GÇó Overview map needs an overhaul. There's too much information that one can gleam by simply going to the over view map. Where is that sniper? Where is that tank? Where is the bulk of enemy forces? Just go to the overview map and its there; anything that isn't blue is an enemy, you don't need the red tags to figure that out. And It shouldn't be that way. I should have to coordinate with scouts, dropships, LAVs, capsuleers, etc. or drop some specialized equipment on the map for scans. There needs to be a "fog of war" to allow information gathering to be critical in game.

To get the "fog of war" I mentioned, we need to change the way the map is presented. We can't simply render the entire map and choose not to render in the enemies because that would not make sense immersion wise. I can see the entire field from a camera above the battle but I cant see my un-scanned enemies? I recommend changing the current live video stream overview map into a contour map that some unknown sensors took a 3D scan from orbit. Don't forget to add elevation, buildings and obstacles, longitude and latitude. A benefit of going this route may be shorter times in bringing up the overview map.

I was pretty happy when I saw the compass on the top of the screen (^ v ^)

I know they previously looked into a grid overlay on the overview map before, but they ran into issues with the terrain being in 3D and the UI being done in 2D Postprocess--this doesn't work well because the terrain has depth and the grid overlay is flat and above the terrain--it will not line up properly.

It can be achieved if the grid is done in 3D. There is one shader technique where you can draw pixels only at the intersection points of polygons. I attempted a small prototype by adding some polygons that intersected with the terrain (they are huge), but the thickness of the 'line' changes too much with slope and distance, so we would still need to figure out a way to improve this (and there are ways). Another way this could be done is by building it into the material on the terrain and use WorldSpace mapping coordinates (basically you can use the x,y value as a texture coordinate). This seems better and I'm leaning more towards it (although the first technique would be useful if we wanted to show a scan effect going across the landscape). Still there is a roadblock in figuring out how to apply this grid material--we don't want to it as part of the terrain material all the time--some of the starmap work is requiring something similar, so I'm hopeful we can learn from that.