The future of Fall from Heaven

[novice]

It didn't help that they stopped 99% from the finish line.

They only finished 1 percent of it?

/math pedant

[sunrise089]

I think they finished about 15% from it (aka 85% done) IMHO. I will reserve final comments for when I've played for high-level games, but it seems unique civ flavors + multiple viable strategies + more dangerous world + better AI play + more were all done very well. But the opening buildup is....almost shockingly bad. I don't want to start an era in because I DO want to have to make meaningful tech choices, but why force the player to hit enter and build warriors for 50t?

[darrelljs]

They only finished 1 percent of it?

I hate when I do that .

Mid game is where FFH shines. Early and late have issues for sure.

Darrell

[Selrahc]

I hate when I do that .

Mid game is where FFH shines. Early and late have issues for sure.

Darrell

But at least in the endgame you get to play around with shiny toys. The beginning of the game really is pretty barren.(Particularly when stretched over a few months rather than a few minutes)

[Bobchillingworth]

The endgame is also where my computer regularly collapses in disgust at trying to manage the hordes of hundreds upon hundreds of units in AI stacks wandering about aimlessly due to the bizarre "barb smasher" script they seem to be enamored with. I've also never successfully gotten the game to run all the way to Armageddon, as the appearance of the Avatar of Wrath and fully half of the countless living units in the world turning barbarian inevitably freezes the program



But yeah you get to play with some awesome toys in the endgame. It's amazing how much the map can change by the time a game finally ends, with the vast variety of terraforming options... it requires an insane amount of micro to optimize even a standard size map, but it also revives the "any land can be made profitable, given enough time & investment" ethos of Civ II which I've sorely missed for many years. And the staggering variety of spells, units, abilities, and strategies you can deploy gives the game far better replay value than vanilla Civ. I've yet to win any two games of FFH with the exact same tactics... I wish I could say the same for BTS.

But at least in the endgame you get to play around with shiny toys. The beginning of the game really is pretty barren.(Particularly when stretched over a few months rather than a few minutes)

Not being able to build or research anything for like 16 turns at the start also makes the game go a little slowly

[sunrise089]

I hate when I do that .

Mid game is where FFH shines. Early and late have issues for sure.

Darrell

I've yet to experience a "lategame."

Tech costs are so ungodly high early that midgame and endgame techs take about the same number of turns due to cottages finally maturing.

I was telling darrell that I had a Lanun (pirate civ, sp?) game on Noble where I was making 1,000bpt without really trying by the end of the game. It's kind of crazy how much the "tech horizontal, not vertical, and choose your path" idea fell aside with that kind of economy - Mithril Working was like a 4t tech. On the other hand it still took me 60t before I could build cottages at the start of the game

Proposed solution - Make cottages buildable from the start, like warriors. That's a real meaningful choice since unlike BtS very early military IS important and cottages take you off hammer tiles. But it at least gives the player an option to get through the early techs quickly.

[Selrahc]

Proposed solution - Make cottages buildable from the start, like warriors. That's a real meaningful choice since unlike BtS very early military IS important and cottages take you off hammer tiles. But it at least gives the player an option to get through the early techs quickly.

That's something that has been toyed with in the various modmods. There is generally an option to enable all improvements without technology. Plantations, pastures, camps, cottages... everything. However without the technology that would normally enable the improvement it builds at only one quarter speed, and doesn't connect resources. It means an early worker has something to do, and that you still need to get to the various technologies.

There are still some problems with this approach however. The main one being that the AI is pretty much unable to deal with it. It spends 30 turns building a windmill on a forested hill, or tries to build lumbermills everywhere.

Not being able to build or research anything for like 16 turns at the start also makes the game go a little slowly

Feh, I'm just talking about the problems that affect me.

[Maksim]

There are still some problems with this approach however. The main one being that the AI is pretty much unable to deal with it. It spends 30 turns building a windmill on a forested hill, or tries to build lumbermills everywhere.

And that, to me, is really going to be the crux of it. Civilization, for all its complexities, is still a rules-based game - as in, you can write one algorithm to look for best city sites, and it will apply to all civilizations equally.

For FFH, the same algorithm would have to be more complex - one civ will work three tiles around each city (Kuriotates) for the first few, and then build settlements only. Another should consider coastal tiles as being worth much more (Lanun's pirate coves - and good luck optimizing their placement!). Every civilization is unique in some way like this.

Once you get into magic and synergies, the game becomes much more situational - e.g., if I go for melee line, then I should aim for Enchantment mana (spell gives +20% for melee units), otherwise it's not too useful.

The rules of FFH lead to optimal strategies being tricky for AI to apply - e.g., it's usually a good idea once you learn Knowledge of Ether to pump out a few adepts so that they have time to get some XP and immediately turn into mages once Sorcery is learnt. There are some magic spheres with a good second-level spell but not so useful first one - Fire - so you can't just give these mages-in-waiting a good first level sphere, and use them for something else. Teaching AI to plan that far ahead is tricky.

All in all, for every spell you'd have to consider how it can be used best, and probably code an algorithm for whether its technology should be researched, a mana node built, adepts trained, wait for mages... And all of this on top of an interesting, nuanced and balanced economic system.

So my personal hope is that they do a "Civ 4: Colonization" style but with Civ 5 engine - get the economic model and the war AI from there, and focus on getting the new stuff right. If they have to do everything from scratch, then I don't have high hopes for them.

[sunrise089]

That's something that has been toyed with in the various modmods. There is generally an option to enable all improvements without technology. Plantations, pastures, camps, cottages... everything. However without the technology that would normally enable the improvement it builds at only one quarter speed, and doesn't connect resources. It means an early worker has something to do, and that you still need to get to the various technologies.

Is this better than my cottages-only proposal? Cottages only is less confusing for the AI and a real trade off (unlike an early camp which makes commerce and hammers) and avoids the very inelegant worker turn nerf.