Barracks during axe rush?

[ben-jammin]

I'm not in the same league as most of you guys so this may be a moot point.

Axe rushes come early...and generally the capital and second city tend to be decent producers throughout the game. It makes sense to build a barracks in a city like this for future wars. The question really seems to be more about timing and whether you build the barracks before the axe rush or after during the recovery phase.

If I have reason to think the combat odds will be this narrow and the margin of victory so small I shouldn't be going to war.


I think the ideal answer depends more on your leader traits (how difficult can recovery get) the target, surrounding AI's, space and quality of land available, etc,.


As a general rule of thumb I like Dirk1302's mention of 2pop whipping. Hell if we are gonna do it, lets build them early and use our overflow hammers. It gets hard to work the overflow magic when building only units (as you do in war). This way seems to earn two birds with one stone.


Macro generally trumps Micro right?

I think thats maybe the best point yet. If the 2 extra axes you'd have gotten by not building a barracks will make or break your rush, is it really a good idea? In that case all it takes is an unfriendly RNG for 2 turns of combat to completely screw you.

In any case, the lower your odds of success the better your odds of dying without even doing damage (if im not mistaken...this seems to happen to me frequently at ridiculously low suicide odds). That's the worst possible scenario and IMO even CR1 makes it less likely.

There's also the fact that you're going to build barracks in a certain number of cities anyway. It's not like we're talking about a hammer investment you'd otherwise not be making at all. As was mentioned above you can also start building your barracks just before bronze gets hooked up and then 2 pop whip it into your first axe. I don't think anybody who doesnt like to argue just for the sake of arguing would deny that's an ideal strategy in any city that's going to produce more than a handful of axes.

@Dirk I was just surprised you said you build catapults in the cities where you don't build stables, because I assumed HA rushes were done in a way that catapults can't keep up. I recall watching one of AbsoluteZero's videos where it was nothing but numidians, but i suppose the melee bonus could change things significantly at immortal and deity where you see mostly metal defenders.

 

[Dirk1302]

HA rushes are done without cats,i'd build stables in every harcher producing city if i were to rush with them. But i was talking about a phant/cat combi and then building stables everywhere is not necessary.

 

[TeraHammer]

With AGG: almost always (unless time of the utmost essence)
With Non-AGG: hm. I wouldnt for a rush. For a longterm military production city power, yes, but not for a quick rush.

 

[pigswill]

Once you've decided on a rush (nearby copper, nearby neighbour) you're likely to have a few turns between discovering BW and getting copper hooked up so why not use the turns building a barracks? You can't build axes until copper is connected.

 

[TeraHammer]

how do you know there is nearby copper if you havent bronze working yet?

 

[vicawoo]

Responses to most of the issues:

Barracks are counterproductive on noble and below, since an unpromoted axeman is faster to build and deals with warriors. Kind of obvious, but worth remembering for maximizing score, and two unpromoted axes instead of a barracks means 2 more capitals.

More unpromoted axemen is marginally better as long as you don't make too many. Your capital will tend to have more trees/hills and produce more axemen, so usually you want one in your capital.
With a close city, you can allocate most of your chops to your capital, which has the barracks, and slow build unpromoted axes from your second city.

However, most cases aren't "ideal". Your second city will have to build something while it grows, and since your copper may not be hooked up, that often means a barracks.

2 pop whipping a barracks is not a good idea, unless you have massive amounts of food. With hills, this hurts you without a granary in the longer run (~10 turns, probably less than that) and since you'll be building most of your axes after the barracks, you won't have time to "recover" the loss of hammers.
If you axe rush, you want to maximize your hammers, so you don't have time to build a granary, and the only time you want to whip is the last cycle or two.

The combat odds for axe rushes isn't highly mathematical or very complicated. Culture goes up by 20%. City raider 1/city garrison 1 go up by 20%. If you have much experience, you realize that most of the time you only have to deal with 3 or 4 values (assuming fully fortified archers, which have a base of +75%): -20%, 0, +20%, +40%. Hills/rivers are +50% and combat 1 is different from cr1, but the odds don't change drastically enough that you can't just round.

Finally, you'll probably attack first with one of your many 1 promotion axes and use the unpromoted to kill the damaged ones, and you only need to worry about breakpoints for the former.

 

[2metraninja]

how do you know there is nearby copper if you havent bronze working yet?

Good question.

And yes - rush means to hurry. When hurrying, no time for building barracks unless aggressive. Either for the discount, and for the second promotion.

 

[Dirk1302]

2 pop whipping a barracks is not a good idea, unless you have massive amounts of food. With hills, this hurts you without a granary in the longer run (~10 turns, probably less than that)

Depends on what massive food is. Second city will usually have cap 4 without a happy resource. Grow to 5 and 2 pop whip, you'll get 60 H immed and lose at most 40H from a plain hill that can't be worked. Net gain 20H unless city can't get back to 4 in 10 turns because it lacks the massive food you're talking about . So 2 pop whip is just a winner in those cases, if you want to whip a barracks with it is second....

Finally, you'll probably attack first with one of your many 1 promotion axes and use the unpromoted to kill the damaged ones, and you only need to worry about breakpoints for the former.

Sure you do but we weren't going to take one city, that unpromoted axe that's winning now is still a loser at 1/2 or 2/2 xp. We're not going to take one city but 4-5 preferably and we will want our axes to get to cr2 with their first or second win.

 

[Dhoomstriker]

how do you know there is nearby copper if you havent bronze working yet?

You have already discovered Animal Husbandry and there is an Unforested Grassland square in your capital's fat cross which is surrounded by Forest squares. It's still a guess, but it's a reasonable one.

Another option is that you might assume that you'll have Copper and will aim to have a Settler to settle near it once Bronze Working arrives, should it not appear within your fat cross. If there is no Copper within settling distance, then either start a new game or change your plans--whichever option is to your preference.

 

[huerfanista]

Once you've decided on a rush (nearby copper, nearby neighbour) you're likely to have a few turns between discovering BW and getting copper hooked up so why not use the turns building a barracks? You can't build axes until copper is connected.

how do you know there is nearby copper if you havent bronze working yet?

Maybe try some reading comprehension? "between discovering BW and getting copper hooked up" means after you've researched BW and before you settled a city to claim it and/or built the mine and roaded to it. [EDIT] I assume you were responding to pigswill's post. If not, just ignore what I wrote.

 

[ben-jammin]

Responses to most of the issues:

Barracks are counterproductive on noble and below, since an unpromoted axeman is faster to build and deals with warriors. Kind of obvious, but worth remembering for maximizing score, and two unpromoted axes instead of a barracks means 2 more capitals.

More unpromoted axemen is marginally better as long as you don't make too many. Your capital will tend to have more trees/hills and produce more axemen, so usually you want one in your capital.
With a close city, you can allocate most of your chops to your capital, which has the barracks, and slow build unpromoted axes from your second city.

However, most cases aren't "ideal". Your second city will have to build something while it grows, and since your copper may not be hooked up, that often means a barracks.

2 pop whipping a barracks is not a good idea, unless you have massive amounts of food. With hills, this hurts you without a granary in the longer run (~10 turns, probably less than that) and since you'll be building most of your axes after the barracks, you won't have time to "recover" the loss of hammers.
If you axe rush, you want to maximize your hammers, so you don't have time to build a granary, and the only time you want to whip is the last cycle or two.

The combat odds for axe rushes isn't highly mathematical or very complicated. Culture goes up by 20%. City raider 1/city garrison 1 go up by 20%. If you have much experience, you realize that most of the time you only have to deal with 3 or 4 values (assuming fully fortified archers, which have a base of +75%): -20%, 0, +20%, +40%. Hills/rivers are +50% and combat 1 is different from cr1, but the odds don't change drastically enough that you can't just round.

Finally, you'll probably attack first with one of your many 1 promotion axes and use the unpromoted to kill the damaged ones, and you only need to worry about breakpoints for the former.

2 pop whipping is never bad if the alternative is working tiles just to stay below the happy cap.

 

[pigswill]

Something else to be considered is survival rate. Building a barracks is a one off cost which benefits every axe built in that city. Promoted axes will have a marginally higher chance of winning (and therefore surviving). Every axe that survives is one less replacement to be built.
I'm not mathematical enough to figure out the model but I'm sure its doable, and there will be a break even point when a barracks saves hammers by building fewer replacement axes (once you've saved two axes you're ahead on hammers). You can also save turns by not having to wait for reinforcements to arrive from your own cities and promoted axes are likely to take less damage so need less time to heal.

editvicawoo. It could be argued that unpromoted axes are counterproductive below noble because you can warrior rush earlier.

 

[vicawoo]

Depends on what massive food is. Second city will usually have cap 4 without a happy resource. Grow to 5 and 2 pop whip, you'll get 60 H immed and lose at most 40H from a plain hill that can't be worked. Net gain 20H unless city can't get back to 4 in 10 turns because it lacks the massive food you're talking about . So 2 pop whip is just a winner in those cases, if you want to whip a barracks with it is second....

Either you regrow in 10+ turns or you don't.

If you do regrow, you also need 26+28=54 food, and of course you'll either have to wait 10 turns to get from size 4 to 5 or work an unhappy tile.

In addition, since you don't have a "massive" food surplus, as in you can convert all your excess food to hammers, you used 54 food to grow which could have been converted to at least 54 hammers. If you grew faster using forested grassland tiles, that in fact cost you 27 grassland tiles turns, or 27 x (4-1) = 81 hammers.

If you choose not to regrow, you have to compare it to saving your whip for your last 2 axes. In that case, you're not gaining 60 hammers and losing 4 hammers per turn for 10 turns, you're just losing 4 hammers per turn since you're going to get your 60 hammers, just later.

Finally there's the option of not growing, since growing is sometimes an opportunity cost (unless you have an unavoidable food surplus, hence massive food).
If your second city is size 2 working the a plains cow or grassland copper mine tile with both possible hills and grassland forests, no farms, say you want to rush 10 turns from now, what gives you more hammers, growing to size 3, trying to grow to size 4 then 2 pop whipping, or working a mine?

Copper mine example, since we can avoid excess food
Mine: straight forward, +4 hammers for 10 turns, 40 hammers.
Size 3: grassland forest for 12 turns, can't even whip, so that case is pretty done

Plains cow
Size 2: +4 hammers for 10 turns, 40 hammers, we can whip on the 10th turn for +30 hammers - 4 hammers. Net: 66 hammers
Size 3: Grassland forest for 8 turns (+8), grassland mine and plains mine for 1 turn (7), plains mine for 1 turn and whip (+34). Net: 49 hammers
Size 4: Grassland forest(s) for 17 turns.

To avoid having to manually labor through every case, let's look at the broader picture: a forested grassland tile produces either +1 food at the cost of 2 hammers or +2 food at the cost of 3 hammers, compared to a mine. So a net -1 yield, if we can assume we can reorganize the rest of our tiles to neutralize our food.
Each turn earlier we grow, we gain another tile, which is +1 yield for a forest, +2 for a mine.
Whipping produces 30 hammers for 2x(10+n) food.

If we can swap food for hammers until we are food neutral, that means we gain are sacrificing 20 + 2n food = 10+n grassland forest tile turns.
And we gain m - grow time extra tile turns, where m is the number of turns we want to rush.
So we lose 10+n hammers from working forested tiles instead of mines to grow.
You gain 2 x extra turns.
If
2 x extra turns + 30 + (10+n) > 4 x (10+n) for plains hills or
6 x (10+n) for 2 grassland hills (unlikely if you have a food source and aren't charismatic).

So 2 x extra turns > 3 n, then it is worthwhile

Spoiler calculation for growing with mines vs forests given unlimited time :

If we can swap food for hammers sufficiently on a 1:1 basis, we compare faster growing time, which is
(20+2n)/(f-1) - (20+2n)/f = (20+2n)/(fx(f-1)), where f is the food surplus with the forested tile
multiply that by the additional tile we can work at size n+1, so
4 x (10+n)/(f x (f-1) )
and subtract (turns to grow with a grassland forest) x the yield lost working grassland forest
or (20+2n) x 1.
So if (20 + 2n ) x (2/(f*(f-1)) - 1), then there's a net gain.

 

[vicawoo]

Here we consider grow to 5 then whip in the long run. I'm sure this has been done before.

So the long term plan, evidently, is to grow to 5, 2 pop whip, then grow to 4 in 10 turns, 2 pop whip, grow to 3, then grow to 4 within 10 turns of anger, repeat last 3 steps or until we go to war.

First we consider growing to 5 after 4, compared to running a hill. Without a massive food surplus, that's 28 food or 14 grassland forest turns = 14 x (4 - 1) = 42 extra hammers from the mine. So we gain 60 hammers from the whip, but we lose 42 hammers getting there, then lose 4 hammers per turn afterwards. So in reality, we fall behind after 5 turns.

Next we consider the 2 pop whip, from size 5 to 3. We need 26 food to grow. If we're running grassland forested mines, we're only generating F - 2 food per turn, so we generate 10 x (F-2) food per turn. Each additional food costs 2 hammers (grassland mine->forest). We lose even more food if we have to run plains mines instead of grassland mines.
So with a 4 food source, we only generate 20 food, so we have to pay 12 hammers for the other 6 food. We lose a plains mine on top of that, assuming our 4th tile is a forested mine.
So we lose 40 + 12 = 54 hammers to gain 60 hammers. This is not including our handwavy free growth to size 5, which as above would have put us at 60 hammers whip for 96 hammers.

If we have a 5 food source, we generate 30 food, 4 of which we convert back to hammers by working a plains mine. So we lose 40 hammers, but get back 4 from our surplus. But we still have to pay for our ridiculous size 4 to 5 growth. 36 + hammers lost growing from 4 to 5.

So it clearly is terrible without excess food. How about a normal size 4 to 2 pop whip?
Ok, we have to regrow 24+26=50 food in hopefully 10 turns (hopefully we have a 5 food tile). We have to run grassland forested tiles the whole time, so we're going to get about 15 hammers from our forested tiles. We lose 30 mine tile turns, in this case 2 plains mine and 1 grassland for 10 turns each, so 110 hammers. We can squeeze out a few extra hammers if we can get grassland farms to grow faster and possibly squeeze in a few extra mine turns.
But 60 + slightly more than 15 < 110.
So a non-handwavy 2 pop whipping without a granary is horrible over 10 turns. You are in fact hemorrhaging around 30 hammers over those 10 turns. A handwavy 2 pop whip looks good, but it still stinks if you don't ignore the growing cost.

In the end, it's pretty simple: 60 hammers for 50 food + 10 lost tile turns + inefficient regrowth tiles is a bad deal over 10 turns.

 

[cabert]

editvicawoo. It could be argued that unpromoted axes are counterproductive below noble because you can warrior rush earlier.

my point exactly

If I'm using "real" troops = chariots, axes, anything better than warriors, I'll build barracks in my first city and possibly (= if there is time to get 2 or more units out of the city after the barracks) in any city after that.

 

[VoiceOfUnreason]

I posted a calculator, and some of the conclusions that I drew from it, over in the Strategy Articles section.

You still have to decide what your success criteria is. But considering 95% odds of capturing a city in one turn, and using Ben-jammin's 115% defense threshold, the data suggests that a barracks isn't worth it.

Yeah, sometimes you need an extra unpromoted Axe, but I didn't see any cases where you need two....

Of course, I may have oversimplified the scenario by considering a single city in isolation, rather than a campaign. And I'm drawing conclusions before anybody else has had a chance to review it for bugs....

 

[ben-jammin]

I posted a calculator, and some of the conclusions that I drew from it, over in the Strategy Articles section.

You still have to decide what your success criteria is. But considering 95% odds of capturing a city in one turn, and using Ben-jammin's 115% defense threshold, the data suggests that a barracks isn't worth it.

Yeah, sometimes you need an extra unpromoted Axe, but I didn't see any cases where you need two....

Of course, I may have oversimplified the scenario by considering a single city in isolation, rather than a campaign. And I'm drawing conclusions before anybody else has had a chance to review it for bugs....

second, third cities would swing it in favour of barracks bigtime, especially when you take into account further promotions and higher defense multipliers.

 

[fyermind]

I may be out of my league here, but I think the deciding factor is how many cities you plan to rush and how many axes you want left. If all you want is one axe left and to capture one city, (normal speed on a fairly small map stealing a capital) and then sue for peace, especially if you are willing to bully your weakest neighbor to do it, a barracks is a waste of time that could be two more sacrificed axes.

If you want to eliminate a potential threat (monty is your neighbor on a marathon game with agg ai? you want him gone before he gets iron working, that means all three of his cities) and maybe continue your campaign to a second civ (don't know if this works above monarch) a barracks looks a lot better.

My rule of thumb is if I am attacking more than five archers total, or if I am continuing my rush to a third city, I build a barracks while I hook up my copper (and research the wheel).

 

[obsolete]

I don't do axe rushes, on deity it's usually too late, unless you have something else going for you. Definitely a no-no if you're not agg without something going for you, though sometimes you can pull off interesting gambits (so I'm not going to say it doesn't work!). Everything has its time and place.

An AGG leader of the Incas is a whole new story... This means Barracks first, then WHAM-BAMB you can knock out a whole AI before you could even get any of your copper set up. It's the best feeling when you find out later that you didn't have any possibility of copper or iron in the first place anyway.

 

[TheMeInTeam]

I don't do axe rushes, on deity it's usually too late, unless you have something else going for you. Definitely a no-no if you're not agg without something going for you, though sometimes you can pull off interesting gambits (so I'm not going to say it doesn't work!). Everything has its time and place.

An AGG leader of the Incas is a whole new story... This means Barracks first, then WHAM-BAMB you can knock out a whole AI before you could even get any of your copper set up. It's the best feeling when you find out later that you didn't have any possibility of copper or iron in the first place anyway.

Quecha starts with combat I regardless of AGG. HC's UU is as good as any AGG version of it.