MA Agartha, or Yet another thread fishing for tips

[alexti]

Evilhomer said:
Just my opinion but i would strongly advice against taking prod 0. Since the latest patch production was made alot better, and you are giving up a very nice +45% production bonus. With production 0 you will be expanding alot slower than with prod 3. I would also advice against taking growth at only 0. In a competitive build it seems like such a bad idea since you are giving up a ton of income due to the exponential way that growth works.

With MA Agartha it's not clear how to convert production into a serious advantage. The problem is that you don't have any troops you'd really like to buy. Heavy infantry has nothing for it except shields and heavy armor - skill-wise it's very mediocre. In the beginning I would rather use light infantry (which is not that light, in fact and also have shields). It costs a bit more gold, but you'll be expanding faster than with HI (even with prod 3). Faster rate of expansion will make up for extra gold spent. Besides, HI has strategic move of 1 vs 2 of LI. This makes expansion with LI is even faster in comparison with HI, because you can avoid wasting time moving (it takes careful planning though to bring replenishments, but luckily there's some synergy in using Golem Crafters heading to search for troop ferrying).

With that Oracle pretender your early expansion will be primarily bound by logistic rather than by lack of troops.

Later on, I can't see planning the game around HI. The core of the army will likely be national summons, backed with Golem Crafters casting something from magma or acid family of spells. HI will likely be just a complementary force. Besides, with easy access to armor buffs for your troops, heavy armors becomes much less important. So production scale is of relatively small benefit to MA Agartha. What are you going to sacrifice to take production scales anyway?

Golem Crafters will be very important mages through the whole game, because of their good choice of battle spells and their high survivability (even in the late game they are quite difficult to destroy due to their ability to forge cold/fire resistence items and armor or cast those protection on the battlefield). The more of them you can have the better - so you don't want to sacrifice income (or to spent it on other things with maintenance). Because of the same mages, you want to have decent research speed, so you don't want to go below Magic 0 (but I think Magic 1 worth it +33% bonus on your primary researcher). For fast initial expansion you need a pretender capable of expansion and Risen Oracle is pretty cheap choice that doesn't require any research and besides gives you easy access to higher level death summons (which will be important for Agartha for several reasons: magic diversification, stealth leaders for umbrals, exploting darkvision, undead priests for fodder summoning).

Growth doesn't really work exponentially, because if you survive long enough to be competitive raiding will quickly negate extra income from growth and effects of the growth early on are rather small. Considering poor mobility of MA Agartha (magical or otherwise), it's crucial to expand (by conquest) rather early which means that you won't be able to benefit from growth at this stage anyway, so the setup should be optimised to make this conquest phase most efficient.

Keeping this in mind, it seems that Luck, Growth and Production are the scales you need least for MA Agartha. I dislike Misfortune 3 from my [bad] experiences with it - I'm not sure if it is statistically justified.

Out of my earlier suggestions for the spare 120 points I lean towards Dom7, E2D4 Oracle and Prod+1 as a better choice (slightly better fear effect, quicker access to death summons and better death searches, slight dominion advantage).

[alexti]

atul said:
Alexti, the guy who did beat me up in almost every dominions 2 MP game I played.

Sorry about that

Good luck in your game!

[Evilhomer]

I was never suggesting any HI. You get +45% LI with that production bonus as well. The benefit from growth is exponential in your original lands (as well as the lands you take over), by turn 100 (if you are still in the game by then) it almost doubles your income (1.006^100+0.06=1.88), in effect giving alot more mages/fortresses. Of course you have to survive more than 20 turns for the growth to start to kick in, but in the build you should plan ahead since you are not building with the intent of loosing the game usually. Growth will help with that old age problem as well.

I stated earlier that i would take the points from the heat scale. Yes it gives slightly more fatigue, but it hurts your opponents just as much, or if you are out of your dominion it doesn't matter...And with growth and order+production gold bonus together (they all actually work together with good synergy) you can take the gold loss without much problem. Magic atleast +1 is a given, and i agree with taking points from luck scale (mis 2, maybe 3 i think).

Of course there is many ways to play dominions, and we all have different ideas how to play.

[alexti]

Evilhomer said:
I was never suggesting any HI. You get +45% LI with that production bonus as well.

Even with Prod 1, you're already gold bound rather than resource bound if you're buying LI (assuming you're buying Golem crafters every turn and saving gold for your next fort). That LI isn't very good investment and you don't want to buy more of it than necessary for the expansion.

Evilhomer said:
The benefit from growth is exponential in your original lands (as well as the lands you take over), by turn 100 (if you are still in the game by then) it almost doubles your income (1.006^100+0.06=1.88)

Heavy raiding usually starts much earlier than that and all those useful spells depopulate the land rather quickly too, and as Agartha you don't have too many defenses against raiding, so this is not very reliable plan. On another side, conquest usually increases your income by 50%+ in 10-15 turns. You should count on strong players conquering at least 3-4 nations by turn 100, so you need to match that pace if you want to have a chance.

Evilhomer said:
I stated earlier that i would take the points from the heat scale. Yes it gives slightly more fatigue, but it hurts your opponents just as much, or if you are out of your dominion it doesn't matter...And with growth and order+production gold bonus together (they all actually work together with good synergy) you can take the gold loss without much problem.

I am not aware of any synergy between order and production gold bonuses. They give 7% and 2% completely independently. Heat is also independent and gives you -5%. So each heat + production scale decreases your income by 3%. Besides, with the good battle mages, extra fatigue is likely a drawback too.

[Evilhomer]

the synergy im refering to is the gold % bonus (order +production) together with growth (constant % bonus +population increasing every turn). Im not sure if the battle mages of ma agartha is so powerful compared to the other nations, not if you factor in the price. They are not bad either i guess, but i would not rate them in the top 20 (among the 62 available).

There are ways for agartha to deal with raiding even if they are few, one way is to use earth attacks to just kill off the commanders (just get 1 earth bonus on your e3 mages).

[Micah]

Don't forget that your heat scale gets pulled by the seasons. If you have a heat 0 pulling hotter or colder both hurt your income by 5%, if you take heat-1 summer will hurt your income by 5%, which isn't any more than normal, but winter actually helps it by 5% by pulling it to neutral, so you lose less than the theoretical value by taking heat/cold scales.

[alexti]

Evilhomer said:
Im not sure if the battle mages of ma agartha is so powerful compared to the other nations, not if you factor in the price. They are not bad either i guess, but i would not rate them in the top 20 (among the 62 available).

Not sure how to rate (what criteria?) them, but you certainly need them on the battlefield casting offensive spells. Part of their strength is a nice power curve. They're getting better and better spells along Evo and by Evo-6 they have very good spells. So they can be efficient pretty early, long before you need to seriously worry about opponents anti-mage spells. Another strength is versatility - they can deal with thugs, SCs, heavy or lightly armoured troops. So at that phase of the game they're pretty high in the mage rankings.

Evilhomer said:
There are ways for agartha to deal with raiding even if they are few, one way is to use earth attacks to just kill off the commanders (just get 1 earth bonus on your e3 mages).

Nothing efficient I can think of - Earth attack is a reactionary measure (and it doesn't work versus many kind of raiders) that doesn't prevent income loss. I haven't tried played MA Agartha in MP, but I would think that the late play would have to be based on raiding using disposable regular troops and stealthy umbral squads of various strength while moving randomly some small Golem Crafter backed squads randomly in your territory. Lack of mobility still hurts though. But even if you defend successfully you still lose income due to unrest. And with raiding, poisoning and destructive province spells, it is in general easier to hurt opponent's income than preserve your own. So I'm generally pessimistic about prospects of preserving the income. On a good side, Agartha is in a decent position as far as being destructive is concerned, and it's also can alchemise if needed.

[alexti]

Micah said:
Don't forget that your heat scale gets pulled by the seasons. If you have a heat 0 pulling hotter or colder both hurt your income by 5%, if you take heat-1 summer will hurt your income by 5%

Wouldn't it pull the scale further by making you lose 10%?

[Micah]

Yes, it would be a total of 10%, but my point was that losing 10% a quarter of the time, 5% half the time and 0% the last quarter of the time is an average of a 5% loss, while a neutral scale will either be hot or cold about half the time, giving you an average of 5% half the time, or 2.5%. So the first point of heat/cold is only really a 2.5% loss, not the full 5 it seems to be. I'm not sure on the exact percentages on the season pulls, so that figure isn't exact by any means, but a 50% chance of a pull seems like a reasonable ballpark estimate. It's a significant amount, although it isn't a HUGE change. Taking a triple scale also kicks in another 'free' percent or two since seasons obviously can't pull you to a 4-point scale, so the only seasonal effect you can get is a beneficial move toward neutral.

[alexti]

Micah said:
Yes, it would be a total of 10%, but my point was that losing 10% a quarter of the time, 5% half the time and 0% the last quarter of the time is an average of a 5% loss, while a neutral scale will either be hot or cold about half the time, giving you an average of 5% half the time, or 2.5%.

Good point. I have tried few experiments and it seems that seasonal effects are rather random. I got average of about 6 out of 36 months when the scale was offset by 1 (Dom 7). Interestingly, that with Dom1, the average was about 9. Is season influence countered by high dominion? Number of experiments weren't large enough to make this observation statistically reliable. It still seems that seasons are not as bad as they initially look - costed only about 1% on average across all my experiments.

This seem to agree with my experiences - wrong scales don't work well if your preferred scale is neutral. Taking heat 2 when your preference is heat 1 is another story. It almost always result in a better income than taking heat 1. The drawback is that Heat 2 loses to Heat 1 in the first few months due to a large proportion of income coming from the capital.