB OCZ SSD, 3 TB Raptor, Thrustmaster Warthog Hotas and Throttle, Saitek Pro Combat Rudder pedals.

Link to comment
Share on other sites

I would love to see a way to collaborate/draw stuff on this map. But this requires more work then just a new layer...

 

Check the "mission planner" link in my sig :) It's not that nice because the map is an earlier version (and in EPSG:4326 projection, so circles become ellipses), but it works now.

 

Eventually, I'd like to put similar functionality on dcsmap.a10c.de, but as you said, that is more work than adding some static data.

Link to comment
Share on other sites

That would be cool, but I feel like I am already in over my head with all this GIS stuff (it's fun to learn though).

 

Exactly my position. And I decided to spend a little time reading about datums and projections and geodetic formats. Still over my head but ... I know little more of the language.

 

Finding a way to transform between DCS: World lat/lon and real-world lat/lon would be a much cleaner solution.

 

This is what I looked at and I wondered if the geo-location data you get in the A-10C cockpit is different to the geo-location data returned by the LUA scripting system.

 

So I googled that and ED forums wins again.

 

It's an interesting thread but this post sums up the issue.

http://forums.eagle.ru/showpost.php?p=896859&postcount=149

 

If igormk is right about the difference in projections then his tip of adding 0.11 minutes to the longitude will be a fix.

 

This is a great tool for testing geo-locations with different projections -> http://www.tool-online.com/en/index.php


Edited by vicx
added link
Link to comment
Share on other sites

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...