GotW: 2700BC Chieftain - How to (7/13/08)

[Sigmakan]

As I promised. A few disclaimers first:
-I apologize to anyone who tries to solve this on their own. It is possible to get a better time than 2700BC (you can get 2800BC) so you still have a chance to best this how-to
-I can't promise that this works on xbox. I did this on PS3, I don't knonw if the random seed is 100% identical, I would assume it is.
-If you don't follow the guide 100% I cant guarantee a win. The slightest changes will change the odds. For example:
Scenario A
You switch from galley production to cossack production ASAP. later this game your warrior, when attacking barbs loses 1 guy
Scenario B
You switch to cossack production a turn later after you could've done it. later this game your warrior, when attacking barbs, doesn't lose any health.
-Knowing this strat will not necessarily help with next weeks maps. There are some common tricks, but all of those have been discussed on the forum elsewhere. Its about executing those tricks and analyzing the map for the optimal route
-If there is a typo in here let me know, especially if it causes you not to win

Here it is:

4000 BC
• Set research to HBR
• Set production to galley
• Set workers to all hammers
• Move WarriorA 1S
3900 BC
• Move WarriorA 1S
3800 BC
• Move WarriorA 1SE to pop hut (gives HBR)
• DO NOT SELECT ANOTHER TECH TO RESEARCH, just back out with cancel button. Never select a research from here on out
• Switch production to Cossack
3700 BC
• MUST DO IN THIS ORDER
• Switch workers in Moscow to all gold. They will stay on gold for remainder of game
• Move WarriorA SE (gets landmark)
• Rush CossackA
• Move CossackA 2E
3600 BC
• Move CossackA 2E
• Move WarriorA 1NE
3500 BC
• Warrior attack barbs to east
• Move Cossack 2NE to pop hut (gives SettlersA)
• Move SettlersA 1SE (gets landmark) and 1E
3400 BC
• Warrior attack barbs to east to earn SettlersB
• Move SettlersB 1E and 1NE
• Move SettlersA 2SE
• Move CossackA 1SE 1E
3300 BC
• Move Cossack A 2SE
• Move SettlersA 1E 1SE to pop hut (gives 25g)
• Move SettlersB 2E
• Sell WarriorA
3200 BC
• Move SettlersB 1SE 1SW
• Move CossackA 2E
• Move SettlersA 1SE 1E
3100 BC
• Move CossackA 1NE (gets landmark) 1N
• Move SettlersA 2SE
• Move SettlersB 1SW 1S
3000 BC
• Move CossackA 2W to attack China and win
• Set Beijing workers to all gold
• Move SettlersA 1E and settle CityA
• Set workers to all gold in CityA
• Rush CossackB from CityA
• Move CossackB 2SE
• Move SettlersB 1SE and settle CityB
• Set workers to all gold in CityB
• Rush CossackC from CityB
• Move CossackC 1SE and 1E to occupy unoccupied Kyoto
• Set workers to all gold in Kyoto
2900 BC
• Sell CossackA and CossackC
• Rush CossackD from CityA
• Move CossackD 1NE 1E to attack America and win
• Move CossackB 2S
2800 BC
• Move CossackB 2SE to attack Arabia and win
• All Palaces captured
2700 BC
• Win.

As I said, 2800BC is possible, I've done it a few times (didnt register on leaderboards though ). It does require a fairly significant change of above strat. I challenge you to figure it out :P

[Axewielderx]

I won this,not sure what year though and pretty sure si nce Iam on the ps3 my score never posted.I have not been able to find any scores on the gotw.

[kylew]

Awesome - thanks for sharing. It never occurred to me to move my settlers that much. Thats some great new strategery to try out . My current score of 2400 was the result of settling the first group right away and selling the two others I got.

[Jamie98s]

who cares about chieftan

[MaxDaddy1]

Well, I went to bed 7th or 8th last night at 2600 BC and woke up with a bunch of new 2700s ahead of me... I wonder how that happened...

Props to Sigmakan for taking the time to put together the write up. The only obvious difference in my 2600 was that I set something to research... and I ended up losing the first battle for Washington.

[MaxDaddy1]

Sigmakan - So this question is for you, or really anyone else who is in the same situation. Your strategy worked on the Xbox 360 version but I didn't save it because it wasn't my strategy. I'll stick with my 2600 BC that I did on my own. When I compare what I did with your moves, I selected only four or five different moves. We both settled cities in the exact same location, both took Japan without a fight, both rushed Cossacks to huts to spawn settlers to rush, etc.

My question is this... and maybe even Elizabeth could chime in on this...

Why does approaching a city from a different spot result in a completely different outcome? For example, Sigmakan approaches China from the East in 3000 BC and wins. I approached China from the South in 3000 BC and lost, my 4 attack Cossacks losing 5 straight times to 2.5 defense Warriors (I kept trying the same thing over and over again thinking that eventually my Cossacks had to win at least once based on the idea that the outcome was random, not fixed), the exact same fight when approached from the East results in the opposite outcome. Why?

I guess my main point (and really the only strong criticism I have of a game that I truly enjoy) is that the game (even the GOTW) should be random enough to allow players with similar strategies to have similar results most of the time, whereas the way the game plays now, you only enjoy success if you approach the "right" city from the "right" location with the "right" unit during the "right" year. You shouldn't have to follow 38 steps in a row to ensure that a battle between a Off +4 and a Def +2.5 results in a victory for the unit whose strength is close to double that of the other unit.

It makes the game much more like a jigsaw puzzle with fixed moves at fixed times as opposed to a true strategy game where similar (but not identical) tactics consistently result in similar outcomes.

[ThreeE]

MaxDaddy1,

I suspect that in order to ensure that the GoTW is played the same for everyone, the game seeds the random number generator with the same seed for everyone. Thus, the same random roll occurs if all the conditions are the same for everyone.

This also makes sure that you can't just play the game over and over with the strategy waiting for the time when you "got lucky" five times in a row and took down a large defensive unit with a low offensive unit.

Now if this was occuring on the non-GoTW games it would suck.

ThreeE

[MaxDaddy1]

But it's ridiculous that the exact same offensive units attacking the exact same defense units produce opposite results based only on a variable of following a certain sequence of events. In all of these cases, there was no strategic advantage (i.e attacking from a hill), it's purely a matter of attacking in a certain order or attacking from one identical square from the South instead of an identical square from the Southeast. It's equally ridiculous that 2.5 beats 4 every time, unless you move one square left or right. The game no longer is a strategy game at that point, it's an exercise of repetition.

[ThreeE]

It would be ridiculous IF it wasn't GoTW. If it was as you described, then all you would have to do is play the game over and over again until your first Warrior unit rolled a successful attack against much stronger defensive units every time.

If the random outcomes differed at all, then there would be no point in the GoTW "competition" for the earliest date. In a nutshell, GoTW IS a jigsaw puzzle. It would seem that you are interested in one of the other game types.

Again, if the other game types did this, it would suck.

[MaxDaddy1]

It would be ridiculous IF it wasn't GoTW. If it was as you described, then all you would have to do is play the game over and over again until your first Warrior unit rolled a successful attack against much stronger defensive units every time.

If the random outcomes differed at all, then there would be no point in the GoTW "competition" for the earliest date. In a nutshell, GoTW IS a jigsaw puzzle. It would seem that you are interested in one of the other game types.

Again, if the other game types did this, it would suck.

Wow, you might try actually reading what I write, it will help you not look like an idiot. My Cossack, attack strength of 4 is losing to the AI Warrior, defense of 2.5... now since you seemed confused, and just to cover the basics... 4 is LARGER than 2.5...

If the Cossack (ATK of 4) attacks from the south, he loses to a 2.5 defense. If the Cossack attacks from the east, he beats the 2.5. Hopefully I typed this slowly enough for you to understand the problem. If not, why don't you go pester someone else with inane comments made without reading what the adults are discussing.

[ThreeE]

A 4 attack does occasionally beat a 2.5 defense. It's called random. Look up what a random seed is if you don't understand.

My replies to you are over now as you seem to be more interested in ad hominem attacks than rational discussion.

[Krikkitone]

Wow, you might try actually reading what I write, it will help you not look like an idiot. My Cossack, attack strength of 4 is losing to the AI Warrior, defense of 2.5... now since you seemed confused, and just to cover the basics... 4 is LARGER than 2.5...

If the Cossack (ATK of 4) attacks from the south, he loses to a 2.5 defense. If the Cossack attacks from the east, he beats the 2.5. Hopefully I typed this slowly enough for you to understand the problem. If not, why don't you go pester someone else with inane comments made without reading what the adults are discussing.

To cover the basics
Winning=CHANCE
Losing=CHANCE

Chance is based on the Random Seed and that depends on which actions you took

if you decided to do ANYTHING different (not just that attack, but even naming a land mark different, or putting a city on food for one turn or starting to research something else.) then the Random # is different

a 4 can lose to a 2.5 if the random number is right ( a 17 can lose to a 2.5 if the random number is right) so because of Something you did, you had a different number by the time you attacked than the OP did.

[kylew]

I believe the random seed is the best design decision. At first glance it may seem unfair for your unrelated actions to affect the battle outcome. But the alternative is worse - having a completely random battle each time. It would result in simply a brute force strategy, rolling enough times to get the good odds. I'd say thats even less "strategic" than the current model. My 2 cents.

[ByeByeToYou]

i will never be top again sob sob.lol

[xbman22x]

for those of you that dont know, it does do this in single player. if you save before you attack someone. then reload and attack them again and the out come will be the same. if you attack them from a different direction or do anything different then the outcome could (but wont nessisarilly) change. sorry to burst your bubble.

[ThreeE]

for those of you that dont know, it does do this in single player. if you save before you attack someone. then reload and attack them again and the out come will be the same. if you attack them from a different direction or do anything different then the outcome could (but wont nessisarilly) change. sorry to burst your bubble.

Good info, but the real test would be if you restart (not save and load) the entire game and enter the same battle. Of course, you can't do this as the game (map, opponents, etc.) is randomly generated.

Again, this is a good thing, because you could just save before every attack and wait till you get the outcome you want. Diety, for example, would be simply a grindfest away.

By the way, if you want to cheat this way (save and reload if losing), just save earlier. There are usually several different paths to attacking a defender of equal distance. Just save at a point where there are as many paths as you want chances and have a go at it.

The only thing that matters is that you don't know the random outcome of any attack before you commit the first time. Kind of like a Civ version of the Heisenburg Uncertainty Principle.

[xbman22x]

just remember... the slightest change will change the seed and could change the outcome. for example. if you go NE then NW then attack you could get something different than if you went NW then NE then attack. so you can still easilly manipulate this.

[ThreeE]

I suppose the game engine uses a hash on a concatenation of all the game state variables for the random seed. Pretty good idea actually. It allows you to explore different options, but not game the random generator.

[Sigmakan]

After watching the CivRev Behind the Scenes video from the PSN I'm convinced that the gotw was not designed to be like this.

I think there is a fairly easy fix that might make everyone happy (or no one happy lol). Firstly, as many have state there should probably be a way to filter results to you can sort by victory. Dont need separate leaderboards, just need a way so that the leaderboard can be filtered so it only shows one sort of victory. Next, goodie huts should be removed from all difficulties besides Chieftain. This would make the other difficulties even more challenging which is quite appropriate for game of the week This would also keep chieftain the way it is. So for people like myself who enjoy the 'puzzle' of the week, we would still be able to do it.

Just a thought!

[Remowilliams]

Firstly, as many have state there should probably be a way to filter results to you can sort by victory. Dont need separate leaderboards, just need a way so that the leaderboard can be filtered so it only shows one sort of victory.

This doesn't work. The reason is that the leaderboard only keeps one score. So, if I decide to go for space race, I can't go for domination, because, obviously, my domination victory would happen faster than the space ship, so it would knock my space score off the leaderboard.

If they do it your way, they need to also save 1 victory of each type, on each board, and not just your fastest time.

The other thing is that if you just shoehorn the victory sort into GOTW, no one will be competing for the other victory types. By making separate boards, you encourage competition.

[Sigmakan]

Good point. At one point I had thought through it all and came to that conclusion, but I forgot lol.

[Thrallia]

I like the idea of no barbs/huts on difficulties above chieftain in the GOTW though.

And Sigmakan, thanks for the writeup...I didn't get to try it before the GOTW changed, but seeing a step-by-step description from start to finish of one of those super-fast finishes sets my mind more at ease about whether or not there's cheating going on in the GOTW as well as the ranked MP ladder.