Starting units given (to conqueror civs?)

[tlaurila]

Okay, so I think I read somewhere that VEM now gives several "wide empire" civs who seemed to be doing consistently poorly a handful of units at the start, including settlers.

This seems to have overcorrected the issue rather massively. The game seems to have become hugely more difficult as a result.

I started 3 new games on Emperor. Small map, continents, epic game speed. In every one of these games I got swarmed and stomped stupid by my nearest neighbor. England, Mongolia and Aztec.

They attacked with 6-10 units, mostly archers, at about the time I had built my first unit/building in my capital. I didn't quite stand a snowball's chance in hell to recover after this assault brutalized my start, even if the silly AI can't figure to conquer my capital after bombarding it to 0hp with archers.

On top of this the AI seems to have had 2-4 settlers from the start, and with those cities made classical era before I could research 3 things. This sounds like Deity more than Emperor.

Yes, I could move down the difficulty, but recent changes to the mod seems to have altered difficulty a lot. And if those units are only give to certain AI, then difficulty depends massively on who starts as your neighbor. Finally, altering difficulty with AI starting units means that if the human player can get a foothold, then the game from mediaval on is likely boringly easy sailing.

If a boost is desired to certain civs when on AI, so that you can't simply modify starting units in Handicaps, then how about giving it something much more gradual than starting units. Especially settlers. Like 20gp per turn for 200 turns (modify with appropriate game speed) it could use to buy things. Of course it would be much better to alter the actual mechanics of the game to balance wide and conquerors vs talls, rather than glue on top this extra units/gold/whatever thing.

Really, by this experience, VEM right now seems rather ridiculous.

 

[Thalassicus]

You might also try Oligarchy if you're having trouble, or 1 lower difficulty level. Building an archer in each of your first 2 cities also holds off rushes easily.

The problem is the badly-designed rush strategy Firaxis hardcoded into the early game in March. Without the bonuses militaristic AIs end up with very low scores because of their poor target selection (they go for major civs instead of citystates). We're unable to change this - it's in the game core only Firaxis has access to. I've been experimenting with ways to compensate for this weakness so the game is about as challenging as it was in February. Vanilla starts militaristic AIs with 2 warriors and 1 scout. I changed this to 3 archers in the mod, and moved 1 settler from Deity to Emperor.

 

[Bridger]

Vanilla starts militaristic AIs with 2 warriors and 1 scout.

This bugs the hell out of me. It's one thing to have the AI "cheat" to make the game more difficult (because making a good AI is incredibly hard for a game as complex as this) but it's another to rub that cheating right in your face. I wind up discovering another civ and seeing 3 military units on turn 5 in Epic (on KING!) and it just ruins the atmosphere of the game. Without it, you can kind of pretend that the computer is not cheating. It also completely screws up my expectations, to the point where I purposefully settled aggressively early in the game against England, and she attacked with an army of 10 units right after I built my 2nd city. What?! I had moved my archer and warrior in expecting an attack, but there's no way I could have had even 5 units by that time in the game.

Is the AI really so bad that they have to give it such a huge advantage? I assume in addition to bonus units all AIs also get production/gold/food/happiness bonuses as well?

 

[Ahriman]

Is the AI really so bad that they have to give it such a huge advantage?

Yes. It is a good way of forcing you to build some military up in the early game rather than mass-expanding. With a couple of archers and a starting warrior you can fight off a very large enemy force.

 

[Dunkah]

Early Rush Advice: Keep your warrior in your city. When the countryside is full of enemy units, don't use your warrior to kill them. Hit them with the city attack or a nearby archer to inflict the killing blow. Attack with the warrior when you are sure you won't leave the protection of the city. (Otherwise the AI will gang up and kill your only melee unit).
Archers can take 1 attack usually. Coordinate your attacks so that you warrior attacks, then city then archer so you have maximum damage inflicted on a single target. Try to keep archer so only the unit that is being attacked can strike back. Usually it will try and heal itself rather than attack.
Then back off until you are healed again.

 

[Ahriman]

I tend to use the opposite (depending on how many units the enemy has); with 2 archers and a warrior, I'll leave the warrior (with drill 1 promotion usually) fortified on a forest next to the city, keep an archer in the city, and keep an archer on a hill adjacent, or on the other side of a river.
I would much rather have my warrior take the blows (and then run or even die if necessary) than my archers.

 

[tlaurila]

Is the AI really so bad that they have to give it such a huge advantage?

Yes, I have to say that it is. It does still matter how the advantage is given.

While this early rush is certainly a way to challenge the human player, I do think right now it's overdone, at least at Emperor where the AI bonuses otherwise aren't humongous. The Rush is very fickle, I've been gang-rushed by several AIs, who correctly view me as weak (2 units against their 15), and then when one declares war, the other sees me as even weaker fruit ripe for the picking. 20-30 units attacking at around turn 70 at epic speed. On the other hand, if you don't get rushed by militaristic neighbors, Emperor is a cakewalk.

The most fun game is one of steadily climbing challenge to a ultimate confrontation, after which you can say you've won and victory should follow relatively quick. In complete contrast, most Rush games are either nigh-impossible or you survived by overwhelming and glaring tactical AI stupidity. Like, the enemy brought in 8 archers and redlined my city for a dozen turns, but I managed to kill the only warrior and then snipe at archers steadily. Or the games are cakewalks where you didn't get rushed because of the starting conditions.

Isn't it really possible to replace AI starting units by unit production and upkeep bonuses? Easier to build units than peaceful stuff should also help militaristic AIs against non-militaristic, it's just a matter of how much? Being attacked by those 30 units at turn 70 or 150 makes for a world of difference in how the human can prepare. Also later rushes would include catapults, so you have to fight somewhat less "exploitative porcupine". And those production bonuses keep with the AI until the very end game, keeping a better scaling challenge.

 

[tlaurila]

Another report:

This time I started with Egyptians, got Aztecs, Rome and Germany as my neighbors. In turn 50 (epic speed), actecs swarmed with 6 jaguars, 1 archer. Just as the attack landed, I had researched masonry, but didn't have the gold to buy walls. When I died, I had ~5 turns until Oligarchy. Nothing I could see I could have done to get walls or oligarchy going before that attack landed.

The only chance to survive would be to get some of the other AIs to attack Aztecs before they swarmed me. I shall try that re-starting the game.

Come on now. This is just too much of a initial rush. It's supposed to be difficulty level, not impossibility level. Oh, and couple this with the fact that if there's no rushers in your neighborhood, the game is a piece of cake. This way the game is never challenging and fun, it's either impossible or too easy, depending on the start.

 

[Ahriman]

Rushing to walls is not a good strategy, they aren't going to help you much. What you need are a couple of military units: archers, or chariot archers.

I wonder though; the people having difficulty are those playing on slower speeds, where the free AI stuff is a relatively much larger deal.
Is there a way that the amount of free stuff could be scaled by game speed?

 

[Seek]

I wonder though; the people having difficulty are those playing on slower speeds, where the free AI stuff is a relatively much larger deal.
Is there a way that the amount of free stuff could be scaled by game speed?

I'm not convinced it's necessary - when people are using unusual and punishing settings like the OP (a small map, which almost guarantees a close neighbor, and epic, where it's difficult to get anything happening quickly) it shouldn't be a guideline for balance.

I think it's more typical that people match speed and size (small/quick, epic/large, etc.) and that's what Thal has explicitly stated that VEM is balanced around.

 

[Txurce]

I wrote recently of being shocked when I stumbled upon a thriving AI civ (with units) 40 turns into the game. A couple of turns later, they attacked me, followed by a second civ. At that moment, I thought: this is too much. Then I played it out, and survived... barely... by being very careful, using the same tactics Ahriman described. (Side note: I've alternately used Dunkah's approach, and it also works.)

I wouldn't want every game to be like this - and they're not. But I do like some games being like this, because the situation is addressable. I wouldn't even mind if it was a bit harder, so I would on occasion get wiped out.

On a side note, I know what Bridger means about "atmosphere," and choose to view these situations as my civ being a latecomer to more established ones.

 

[tlaurila]

I wrote recently of being shocked when I stumbled upon a thriving AI civ (with units) 40 turns into the game. A couple of turns later, they attacked me, followed by a second civ. At that moment, I thought: this is too much. Then I played it out, and survived... barely... by being very careful, using the same tactics Ahriman described. (Side note: I've alternately used Dunkah's approach, and it also works.)

I wouldn't want every game to be like this - and they're not. But I do like some games being like this, because the situation is addressable. I wouldn't even mind if it was a bit harder, so I would on occasion get wiped out.

Very nice. So you did everything to the absolutely optimal you could game it, and won by the skin of your teeth. Sounds like near max difficulty to me. And how did the game develop from there on? After you barely survived the start, was the rest of the game a continuing challenge, or a cakewalk? If the start is max difficult, then the difficulty should scale to max towards the later game as well.

What I'm saying is, essentially moving the AI free start settler to easier difficulties from deity unbalanced the AI start bonuses as compared to across the board bonuses. The start with extra cities belongs to deity for a good reason. If I nudge down difficulty so that AI doesn't get overwhelming start bonuses as it now is in VEM, then their bonuses across the board are too little to make any challenge in the longer run.

And it's a good catch that this likely depends on game speeds, Ahriman. But I'm doing "only" epic, so should even be more different in Marathon. And I'm not entirely convinced the game would simply be harder to start on slower speeds, as slow speed gives more time for human tactical maneuvers against the rush before AI spams in even more units.

 

[blightmoon]

For early game bonuses, I would suggest that instead of giving out units to the AI from the start, they ought to be given as production bonuses (extra) instead for the first few units. I mean the units would be halfway done or something, although that may be easier said than done. That way, the diplo attitudes are not hostile from the start upon seeing your weak army.

I agree with tpam about marathon. In that mode, there's no room for error. Any mistake in the early game is fatal and is at least boring to recover from (loading from the last production cycle started).

 

[mitsho]

So some civs are hardcoded to attack? Seems like bad design. But I agree, it is survivable, you just need to start building with military units and get one archer up as soon as possible. It's incredible what one archer safe in a city can achieve. As for a boost for Military civs, I'd rather than have them more units at the start give them boosts midgame (f.e. when they build some building in the capital similar to the Korean trait). Generally, it seems to me they just attack to early. What good does a Warrior Rush do to the Mongols? But the main problem is their poor tactics, nothing you can do about that.

Additionally, imho the lower civs reduced to few cities should receive some freebies Mario-Kart style, though not so heavily as it was done there. If Civ 5 veers heavily from the "Historical Simulation" to "Gameplay with historical flavours", we can go all the way at least.

 

[Thalassicus]

Isn't it possible to replace AI starting units by unit production and upkeep bonuses?

The AI already has a 175% unit production bonus on Emperor... increasing it further would clog the map with units in the mid to late game.

It's worth emphasizing what Seek pointed out; it's important to match game speed to map size:

• quick + small
• normal + normal
• epic + large
• marathon + huge

I hadn't noticed you used epic speed with a small map. The 150% movement speed on epic clashes with the 66% land area of a small map, which essentially speeds up warfare to 225% of vanilla. This is probably why you have such a hard time getting anything built before armies arrive.

@mitsho
Yes, all leaders are hardcoded in vanilla to rush a major civ right away if the military situation is favorable. This was added to vanilla in the March patch. I haven't been able to figure out a way to stop it, or redirect the rush to target citystates. It'd be so much better if they took a few citystates instead.

 

[tlaurila]

It's incredible what one archer safe in a city can achieve.

Yes, yes it is. One archer can kill 10 units, but it's only because of how incredibly stupid the AI can be with his attacks. I've had 5 AI archers redline a city for a dozen turns, but they can't figure in a melee unit to finish the job. And it's not a one time event, either.

 

[Ahriman]

I've had 5 AI archers redline a city for a dozen turns, but they can't figure in a melee unit to finish the job.

Yes, getting the AI to do combined arms better to avoid this kind of thing would be a huge improvement. I have observed this often.

 

[Txurce]

Very nice. So you did everything to the absolutely optimal you could game it, and won by the skin of your teeth. Sounds like near max difficulty to me. And how did the game develop from there on? After you barely survived the start, was the rest of the game a continuing challenge, or a cakewalk? If the start is max difficult, then the difficulty should scale to max towards the later game as well.

What I'm saying is, essentially moving the AI free start settler to easier difficulties from deity unbalanced the AI start bonuses as compared to across the board bonuses. The start with extra cities belongs to deity for a good reason. If I nudge down difficulty so that AI doesn't get overwhelming start bonuses as it now is in VEM, then their bonuses across the board are too little to make any challenge in the longer run.

I already noted that particular game was near max difficulty for me. I also noted this was my toughest start - not every game has gone like this. How did it develop from there? My progress toward a Science victory was slowed by the early over-emphasis on defense (two bowmen plus walls, which I didn't mention earlier). Then I settled into my normal rhythm, began gaining ground, and eventually pulled ahead like I usually do.

I don't want to deny the jolt of competing against metastasizing warmongering AI early on. I don't play on Immortal now because I had no desire to face the Deity bonuses earlier. But keeping in mind the relative difficulty of VEM vs vanilla levels, playing on King in VEM ought to provide Emperor-level vanilla difficulty. That's not a walk in the park in the early going. It's worth playing a few games on both levels and then making an informed comparison. It could be useful to Thal.