G-III Upsilon

[Marsden]

Hello and welcome to the latest Gauntlet for the Civ III HOF. This month, with holidays coming up for most of us, we're going to turn down the difficulty all the way. The size will be up all the way, so it isn't over too quick.

• Mapsize: Huge
• Difficulty: Chieftain
• Condition: Conquest

Good luck and first one to smack them all gets the gold! All submissions must be in by 12/17/07.

 

[Elear]

Figures

 

[AutomatedTeller]

huge chieftain conquest - jags are awesome, but can run into a problem late with spears showing up in your few cities. I suppose the war chariots would solve that, but they aren't as flexible, they are more expensive and they need roads to get through the jungle belts in the middle of the continent.

It's a pretty fast game, though.

I'll try to write up my 1050 BC game - it was basically "build a barracks and build jags and kill"

 

[Niklas]

Hmm, not a big fan of games where you don't know until a good deal of turns into the game if it's at all worth continuing. I tried one of these a while back, I think it was Huge Warlord, Conquest was the goal anyway. Was off to a really good start, with two settlers popped from huts, only to discover a good deal later than one AI was off on a tiny continent all of its own.

AT, you mean stick to one town and just go on Jag spree? I guess to conquer the whole map in 78 turns like you did, there's no other alternative. Might make for a fun, different game.

Btw, why Chieftan and not Warlord? That date looks infinately more improvable...

 

[AutomatedTeller]

I think that game took all of an hour...

I may have built another city, I'm not sure.

and I had the various other civs.

I think it's improveable - the big challenges are:

finding the AI's
dealing with spears.

 

[Lord Emsworth]

Are spears really that much of a problem? I mean if you make sure that none of the other civs is scientific and starts with bronze working you ought to see more warriors.

 

[Elear]

They can be...

It really depends, here's what I have found so far:

Most capitals will be defended by 1 warrior. To be safe, take two jags. Take it over and wipe out the civ. If they have a second warrior after the first, retreat, rest, attack again.

If they develop spears, you need at least 4 jags for every spear. A jag can take anywhere from 0-2 hp off a spear, but 1 is fairly common. The other problem is, when your jags die, sometimes they will promote the enemy, and that just happens. Elite spears are not uncommon in mass jag sieges near the end.

In my attempts, I see about 5 of the 8 civs have warriors. Ideally 6 or 7 would be better, but it's all luck there.

It's not really a huge issue, so to speak, just very annoying when going for a 1000 BC or quicker date.

The main problem is something that happens on all difficulty levels: AIs will panic and go max to bronze working, then whip spears.

Occasionally, a Chieftain AI will self research BW or trade with a neighbor who researched it.

Above all, a civ stuck on one of those 'pangaea islands', is what will kill your attempts.

 

[Marsden]

Figures

Sorry. I promise something harder in the future.

 

[AutomatedTeller]

It is very likely that at least one civ will end up with spears by 1000 BC or so. Particularly if you end up with a civ a long ways from you that have a decent start - they might poprush some spears. and if you get unlucky...

I think my 1050 BC game, I attacked an Indian spear with 6 jags and lost at the end - that's just bad luck, but it took me quite a few turns to build up forces.

 

[Chamnix]

This map was not exactly what I was looking for...

 

[Niklas]

I had one of those too, a really great start too with plenty of grassland cows around, better than any other map I've generated. Maybe there's a built-in handicap for gilligan starts. Though yours doesn't seem too great in that regard...

 

[Elear]

Frustrating.

My first real try would have ended in 600 BC (plenty of troops scattered around) but the 8th AI was on some tiny island in the middle of nowhere.

My second real try would have ended in 1200-1300 BC (over 50 jags just rampaging the countryside). Unfortunately, guess what? Yep, another AI stuck on an island. Other problem to the game: the only two AI who ever made spears were India (only 1) and Carthage. Yes, Carthage, so Numidian Mercenaries of all things. Otherwise the date could have been as fast as 1400 BC.

Don't you just love this?

 

[Marsden]

If you choose pangea with 80% water, odds* are that's going to happen.


* odds based on personal experience and not on any scientific facts.

 

[Elear]

Would not pangaea 80 percent give the fastest completions however? It's bad enough as it is to find all the civs and destroy them before they make spears.

 

[Marsden]

Would not pangaea 80 percent give the fastest completions however? It's bad enough as it is to find all the civs and destroy them before they make spears.

I would think so, but you'll need to get good luck and have them all on the same landmass, I don't think building a navy to attack favors a fast win.


The reason we picked this type of game is not to see if somebody could do it but who could do it best, unlike some high level games where it's a challenge for some, like me, to even finish.

 

[Elear]

My dear friend, the HoF is half about luck anyway. I think it tests your tenaciousness to find maps!

Speaking of islands...

Spoiler :

At least I got a settler!

 

[Chamnix]

I've been wondering if it is that necessary (or even desirable) to use 80% water maps. I've been playing 80%, and I've also gotten several games where some AI was isolated on an island (and some games where I've been the one isolated on an island ).

For a low-level conquest game, it doesn't really matter how much land there is total; what matters is how long it takes you to find and walk (or swim) to the last AI. Are 80% water maps necessarily more compact than 70% water? I guess they probably are on average, but it often seems like 80% maps are often still fairly spread out, but they have more choke points/peninsulas which I think are bad features in general for this type of game.

I think tonight I will be running a bunch of 70% water maps, and I'll see where that gets me tomorrow.

 

[Elear]

Chamnix,

In theory, that sounds good.

However I recommend that you stick with 80 percent. I have spent some time experimenting with 70 percent, and unfortunately the land size is A LOT higher. It takes a lot longer to get anywhere and find any civs. More civs get spears/bronze working. By 1000 BC, finding and destroying about 6 civs is realistic. I suppose you could get 7 if you are lucky.

I think the key is the find the right map where no civs start on a tiny island on 80 percent. This should get a 1300 to 1500 BC ending.

 

[AutomatedTeller]

I would agree.

My game could have ended at least 100 years earlier.

I think you have to go 80%. I don't think it's all that uncommon to get them all on the main map.

 

[keldr]

I think it's really a matter of patience.

I've tried a few 80% pangea games that all ended with at least one civ on an island far away...
I decided to do a little research. I started 25 games at both 80%, 70% and 60%, just to get an idea of the probability of all civs ending up on one big happy continent. I just played the first turn of the game and then retired to see if all the civs were on one island.

The results were:
80%: 2/25 = 8% probability
70%: 10/25 = 40% probability
60%: 15/25 = 60% probability

Of course this is not in any way scientificly representative or anything like that...

In theory, you should be able to get a faster finish with 80% if you can get all civs on one continent, but i would have to agree with chamnix that even at 80% it still seemed that the continent were not necessarily more compact than on 70%, there is just fewer tiles to explore...

Given my patience i think i will try a few 70% games given that the probability of all civs on one continent is about 5 times higher than 80%