Magellanic Metro (Map)

[Lost Seedling]

The Magellanic Metro
After conquering The Green Planet, and in order to survive, the Seedling Empire must expand their territory in that previously unexplored section of the Galaxy.

What new worlds and life-forms are waiting to be found? What artifacts of alien civilizations or astronomical wonders will they encounter?

Two things are known- ruthless enemies await them, and that region of space holds a bizarre secret not yet seen by seedling eyes!

Features:

Merchant AI version V1.1
Full-time Game Statistics Display (Chart Stats Function)
Random Number of Enemy Empires (2-5)
Easter Eggs
Difficulty Level? Dependent on random factors- number of enemy factions, spawn locations, and most importantly- player skill!

Notes:
Please report your experience with the Merchant AI. Does it stay active throughout the game?
Comments, suggestions for improvement, or reports of problems are always welcome and requested.

I am aware of a graphical anomaly which occasionally will appear while Charts are activated. This seems to be associated with extensive use of Drawing functions, or perhaps a memory problem with my computer- not sure which. Nevertheless, it does not affect gameplay. If you played Visit to a Green Planet, you will note an improvement over the illustrations in the game. Through use of the Eufloria Arts Project, what would have taken many hours to draw (if at all possible) now takes mere seconds to minutes to complete.

I hope you find it entertaining.

Fix/Updates:
01/13/12
Fixed Start and End Splash Screen Graphics to maintain alignment at different screen sizes.
Fixed Asteroid Speed changing too early if player spawns on moving asteroid.

[Rudolf]

When the iOS and other ports are done I will have about 2 years worth of wonderful custom maps to catch up on.
0_0

:-)

[Aino]

This is just awesome, but I hope you have modified Merchant AI to respect Moving asteroids, because it doesn't do that. Infected AI scan neighbours on the fly, which is what causes the lag if you have many asteroids. But I see I'll have to add a scan method for individual asteroids

[Lost Seedling]

If you play the map, you will see I took measures to alleviate any problems with moving asteroids. I haven't inspected your AI code very closely, but I'm surprised that would even be an issue. Nevertheless, in my play-testing, I never noticed any particular problem with that. I'm very interested in hearing what others think of the AI behavior, though.

I wish Mihhaelo and Pilchard had time to advance their AI's.

[Aino]

When the iOS and other ports are done I will have about 2 years worth of wonderful custom maps to catch up on.
0_0

:-)

Rudolf, you and Alex should allow us to code new game elements, like a new tree or seedling.

We atleast need a simplified way to call a seedling, and have functions that are called whenever a seedling die, a tree is destroyed, Asteroids loose health, a seedling is attacked, a player command is done, etc...
This will allow for much much much more diversity throughout the custom maps. Also keybindings and drawing in layers(or prioritizing the drawing, meaning things will overlap).

[Lost Seedling]

Aino- that's why you need to create your own game! I know you'll do that sooner or later. Can I reserve a copy now?

BTW, if anyone noticed my spelling error you can download the corrected version.

[Aino]

Aino- that's why you need to create your own game! I know you'll do that sooner or later. Can I reserve a copy now?

Hehe, I have already done a lot of games, one is actually Eufloria with magics, like bases of the basic elements with units. Yet, there is much I don't know and I have no real project I'm working on, I'm just adding things as they come to my mind, which is not a good way of working with stuff. I still don't know what everything do either, so experimenting is something I'm still doing

But I want this in Eufloria too, because it's simple to add(I don't know how simple it is to add into the Lua integration) and it's so useful for everyone!

[Lost Seedling]

Aino, I've played this map now a good number of times and have noticed a few definite strengths and weaknesses in MAI.

A very good tactic it uses, intentionally or not, is to keep the seedlings in near-constant motion for a good portion of the game. Although not necessarily attacking an enemy asteroid, by keeping the seedlings in between asteroids, it avoids having asteroids reach the seed cap and thus no longer generating new seedlings. As you can see in the attached (labeled) picture, even though MAI was not expanding to new asteroids between the times of C and D, it continued to generate seedlings between times A and B beyond what the asteroids would normally hold. It did this by constantly keeping the seedlings in motion.

The unfortunate thing it does, however, is fail to expand aggressively when it has the opportunity. As you can see, it had numerically superior forces for the majority of the game, but would not attack en mass to overcome my weakly-defended front. This is something Infected is very, very good at. When MAI does attack, it unfortunately splits it's forces as if it is still expanding it's empire. So, even though it is numerically superior, it often would only send half of it's force to my asteroid, and send the other half back to the rear. This is very effective when expanding it's empire, but not so much later in the game.

I should point out that this technique of splitting forces more evenly around the empire does have an advantage when there are multiple fronts. One way of defeating Infected is by baiting it to send all it's forces in one direction, while amassing an overwhelming force to attack it's then weakly-defended area. So both methods have their benefits. But MAI fails in that when it DOES attack a heavily defended asteroid, it often fails to send a numerically superior force when one is readily available.

I've also noticed that although your recent update to MAI has improved it quite a bit, it still seems content at times to go into a sort of deep-freeze, as if it is no longer "thinking". Either that, or it just keeps moving seedlings around it's own Empire, as noted above.

That all said, I do find this map challenging to win against MAI. If MAI stayed more active in the latter stages, it would (for me) be very difficult.

I'm looking forward to observing more AI behaviors once the Web Wide Wars is finalized!

[Pilchard123]

On a semi-related note, are there any AI test maps around? I seem to remember a couple, but don't know where they are. This one sounds (haven't played yet) like a good one, if Lost's post is anything to go on.

[Aino]

You can always make your own

But there's tons of others too, I'm just too lazy to look for them :S

[Lost Seedling]

...are there any AI test maps around?

AI Test Map? Not sure what that means but these maps were made for AI contests of sorts:

Mihhaelo's "Man vs. AI"-
http://www.dyson-game.com/smf/index.php?topic=1456.0

Aino's "Square"-
http://www.dyson-game.com/smf/index.php?topic=1322.0

Aino's "AI War"-
http://www.dyson-game.com/smf/index.php?topic=1231.360

Annikk's "AI War Organic"-
http://www.dyson-game.com/smf/index.php?topic=1231.msg10246#msg10246

That's all I can think of right now.
Magellanic Metro would be a very poor map for AI testing.

Hope that's helpful.

[Aino]

Aino, I've played this map now a good number of times and have noticed a few definite strengths and weaknesses in MAI.

A very good tactic it uses, intentionally or not, is to keep the seedlings in near-constant motion for a good portion of the game. Although not necessarily attacking an enemy asteroid, by keeping the seedlings in between asteroids, it avoids having asteroids reach the seed cap and thus no longer generating new seedlings. As you can see in the attached (labeled) picture, even though MAI was not expanding to new asteroids between the times of C and D, it continued to generate seedlings between times A and B beyond what the asteroids would normally hold. It did this by constantly keeping the seedlings in motion.

The unfortunate thing it does, however, is fail to expand aggressively when it has the opportunity. As you can see, it had numerically superior forces for the majority of the game, but would not attack en mass to overcome my weakly-defended front. This is something Infected is very, very good at. When MAI does attack, it unfortunately splits it's forces as if it is still expanding it's empire. So, even though it is numerically superior, it often would only send half of it's force to my asteroid, and send the other half back to the rear. This is very effective when expanding it's empire, but not so much later in the game.

I should point out that this technique of splitting forces more evenly around the empire does have an advantage when there are multiple fronts. One way of defeating Infected is by baiting it to send all it's forces in one direction, while amassing an overwhelming force to attack it's then weakly-defended area. So both methods have their benefits. But MAI fails in that when it DOES attack a heavily defended asteroid, it often fails to send a numerically superior force when one is readily available.

I've also noticed that although your recent update to MAI has improved it quite a bit, it still seems content at times to go into a sort of deep-freeze, as if it is no longer "thinking". Either that, or it just keeps moving seedlings around it's own Empire, as noted above.

That all said, I do find this map challenging to win against MAI. If MAI stayed more active in the latter stages, it would (for me) be very difficult.

I'm looking forward to observing more AI behaviors once the Web Wide Wars is finalized!

Seedling, the attacking on large armies are probably caused by them exploring the asteroid(but they send the whole army, which is lame). MAI has to see whats on the asteroid before it will decide what to do, just like players do. Infected AI doesn't have this limitation and that promoted Infected over MAI whatsoever. Annikk, fix Infected's restrictions
Also, that thing about splitting the army is actually very useful(in some cases, some not) because then the AI can have one big mobile army that can split up to attack each front.

Speaking of MAI, I'm currently adding tree and seedling count before sending armies. This means the asteroid with most trees will get the seeds and the asteroid with least enemy seedlings will get the seeds. I don't think attributes will come to account, but radius could be interesting adding(radius is commonly attributes all together anyways)