Accelerated production
- Thread starter Patoz
- Start date
ThinkTank
RL Addict
The AI gets the same bonus. This implies that it just stretches the difference. At chieftain, the human player has a 2:1 production advantage over the AI, with AP this becomes 4:2, which is the same ratio, but the absolute difference is larger. If we focus for the moment only on unit production, this compares to 2 archers fighting 1 spear vs. 4 archers fighting 2 spears.
Edit: calculating the odds, it turns out the 4 archers fighting 2 warriors have a better chance of winning compared to 2 archers beating 1 warrior.
If we assume veterans, the odds of an archer beating a warrior (out on the grassland) are about 80%-20%.
That means that 2 achers have a 96% probability of beating 1 warrior:
First archers wins: 80%
Second archer wins: 80% of 20% = 16%
Now 4 archers beating 2 warriors:
w-w : 80%x80% = 64%
w-l-w : 80%x20%x80% = 12.8%
w-l-l-w : 80%x20%x20%x80% = 2.56%
l-w-w : 20%x80%x80% = 12.8%
l-w-l-w : 2.56%
l-l-w-w : 2.56%
Total: 97.28%
2nd edit: corrected the text since the calculation is about archers vs. warriors, not spears.
Edit: calculating the odds, it turns out the 4 archers fighting 2 warriors have a better chance of winning compared to 2 archers beating 1 warrior.
If we assume veterans, the odds of an archer beating a warrior (out on the grassland) are about 80%-20%.
That means that 2 achers have a 96% probability of beating 1 warrior:
First archers wins: 80%
Second archer wins: 80% of 20% = 16%
Now 4 archers beating 2 warriors:
w-w : 80%x80% = 64%
w-l-w : 80%x20%x80% = 12.8%
w-l-l-w : 80%x20%x20%x80% = 2.56%
l-w-w : 20%x80%x80% = 12.8%
l-w-l-w : 2.56%
l-l-w-w : 2.56%
Total: 97.28%
2nd edit: corrected the text since the calculation is about archers vs. warriors, not spears.
Nergal
Prince
Its been posted in another thread but it gives the AI a huge advantage at higher levels. for exactly the reason given above ... SID is bad enugh .. SID with AP doesnt bear thinking about ..
ThinkTank
RL Addict
No, it will enhance the advantage of whoever already has the advantage. On regent, everything is even. Below regent the human player is in the advantage, and AP will only increase that. Above regent, the AI is in the advantage, and AP will increase that.
PaperBeetle
Emperor
Your broad assertion seems to be that if you have a proportional advantage in numbers, it is furthermore advantageous to you for everyone to have proportionately more units... an interesting idea. I'm not sure I'm convinced, but certainly I will run it past Offa.
However, your choice of archers and spears as an example is peculiar. An archer certainly does not have an 80% chance of beating a spear in the open. It has about a 45% chance. 80% is more like archer versus axe I think?
ThinkTank
RL Addict
Yes, the intuitive idea being that unfortunate events in the proportionally larger case are less destructive for the larger force.
You're right, it is archer against warrior. I was using BomberEscort's combat calculator with a defender value 1 with 4 hit points, and an attacker value 2 with 4 hitpoints, but I was thinking spear-archer.
Edit: I have corrected this in the original post.
TheOverseer714
Overseer
AP tends to make war more difficult because it magnifies the importance of the RNG. If you don't take an enemy city quickly, it can build a new defender every turn and in a few turns be a fortress. If you get a couple of bad RNG results, your attack could fail. Say you have 8 archers attacking a town with 4 spears. If one spear wins and promotes, then wins again, the whole tide changes. You only have enough if every other battle is a victory. AP can also lend itself to extreme unit costs, which can really cripple your economy. In my only major AP experience, in 3 of our 5 cities, we were getting tanks every turn. They were building up so fast we could hardly get them to the front fast enough to not hamper our economy. I would say AP is not a problem at Regent or below, but as you go higher it becomes an even bigger hindrance to quality play. In short, if you are thinking about using it...don't!
Pyrrhos
Vae Victis
- Joined
- Mar 15, 2007
- Messages
- 712
An example:
You want to build a marketplace in a city that produces 5 shields net and +3 food every turn. The Marketplace costs 100 shields under "normal production", which means that it takes 20 turns to build.
If you want to pop rush it (each citizen worth 20 shields and cannot sacrifice more than 50% of pop), you'd usually have to produce at least 40 of those shields, which takes 8 turns, before sacrificing three pop - town must be at least pop 6. To replace those three pop at +3 food every turn would take another 10 turns - provided you have a granary. Total cost 18 turns.
Under accelerated production, the shield cost for the marketplace is 50 which means ten turns to build. Pop-rushing, you'd only have to produce ten shields (two turns) before sacrificing two pop (= 40 shields) of your pop 4 town. To replace each pop lost with a granary at +3 food per turn would take only five turns as you only need half the amount of food under accelerated production. Sum seven turns.
Without pop-rushing, accelerated production is twice as fast. With pop-rushing it's four times as fast if you only count the time until the marketplace is built, 2½ times as fast if you include the time it takes to replace the pop.
As far as I am aware, the AI doesn't pop-rush which means that accprod is a great advantage to the human player.
You want to build a marketplace in a city that produces 5 shields net and +3 food every turn. The Marketplace costs 100 shields under "normal production", which means that it takes 20 turns to build.
If you want to pop rush it (each citizen worth 20 shields and cannot sacrifice more than 50% of pop), you'd usually have to produce at least 40 of those shields, which takes 8 turns, before sacrificing three pop - town must be at least pop 6. To replace those three pop at +3 food every turn would take another 10 turns - provided you have a granary. Total cost 18 turns.
Under accelerated production, the shield cost for the marketplace is 50 which means ten turns to build. Pop-rushing, you'd only have to produce ten shields (two turns) before sacrificing two pop (= 40 shields) of your pop 4 town. To replace each pop lost with a granary at +3 food per turn would take only five turns as you only need half the amount of food under accelerated production. Sum seven turns.
Without pop-rushing, accelerated production is twice as fast. With pop-rushing it's four times as fast if you only count the time until the marketplace is built, 2½ times as fast if you include the time it takes to replace the pop.
As far as I am aware, the AI doesn't pop-rush which means that accprod is a great advantage to the human player.
Nergal
Prince
I've seen the AI pop rush. Especially if it goes to Communism or Fascism. Which are its favoured war govs. Its bad enough on Monarch without giving it an advantage numerically.
Sweetchuck
King
- Joined
- Nov 26, 2006
- Messages
- 649
I always use AP.
Not because (or in spite) of the aforementioned, I just hate waiting a billion turns to build a temple.
It was once suggested that workers won't able to keep up with cities under AP, but I just build more workers proportionately. Good for commerce.
Not because (or in spite) of the aforementioned, I just hate waiting a billion turns to build a temple.
It was once suggested that workers won't able to keep up with cities under AP, but I just build more workers proportionately. Good for commerce.
ThinkTank
RL Addict
That's interesting, my intuition is exactly the opposite, that it reduces the importance of the RNG, in case of combat, as my example in post 4 shows.
However, I now see that my calculation was a bit too simple: I neglected possible promotions. I also negelected the fact that a loss may still do some hitpoints damage so that the odds for a subsequent wins become better.
ThinkTank
RL Addict
Chieftain is already so easy that some people think you should not play it because you will become spoiled, that is, used to strategies that will never work on higher levels. Sid is way too difficult for most us round here. AP will make chieftain even easier and sid even harder. So if it's about difficulty, there's no need to use AP.
MAS
Deity
The game was originally intended and designed without such a thing as AP.
AP was added as an after though in an attempt to make multi-player games quicker. (but I wonder if that was a good idea, even for MP games)
Being a bit of a purist, I would like to suggest not.
The game was originally intended and designed without such a thing as AP.
AP was added as an after though in an attempt to make multi-player games quicker. (but I wonder if that was a good idea, even for MP games)
Being a bit of a purist, I would like to suggest not.
How come it was introduced for Multiplayer, but is present in Vanilla?
PaperBeetle
Emperor
Yes, that was the omission which worried me. The effect does it make tough to perform straightforward probability calculations about combats with multiple units - the more so the more units you pile on.
Nevertheless, your intuition seems to be borne out. I used Offa running at 100000 trials to get the following win rates (vet archers attacking reg spears, unfortified on the open plains):
2 archers vs 1 spears: 91.2%
4 archers vs 2 spears: 91.0%
6 archers vs 3 spears: 92.2%
8 archers vs 4 spears: 93.5%
10 archers vs 5 spears: 94.6%
12 archers vs 6 spears: 95.4%
14 archers vs 7 spears: 96.3%
MAS
Deity
More accurately, it is present in single player.
They patched up vanilla when they released the XP's, and besides fixing bugs and improving compatibility with the XP's, they also added some XP features to the vanilla version for free. (isn't Firaxis a nice company?
)For example, the earlier vanilla versions also didn't have stack movement.
