It's not clean. It's not elegant. But if it works, it would re-open up that thing called water to the game. Here it is:

copy and paste the mining gadget. slightly alter it so it stands out as water_mining or whatever. With a separate definition system, simply have the water miners go to the water drop off points. As two different systems, they shouldn't intermingle... in my mind. Of course not knowing exactly how it works, I could be wrong.

If it does work, I'd like to implement this so we can get to work making water maps and new water units.

boats or subs?

As miners? Probably subs. Maybe boats.

having duplicate code is fail, you have to update stuff in 2 places.
to seperate land/water mining would just need a is_in_water check in getnearestdropoff() and searchforrocks()
but water will have many other problems like shallow water (rocks falls in, too shallow for mining boats to reach), cruisers being able to be landed in water etc so you should really plan out what you want.

If you know how to enable it, can you update the code then? I see these other problems as minor things to be worked out. The current system makes our game unplayable on maps with a lot of water, and that is both not cool and limiting.

Units wise, what I'm thinking right now would be a submersible mining unit, and the drop off would be a unit mimicking the "cruiser" drop off point, just a floating version that morphs into a mobile "boat" form instead of a "cruiser" form.

Hover Bengi?

Isn'tNavy disabled anyway until land balance is finished?

The idea is to basically allow mining of different resources ("rocks", "wood", "gold") all with their own miners/dropoffs etc and then land/water rocks would internally be handled as two different resources but give the same "real resource", metal. But that must wait atm because other thing and rl blabla.
Of course you can try your idea in the meantime, to see how it plays out balance wise or whatnot.

Butis not really true anyway I think. TA is the only game I can think off where you can build a complete eco on a water-only map. In other games with sea warfare ie age of empires, you always need to some land to collect wood etc...

What he was was getting at was that you lose so many drops because they hit the water and vanish.

Land balance is more or less fine for now. I want to get the ball rolling for re-entry of the navy. And a good first step is to make an economy that can exist on a map with more than a pond or two.

A naval mining unit and drop off point will be put on my short list of things to model and get in-game. When I get those ready, you'll be able to get water mining going.

To Knorke: How difficult would it be to have the spacerocks gadget check for a viable landing location? Like just dry land. Could it check for dry land, and if the location is over water, immediately choose another spot and continue the check.

Perhaps this is something where you could put some land/naval differentiation? Have the sea-going economy work a little different?

Personally, I'd make the standard collector amphibious. That way you can get things that fall in small ponds around your base that aren't worth investing in a major unit to access. But the depot is still land based, and they're slow underwater, so this isn't optimal.

It also functions as a fallback if you lose your naval harvesters.

Then, for then naval harvester, I'd go with a completely different harvesting approach. Some big, expensive mobile harvesting platform that doesn't require a depot and has a long reach for harvesting nearby ore.

tl;dr - Pxtl spams out more random crappy ideas.

Snoop and I talked a bit and we've more or less decided that as long as each meteor storm falls on dry land, there is no significant need for a naval economy. The biggest push for water mining was to open up the playability of water dominant maps, or just maps with good chunks of water.

erm unless you have 5 million kinds of unit type that are capable of mining I dont see why you cant just define a list of pickup points and the types of miners able to reach them.

Failing that just do a test to see if the unit is capable of reaching the position of the pickup, which has the added bonus of protecting from miners on island A being stuck trying to reach the pickup point on island B

thats what i have planned to do.


how

I dont know the API call specifics but cant you query the pathfinder in lua gadgets and do a test?

http://springrts.com/wiki/Lua_SyncedRead#PathFinding
well, maybe Spring.GetUnitEstimatedPath returns nil or something if no path is found.

Split the map into a grid, use a bucket fill algorithm to fill in each square that is land or water, resulting in landmasses and water masses.

http://www.youtube.com/watch?v=mxwAiGRcaCw

Not really, its actually a simpler and faster algorithm than a pathfinder, and should come in handy elsewhere