Roads give movement bonus and NOTHING else?

[Harovan]

In very narrow limits, the Civ3 AI already had the ability to defend strategically important tiles. I have seen it several times, that it would place a fortified spearman on an isthmus even in the ancient age. I wonder, if it has some kind of underlying map layer, where every tile has a numeric value about it's strategical importance, or perhaps a set of flags. If it has (at least in Civ4), it could be used for this purpose. If not, it would have to be created.

[Willem]

Originally posted by nbarclay
If invasions were possible only in particular places, it would probably take serious modding to the AI to have it focus its coastal defense specifically on places that are vulnerable. There would also likely be problems where until amphibious assault units are available, players could stop invasions merely by parking units on all the invadable tiles. (As things stand now, there are generally too many invadable tiles for it to be practical to park units on all of them.)

I don't think these issues are insurmountable though. The AI could treat the coastal squares the same way they do in Civ 3 with Victory locations, only using more units to defend them and possibly building Forts. And adding earlier amphibious units wouldn't be any problem. I think it would be cool to have our own little versions of Normandy in a game.

And of course you could always arrange an ROP with some nearby civ and use their landing areas to launch an attack on an enemy.

[The_Aussie_Lurker]

What gives me hope is the fact that many terrain tiles now have TWO designations-such as Desert/Hill, Tundra Hill and Forest/Hill-just as examples.
If this is truly the case, then maybe it is possible to designate coastal tiles as Coast/Plains (Beach) or Coast/Hills (Sea Cliffs) or even Coast/Mountains high Sea Cliffs). Coast mountains would automatically be denied to both sides-as mountains are now impassable, wheras coast/hills would give the defender the hill bonus. As for putting reefs in, this would be a case of adding a 'reef' terrain, and have it only appear with coast or sea/ocean tiles.
I think it is more do-able than ever before! In fact, maybe Marcos or Solver can let us know if cliffs, reefs and beaches are already IN the game!!

Yours,
Aussie_Lurker.

[Solver]

What gives me hope is the fact that many terrain tiles now have TWO designations-such as Desert/Hill, Tundra Hill and Forest/Hill-just as examples.

Correct.

[korn469]

Originally posted by Solver

What gives me hope is the fact that many terrain tiles now have TWO designations-such as Desert/Hill, Tundra Hill and Forest/Hill-just as examples.

Correct.

Nice to know! The implications of this feature make me very excited.

[snoopy369]

Originally posted by The_Aussie_Lurker
What gives me hope is the fact that many terrain tiles now have TWO designations-such as Desert/Hill, Tundra Hill and Forest/Hill-just as examples.
If this is truly the case, then maybe it is possible to designate coastal tiles as Coast/Plains (Beach) or Coast/Hills (Sea Cliffs) or even Coast/Mountains high Sea Cliffs). Coast mountains would automatically be denied to both sides-as mountains are now impassable, wheras coast/hills would give the defender the hill bonus. As for putting reefs in, this would be a case of adding a 'reef' terrain, and have it only appear with coast or sea/ocean tiles.
I think it is more do-able than ever before! In fact, maybe Marcos or Solver can let us know if cliffs, reefs and beaches are already IN the game!!

Yours,
Aussie_Lurker.

You're right, i forgot about the mountain thing. I think what you describe is exactly correct - and quite moddable even if it's not actually in the game. The dual designation is clearly intended to be used in this sort of way

[SpencerH]

The trick will be whether the random map creation can be modified to include the dual designation i.e. we dont want a coastline to be made up of something bizarre such as alternate coast tile types.

[snoopy369]

true ... but at minimum you can create a map that uses them

Actually, thinking more about this, it's probably moddable in Python. If they use the Markov matrix-style terrain generator I suspect they do ... they'll probably give you the option of modifying this via SDK. Then it's just a matter of giving each coastal terrain type a high self-probability, and a much lower other-coast-probability; you can also use the probabilities relative to the other terrain types (both water and land) to make it more realistic (ie cliffs happen when there are hills or mountains nearby and/or deeper ocean, while beach happens near shallow coastal waters and/or plains). That's a pretty simple thing to modify, anyway, it just takes a lot of work to BALANCE it.

But that's why we have a dedicated community

[SpencerH]

and we look forward to your new terrain art

[alva]

Different dog.

[SpencerH]

Thats confusing.

[Jaybe]

Spencer, snoopy369 is not Sn00py. Sn00py does the artwork.

[Adagio]

Confusing dogs

[Nikolai]

/me awaits Sn00p Dogg.

[SpencerH]

Originally posted by Jaybe
Spencer, snoopy369 is not Sn00py. Sn00py does the artwork.

Thanks, I understood Alva's comment, I just wonder why people choose similar names.