GK2- The Training Day Experiment

[SolarKnight]

That sounds like a good plan scoutsout, looking at the opponents we have, looks like this game could end up being pretty violent towards the end, we have the russians, germans, english and french to find, the first two especially will probably give us a lot of grief later on, so we may have to keep an eye on them.

 

[Sir Bugsy]

Well thought out plan.

Note: Since knights and Impi are both movement 2, knights can't retreat from Impis either.

 

[scoutsout]

Originally posted by Sir Bugsy
Note: Since knights and Impi are both movement 2, knights can't retreat from Impis either.

True, but I'd rather attack the Zulu with 4/3 Knights than 3/1 Mounted Warriors... and unless we want to let Shaka hang around until Military Tradition, we're not going to get anything that hits harder than an attack value of 4....

 

[denyd]

But if you aren't going to be able to retreat anyway, why not hit him with 40 shield MDI (upgraded swords) instead of 70 shield knights (upgraded MW).

Try to save the faster units for open terrain (plains, grassland, desert) and let the slower units do battle in the tougher terrain, they'll also get the defensive bonus on the way their if they use the hills, mountains, forrest & jungle.

One more thing of note, you don't have to use your UU to start a Golden Age. What wonders could you combine to get the GA?

 

[Sir Bugsy]

Since this is vanilla civ, they won't have MDI. But Denyd brings up a good point. The swords will be best used in the jungles. The mounted warrior and any fast movers are best in open terrain.

 

[ControlFreak]

Good job planning scoutsout!

Of course, the farther out the time frame, the more potential for deviations. We will see how close you the team stays to whatever plan is decided and how to change the plan as needed. (I'm guessing MB threw in a few more shockers.)

To summarize your plan:

1. Fill area from lake to jungle, peacefully.
   (One question, are you sending units to block the chokes or are you sending settlers to make cities at the chokes as first priority?)
2. Using warriors that have been defending/MP, upgrade to swords and attack Japan, avoiding our GA but triggering Zulu's.
3. Finish Japan, Immediately hit Persians using MW.
4. Upgrade MW to knights and hit Zulu.

Some thoughts:
Others have said that Japan's UU is a knight replacement so there is some time to deal with them.
Japan is in a grassy area, good for MW movement.
Persia gets a swords UU but still needs iron connected to build it.
Persia is behind a jungle: MW have no advantage until roads, AI will be reluctant to settle the poor jungle tiles.

What I like about the plan:
Japan gone before Samurai.
Zulu GA done early.
It's a plan.

What I worry about:
Do we have iron? What do we do if we don't. All those warriors will stay warriors.
Not using MW on Japan wastes their advantage of a quick strike force. Delayed use of AA UU reduces their effectiveness.
Buying in Zulus requires writing. (May not have been researched yet.)
Buying in Zulus forces a 20 turn war against Japan.
Using MW against Persia with Immortals, in the jungle gives them the advantage. Their 4.2.1 Unit vs our 3.1.2(1) unit, movement dropped to 1 due to jungle (I think retreat is still an option.)

Originally posted by scoutsout
The research strategies are not carefully thought out.

Some thoughts on Research:
We've got 30 turns left until Polytheism. What techs do you think the three neighbors are going to research?
Persia is undoubtably going for irowworking since they get a UU with it and started with bronze. But since they only offered 17g for the wheel and probably didn't meet Japan yet, maybe they just got started on IW a few turns ago.
Since Japan started with CB and we've sold it to the others, Mystism may be the choice of the others. It enables the oracle and since we have it, their research cost is cheaper.

That brings up an interesting fact: I don't think anyone has alphabet. Is this true? (Forgive me if I missed it since I can't open save games.)
So, for the MA to work, the AI will have to get Alphabet and writing by the time our poly gambit is done. Without alphabet, I would venture a guess that the AIs will prefer to follow you up the poly branch after ironworking and HBR and you might have nothing else to trade for.
Very interesting choice of starting techs mad-bax.

It's interesting how much the tech rate can affect these long term plans. The plan makes some definite assumptions about techs:

• We get ironworking (and have iron) before our wars start. (Guessing Persia gets it first.)
• We get writing before the Japanese war is concluded (to get Shaka in MA). Unless we find a commercial civ, this could be late.
• The AI doesn't get Map Making before we claim our land. MM is even farther out since no Alphabet.

Not that any of those are a big deal, just that they affect the plan.

So what do we do once Poly comes in? Yes it depends on what else is available.

Assuming we get HBR, IronWorking, Alphabet and Writing, scoutsout said MM then literature. Why?

What about getting out of despotism? I would say this is a priority but should you research a government, or hope the AI does and buy it?

ACK! Rereading this, I just rambled. Oh well, take from it what you will.

 

[ControlFreak]

Originally posted by denyd
But if you aren't going to be able to retreat anyway, why not hit him with 40 shield MDI (upgraded swords) instead of 70 shield knights (upgraded MW).

Try to save the faster units for open terrain (plains, grassland, desert) and let the slower units do battle in the tougher terrain, they'll also get the defensive bonus on the way their if they use the hills, mountains, forrest & jungle.

One more thing of note, you don't have to use your UU to start a Golden Age. What wonders could you combine to get the GA?

Great input denyd! I agree with the open terrain and am interested about the wonder triggers.

As far as "wasting 70s knights" I've been thinking about that.
Assuming all units are 4HP veterans these are the results from BomberEscorts combat calculator for us attacking zulus.

THIS TABLE IS NOT CORRECT, SEE CORRECTION IN LATER POST

             Win  Loss  Retreat  S.Cost  S.Loss  S.Gain S.Net
4HP Sword   54.7%  17.2%  28.1%   30      5.16   10.94   5.78
4HP MW      71.0%  29.0%          30      8.7    14.2    5.5
4HP LB      60.0%  17.0%  23.0%   40      6.8    12      5.2
4HP Knight  90.0%  10.0%          70      7      18     11
4HP Cav     93.0%   7.0%          70      4.9    18.6   13.7

S.Loss is the sheild cost of the unit times the possiblity they might die.
S.Gain is the sheild cost of an impi (20s) times the possiblity that the Impi might die.
S.Net S.Gain-S.Loss

This assumes that any retreat results in both units seeking cover and healing back to 4HP.

While the slower sword/longbow are cheaper than the knights and thus lose few sheilds, they allow the Impi to escape and won't destroy as many Zulu shields. Yes you risk the knights dying because they can't retreat, but on a 4 attack vs 2 defense, the odds are in the knights favor and why not make the impi battle it out for the last HP.

Of course I may sing a different tune if you're bringing enough units to kill the retreated 1HP impies but they'll be hard to catch with the swords. I think mixed stacks of swords/MW would be a good impi killing tool. The sword makes the first attack, kills or retreat the impi and then the MW finishes off the retreaters. The MW has the same attack as sword so if there are more units than the swords can handle, finish with the MWs.

Personally, I'm not afraid of an early GA. I would consider Sword/MW combo, blitzing through Japan and Zulu, leaving Persia on the other side of the Jungle. Use the GA sheilds for a ton of MW and then take the immortals with upgraded knights.

Of course who knows where the other civs are and what challenges they bring.

 

[scoutsout]

Thanks for the compliment, ControlFreak...as usual, LOTS of good stuff here.

(One question, are you sending units to block the chokes or are you sending settlers to make cities at the chokes as first priority?)

...um "yes".

Others have said that Japan's UU is a knight replacement so there is some time to deal with them.
Japan is in a grassy area, good for MW movement.

But they are also the ones we're going to be competing with in the land grab, so they'll be the first to feel pressured or be tempted to pressure us militarily.

Persia gets a swords UU but still needs iron connected to build it.
Persia is behind a jungle: MW have no advantage until roads

Attacking Japan first gives us an opportunity to get some workers down there to deal with the jungles...

Do we have iron? What do we do if we don't. All those warriors will stay warriors.

One reason we need to acquire iron working soon... because if we don't have iron we're looking at either an archer rush or a fundamental shift in strategy.

Not using MW on Japan wastes their advantage of a quick strike force. Delayed use of AA UU reduces their effectiveness.

If we were talking about delaying it until well into the Middle Ages, I'd agree. Using them against the Persians really late AA or very early MA will give us an early boost in the Middle ages

Buying in Zulus requires writing. (May not have been researched yet.)

It is on the way to map making... which is the other research priority in the early stage (suicide galleys...)

Using MW against Persia with Immortals, in the jungle gives them the advantage.

Assuming we let the Immortals close with the MWs in the jungle, yes... again, get some workers down there while we're fighting Japan, and we can make the terrain more favorable to us...

...and I'm rambling back...so I'm going to finish this and come back later when I can make a more carefully considered post...(I'm sure I'm cross-posting by now...)

Edit: One other thought... if we take Japan to the land bridge, geography will be on our side defensively, and we can put off taking the Zulu... who tend to be easier to take with the passage of time. A stack of Immortals will come unless we go after Xerxes first. I'm betting Xerxes has iron... why bother putting Persia down there with no iron? What would be the sport in that?

 

[denyd]

CF: How much would the battle odds change if the Impis could not retreat (garrisoning a city)?

When looking at battle odds, I rarely worry about open combat, where the defender has no bonus to help out, those should be wins for the attacker. When you get to the cities, is where you have to look at the odds before attacking. If that city has walls/+6 size and the unit is fortified, then the odds get worse for the attacker. I would hope that the attacking stack would have a couple of defenders to keep the Impis at bay until the Zulu cities could be reached. If add a couple of knights (combined arms) to the stack you'll be able to pick off any stray archer/sword before they reach your attackers and still have the extra move to return to the protection on the stack.

BTW CF, nice analysis on the attacking odds. I don't normally consider the +/- shields when looking at battle odds.

 

[ControlFreak]

Originally posted by denyd
CF: How much would the battle odds change if the Impis could not retreat (garrisoning a city)?

I just did worst case scenario of Veteran Impi in Hill City/Town with wall:
THIS TABLE IS NOT CORRECT, SEE CORRECTION IN LATER POST

        Win    Loss   Retreat S.Cost    S.Loss    S.Gain    S.Net
Sword   23.10% 70.10%  6.80%     30    21.030     4.620   -16.410 
MW      29.00% 71.00%  0.00%     30    21.300     5.800   -15.500 
LB      34.30% 55.60% 10.10%     40    22.240     6.860   -15.380 
Knight  43.60% 56.40%  0.00%     70    39.480     8.720   -30.760 
Cav     65.30% 34.70%  0.00%     70    24.290    13.060   -11.230

Now the cost of knights make them a real liability. Good catch denyd. This tells me, you don't want to attack cities unless you plan on losing shields relative to your enemy. It also says avoid cities on hills with walls unless you can bombard them first.

EDIT: Cavs can retreat from Impi which makes them:
THIS TABLE IS NOT CORRECT, SEE CORRECTION IN LATER POST

Cav     64.50% 27.50%  9.00%     70    19.250    12.900    -6.350

 

[shoguntaka]

Wow, what a wondeful set of information! I like learning new things about CivIII every week!

One thing I'm confused is the retreat chance for the Impi,
I've always thought that it was 50%.

Does higher attacking unit reduce the target's retreat %?

I noticed the different retreat % for Longbow and the swordman.

And I noticed that the Swordman and MW has different Shield net, but I don't think it's true. The Impi can't retreat when fortified (can they?), so the swordman and MW should have same shield net.

And one last question,
how come the MW has higher % chance of winning than the swordman, since they both have same attack rating?

 

[Sir Bugsy]

Originally posted by shoguntaka
Does higher attacking unit reduce the target's retreat %?

The way I understand retreats: When the higher movement unit hits 1 hp left the RNG decides whether the unit will retreat. The unit will never retreat if the other unit is also at 1 hp.

how come the MW has higher % chance of winning than the swordman, since they both have same attack rating?

Since the Impi can't retreat from a MW the last hp must be fought, which the MW has a higher chance of winning.

I'm not sure about the answers to the other questions.

 

[shoguntaka]

Then, the Imp's retreat % should be higher when attacked by a Longbow than a swordman, but the chart showed the other way.

 

[Gengis Khan]

You got me there, I was actually wondering pondering the same thing.

 

[Sir Bugsy]

CF?

We are now officially out of my league.

 

[Gengis Khan]

Doh! I was really hoping you'd pick that one up for me bugs.

 

[scoutsout]

Hey guys...I hate to sound short-sighted, but can we get back to the part where we were discussing where we're gonna put that second city? After all, we're gonna pop out a settler on the next turn...

 

[Sir Bugsy]

Oh! Yeah! A game to be played! Good distraction scout! Where do we put that city?

 

[scoutsout]

Well...Shogun and I seem to agree on one thing... the floodplain sites have possibilities. Though we don't agree on the order (or the exact placement of one) harnessing the power of a floodplain city seems to be a recurring theme...

I see 2 possible advantages to settling the southern city early:
1) Workers pumped from that city could build a nice road towards the choke point
2) if we could get some settlers out of that thing, we could concentrate the efforts of our capitol settler-pump towards the north...

The site to the west: Workers pumped from this site could quickly build infrastructure northward towards Japan...(an early military target)

My only reasoning for settling the southern site first is because I want about 5 settlers out of it plus a few workers, and it might take a little more effort if a granary would be required there to pull that off. The western site appears to be a good one for a worker pump, but not as nicely located for settlers...Just my thoughts when I drew up the dot map.

 

[ControlFreak]

Originally posted by shoguntaka
the Imp's retreat % should be higher when attacked by a Longbow than a swordman, but the chart showed the other way.

Originally posted by Sir Bugsy
CF?

We are now officially out of my league.

Obviously we are now out of my league too. The numbers were wrong. Chalk it up to a learning curve on BomberEscorts utility. It covers every possibility for a combat (including C3C terrain bonuses). But it changes some fields when you click other fields. I wasn't paying close enough attention to that when I switched from fast attack to slow etc.

Anyway, I hope this makes more sense:

Open, no bonus       Impi S    20
Vet Impi vs  Win  Loss  Retreat Cost S.Loss S.Gain  S.Net
Vet Sword    53.4 28.5    18.1  30    8.55   10.68    2.13
Vet L.Bow    60.0 17.0    23.0  40    6.80   12.00    5.20
Vet M.War    71.0 29.0     0.0  30    8.70   14.20    5.50
Vet Knight   82.7 17.3     0.0  70   12.11   16.54    4.43
Vet Cav      92.3  5.7     2.0  80    4.56   18.46   13.90

Hill City or Hill Town w/walls
Vet Impi vs  Win  Loss  Retreat Cost S.Loss S.Gain  S.Net
Vet Sword    23.1 70.1     6.8  30   21.03    4.62  -16.41
Vet L.Bow    34.3 55.6    10.1  40   22.24    6.86  -15.38
Vet M.War    29.0 71.0     0.0  30   21.30    5.80  -15.50
Vet Knight   43.6 56.4     0.0  70   39.48    8.72  -30.76
Vet Cav      64.5 27.5     8.0  80   22.00   12.90   -9.10

So in the open, the MW is the way to go until cavs. In fortified cities, L.Bows are most cost effective until cavs.

Originally posted by scoutsout
can we get back to the part where we were discussing where we're gonna put that second city? After all, we're gonna pop out a settler on the next turn...

Not being able to play, the theory is everything. But you're right scout, you've got a game to win. Discuss the differences between the two FP cities or a possible high sheild city and leave the next play to choose the teams destiny. Part of an SG is the fact that the next play has the ultimate say on what gets done. You probably have enough thoughts to make a reasonably good choice and the team can find a way to continue from what ever decision is made.