Troops vs magic, what am I doing wrong?
 

OK, this has been bugging me for a while, and I can't seem to figure out what I'm doing wrong.

I typically play SP against several Difficult AI opponents. No matter what I do, it seems like troop strength seems to matter way more than magic, at least up to level-6 research. At what research level is magic really relevant?

For example, I was recently trying EA C'tis. I thought a good easy strategy to use magic would be to hand a few Reborn some death gems and let them cast Shadow Blast. But when I do the math, it doesn't add up:

Reborn: 100 gold
Death gem: equivalent to 25 gold
Shadow Blast: hits 7 squares, I estimate smacks a net 15 gold worth of enemy troops per square (between misses, hitting my troops, and similar effects)
7 * 15 = 105, < 125 gold.

For 125 gold I can field 9 Elite Warriors which are going to do a lot more damage.

Only other thing I can see to do is hand the Reborn three gems, to get a fatigue reduction on the first casting so they can cast it a second time. But that's equivalent to 175gold cost, which is still not that efficient against 210 damage.

Seems like I should skip recruiting mages altogether, take the extra cash and build castles, and just build a swarm of Elite Warriors. I've tried this, and it works really well. What am I missing in my attempts to use magic effectively?

Karl

Re: Troops vs magic, what am I doing wrong?
 

well, your math is fine, its your strategy thats wrong.
with ctis try spamming skeletons, avoiding the gem cost.

even so, you do have a high opportunity cost - those mages could be site searching etc. Research is the way to increase the effective value of a mage, ie, as you research more and more
the value of mages increases vis-a-vis normal troops.

Re: Troops vs magic, what am I doing wrong?
 

Magic matters more then anything else in this game, you just need to learn choosing right spells. Evocation school isn't just the best choice if you are using death mages. Try researching some Enchantment instead for a start, and make your mages cast Rise Skeletons five times - you'll see they are much more effective this way.

Re: Troops vs magic, what am I doing wrong?
 

If, for whatever reason, you really *do* want to have Reborn cast shadow blast (vs. just recruiting more Sauromancers instead), you might want to invest in some Skull Staves. At 7D w/hammer, you're going to come out ahead vs. using gems to cast very shortly. I'd recommend just going with the Sauromancers, though, and, as recommended above, doing skellispam.

Re: Troops vs magic, what am I doing wrong?
 

In general, magic is used to create combos that national troops cannot handle and to counter combos that your national troops cannot handle. In your question you describe the first case, where at research level 7 you start to see battle winning spells (Mass Protection, Fog Warriors, Darkness, etc.) that let you massacre unsupported national troops. And so long as your opponents just use human-level troops against you, you can get away with not using magic, or in waiting until you've invested enough use the high level spells.
However, if you are facing the second case where your opponent has something your national troops cannot handle (elephants, combat pretender, blessed sacreds, etc), magic is needed to counter their advantage. You see this primarily in MP, since in SP the AI isn't smart enough to put together vicious combos.


For the particular case you gave with EA C'tis, I'd recommend that old stand by of skelespam. While your casters are initially expensive, they do not take any casualties so long as they win the battle. With a normal army, you lose a portion of your troops after each battle, but the casters can keep fighting indefinitely without losses. And once you reach Evo 7, you can combine the skeletons with Cloud of Death, and then things become really fun.

Re: Troops vs magic, what am I doing wrong?
 

Nevermind, I was ninja'd by like 3 other people. :)

Re: Troops vs magic, what am I doing wrong?
 

Right, I can see at high level research there's a mess of battlefield spells that look effective, provided you can cast them against the enemy's army and not his PD. I seem to have a hard time intersecting his armies with mine... only thing that works reliably is to park my army on his capital and make his armies come to me :) And yeah, the AI rarely throws anything at me that my national troops can't handle other than "too much" which is what I was hoping mid level magic could help with. It seems like to vaporize his 300-soldier hordes, I have to have those L7+ spells. Skelspam really isn't going to stop an army of that size... nor is falling fires or thunderstrike or any other mid-level spell cast by a similar gold-worth of mages.

I guess the trouble is trying to figure out how to balance the game right in AI. Any AI that is weak enough for me to survive against while spending lots of gold to research high-level spells is also weak enough for me to skip the research and just crush with troops, it seems.

Re: Troops vs magic, what am I doing wrong?
 

Against the big AI armies, you need to use a lot of mages to have an effect.

The key is that you may have to use a similar gold cost of mages to destroy a good sized AI army, and you could also beat them with the same gold cost of troops, but you'd lose a lot of troops and the mages will all survive. (You may lose some of the troops you use to keep the AI hordes away from your mages, but they're replaceable.)

It may take 3000 gold worth of mages & 500 gold worth of meatshield troops to beat that 300 unit AI army, but you'll still have the mages and he'll have lost the army.

The other approach is buffing your troops, but I don't think C'tis works well for that. Quickened, lucky troops with strength and protection bonuses work quite nicely.

Re: Troops vs magic, what am I doing wrong?
 

Speaking just of SP and facing down the AI, the most dramatic way to see the power of spells is to take EA Abysia and expand and hire mages to research as normal until your mages have researched both phoenix power and falling fires. Then take a huge chunk of your research mages and join them up with a reltively small army. Script them to PP then FFx4. If you have a critical mass of them, it doesn't matter what the enemy is throwing at you, they'll burn it to a crisp in a few rounds.

The difference between this and an army, again speaking purely in SP terms, is that an army will take casualties against the AI. A critical mass of mages with some blockers will not.

Re: Troops vs magic, what am I doing wrong?
 

Skellispam is definitely the most cost efficient move for C'tis. You're going to get much more than 5 casts of Raise Dead per Sauromancer because as you flood the battlefield with skeletons, your mages have time for their fatigue to dip below 100 and cast it again. Also, your opponent's fatigue starts to build and as the battle drags on they start taking more hits.

Shadow Blast is much more useful for removing difficult units from the battlefield than it is for just general destruction. For instance fighting double blessed troops that could kill 10+ Elite Warriors before going down would be a good time to use Shadow Blast. But the computer obviously doesn't do that so I don't think there's ever really a need for that spell against the AI.

To take your argument to extremes, it's much more cost effective to put 125 PD into every province than it is to recruit Elite Warriors. For a one time investment of 8,000 gold your province becomes impregnable for the rest of the game. The AI almost never builds the right troops in the right numbers to defeat a free army like that.

The problem with Elite Warriors and PD is that they aren't time efficient. It's going to take years and years to conquer the enemy a few provinces at a time. With magic, once you get the spells you need you can teleport in raiders, wipe out huge armies with just a couple spells, basically redraw the map in a couple turns. It's not cost effective for a long time, but it is efficient.

PD: Cost Effective, not efficient
Elite Warriors: Somewhat cost effective, somewhat efficient
Mages: Not cost effective, very efficient (becomes moreso at higher research levels)

Re: Troops vs magic, what am I doing wrong?
 

[quote=Sombre;742055]Speaking just of SP and facing down the AI, the most dramatic way to see the power of spells is to take EA Abysia and expand and hire mages to research as normal until your mages have researched both phoenix power and falling fires. Then take a huge chunk of your research mages and join them up with a reltively small army. Script them to PP then FFx4. If you have a critical mass of them, it doesn't matter what the enemy is throwing at you, they'll burn it to a crisp in a few rounds.
[quote]

Abysia is a "special case" -- I have tried sending out Anethemant Salamanders with a small army and watched with glee as they set the world on fire. I call it a special case because of one key element -- abysian troops are fire immune. So when my mages try to incinerate my own troops (which they seem to love to do!), they fail.

Anethemant Salamander: 130gold. 3 Falling fires, area effect 3, doesn't incinerate my own troops, so net 60 per casting, 180 total, a nice 50% bonus over troops.

I think the trouble is that I'm slightly spoiled by how effective this is, as well as MA Ulm's magma eruptions. I was hoping to find something similar for most other nations, but it's just not to be.

Karl

Re: Troops vs magic, what am I doing wrong?
 

LOL!!! You know, now I'm going to have to try this and see what happens!!! This may be the ultimate AI opponent killer strategy! ;->

Re: Troops vs magic, what am I doing wrong?
 

I run the kind of games where AI armies just roll over 50 PD. I doubt 125 PD is going to stop Impossible AI beyond year 3 or so.

Re: Troops vs magic, what am I doing wrong?
 

Just play a single MP game, just one, and you'll learn a lot more about the power of magic than possible in a forum thread.

Magic is cruical for all nations, in one way or another. The Abysian approach is obvious, but not the most powerful.

In SP many strategies will seem cost ineffective simply because the AI is stupid. You'll never get the kind of resistance where you *have to* use magic competetively.

In MP you are faced with the simple condition: maximise your magic potential or die. Which is what the AI will do when you bring this new understanding back into SP.

Re: Troops vs magic, what am I doing wrong?
 

Your maths is not correct at all, because the mages (should) survive the battle. So you can't calculate the cost of a shadow blast as being (cost of mage) + (cost of gem).

I suppose if you wanted to analyse it like that, you'd have to consider the expected number of castings the mage would make over his entire life. Say you'd expect N. Then the question is whether the troops killed by N shadow blasts cost more than N death gems and one death mage.

Anyway, Shadow Blast is an awesome spell, but should be saved for special occasions, when you really NEED to win. Mostly, as other people have mentioned, you should use skelly spam.

Re: Troops vs magic, what am I doing wrong?
 

As a further example... it is relatively easy to take out 8000 gp worth of pd. It can be done with as little as 100 gp worth of troops....

Re: Troops vs magic, what am I doing wrong?
 

I can't think of a 100gp troop that the AI uses that will kill 125 PD. But feel free to prove me wrong.

Also, I'm not sure that the game's difficulty level affects this much. They get more gold and resources but they aren't any smarter. Most of the time the PD is such a deterrent that they will not even attack you. Even if they do have a huge army on your border, they rarely attack with the full stack and/or they use poor orders. The difference between 50 PD and 125 PD is unbelievable, just make a SP game and try it out.

Re: Troops vs magic, what am I doing wrong?
 

Late-game AI on impossible difficulty will wipe through 125 PD without effort. 125 PD w/ a dozen decent battlemages backing it up though will stop most late-game AI armies, but will suffer occasional mage casualties to enemy fliers and spells and be defeated occasionally by particularly strong armies. Also useful for stopping AI armies in defensive situations: PD + half a dozen battlemages + some middling zero-upkeep-unit-producers (unholy priests, fairy queens, whatever). If the province gets left alone for a while it will start building up a few hundred zero-upkeep no-need-to-eat soldiers, which should hopefully make it impossible for the enemy to take it later on as long as the combat mages stay to support the troops.

In my experience vs AIs you generally want troops until about research levels 4-5, then a mix of mages and troops (possibly using PD or summoned troops to minimize upkeep) until research levels 8-9, then an end-game strategy (massed clams if not using a mod that makes them unique, massed juggernoughts, massed vampire lord or wraith lords, some wished up seraphs or chayots w/ full items).

Re: Troops vs magic, what am I doing wrong?
 

Huge PD is rarely worth it even against the AI. Maybe at a chokepoint if you want to concentrate on other fronts.

Thugs are the real way to go against the AI. It's not hard to build high end thugs/ low end SCs that essentially can't be killed by normal troops. The AI isn't creative about killing them, so it's worth investing the gems. You know the main paths that nation has, so protect against them and the AI won't surprise you like a human player will.

Re: Troops vs magic, what am I doing wrong?
 

Have you ever seen AI armies in the four-digits?

Re: Troops vs magic, what am I doing wrong?
 

I'm on year 10 in my current game (EA T'ien Ch'i). Typically 1000 guys will roll up to my border starving and demoralized, attack 200 at a time for 3 or 4 turns and then sit in their castle until they get reinforced by another army. The only threat is Ermor because they can't starve and they can skellispam from the back so 1000 guys is actually more like 1500 by the end of the fight. I do lose to the big Ermorian armies without mage support, but most nations buckle like a pilgrim's shoe.

Re: Troops vs magic, what am I doing wrong?
 

Do you have supplies set to 300?

Re: Troops vs magic, what am I doing wrong?
 

Is there a way to bypass the artificial supply multiple limit of 300?

Re: Troops vs magic, what am I doing wrong?
 

No, but you could have a map where every province has a supply bonus site, or you could make all commanders give a huge supply bonus, if you want to remove that part of the game entirely.

Re: Troops vs magic, what am I doing wrong?
 

I've been playing a 300 everything EA game on Faerun map, and not a turn had passed without AI wiping a 125 PD of mine here or there. Everyone from Sauromatia to Atlantis to Arcoscephale to the japanese. Not with particularly large armies even, typically in 500-1000 range, and that's with normal difficulty.

Of course, I played Niefels, and their PD is notoriously sucky, or so I've heard.

Re: Troops vs magic, what am I doing wrong?
 

I do play on 300 supplies, but the really large armies still starve if they province they are sitting in doesn't have enough population. I guess I am an anomaly, or I've just been using nations with powerful PD. The worst things for PD to have is low morale, and/or flyers. So Mictlan, Caelum, the giants, etc. are going to get rolled eventually. But most human nations do fine: T'ien Ch'i, Eriu, Ermor, Pythium, Utgard, etc.

Re: Troops vs magic, what am I doing wrong?
 

The way I'm doing the math is based on the observation that battles against the AI tend to be close to "all or nothing" affairs. If I come to the fight with 50 troops, and he has 100, I'm going to lose 40 and he'll lose 10. If I can add 100 troops to that battle, he'll lose 80 and I'll lose 20. My losses when I win a battle are relatively minimal, and my losses are catastrophic when I lose a battle.

I'm looking for mages to provide a greater likelihood of winning... too often, it seems like they instead increase my losses -- my army of 50 troops and 5 mages meets his 50 troops, and my mages incinerate half of my troops in the process of killing his :)

Re: Troops vs magic, what am I doing wrong?
 

This makes sense only on normal difficulties. On Impossible settings, with custom advantaged AI-Gods, the AI will outbuild you 10-1 on conventional units. Units like knights will achieve a very favorable KD ratio, but there is only so much they can do before the AI swarms all over you.

Re: Troops vs magic, what am I doing wrong?
 

To answer your original question, magic is useful at ALL levels, but it depends on what you're trying to do with it. Most lower level magic is best used on small numbers of very expensive or otherwise rare things, rather than trying to be useful in army combats. For example spells like earth meld or mind burn are very useful on relatively small numbers of enemy elites. Under CBM, fire prison is an amazingly good spell. Try it sometime. It can be a real battle winner, especially if you rely on ranged units or something like summer lions. There are some exceptions to this though. Thunderstrike is an amazingly good spell no matter the number of troops the enemy has... it demolishes anything that does not have full lightning immunity, is pretty accurate, and can be spammed by thuggish mages who can survive without really needing troops to guard them. Try playing tir na nog, eriu, or ea caelum to get an idea of what I mean.

By and large though higher level magic is just much more efficient, especially buffs. Fog Warriors is one of the most powerful spells in the game IMO.

One thing that can help make your mages more efficient is using trick armies. Like say you're playing marignon. You can use summer lions to occupy the enemy army while your mages rain fire down on their heads, or have all your mages cast resist fire and then cast fire storm. Or with your C'tis example, using undead in combination with the raise dead spells (yes I know everyone's talking about these...), shadowblast, cloud of death, etc. Undead are immune to the effects of death spells so when you accidentally hit them they really don't care. A few wights or a bane lord or two can absorb all the missile fire they throw at you, and are quite tough to kill, and can hold the attention of nearby troops long enough for your mages to create an impenetrable wall of skeletons while dropping painful spells on them from afar.

Have you tried thugging? That sounds like something you'd feel is way too inefficient, but it works very well, and is really the best way to fight the AI. If you haven't and want to get a good idea of how effective thugging can be, try the following: Play Tir na N'og with an e9n4 bless. Try making armies that consist of 2 sidhe thugs, and 3 mages with air magic to at least level 3. Thugs should be equipped with a vine shield, a frost brand, and possibly rainbow armor to help with fatigue. Nothing else. Script bless, air shield, mistform, resist lightning, barkskin, attack closest. Script the mages to air shield, bless, aim, thunder strikex2, spells. As long as they don't have any flanking units, an army like this should be more or less invulnerable to conventional AI armies. The mages aren't really even necessary to be honest, they are just there to prove that it can be done :P Something like that is safer with death mages because of the skelly spam acting as a barrier to keep them safe.

Re: Troops vs magic, what am I doing wrong?
 

Exactly, as I try higher difficulty levels, I just get swarmed by enemy troops, so I'm looking for strategies to wipe out his swarms. What I find is that most of the magic is geared toward taking out small numbers of quality troops, not huge amounts of chaff.

What is all the excitement regarding Thunderstrike? I've tried it several times, and it never seems to work for me. My mages nearly always target the periphery of an enemy squad instead of the center, so I kill the three guys in the square it hits and injure half a dozen nearby. Very underwhelming. Is it critical to cast an aiming spell first to make this useful?

Re: Troops vs magic, what am I doing wrong?
 

[quote=rdonj;742583]Have you tried thugging? That sounds like something you'd feel is way too inefficient, but it works very well, and is really the best way to fight the AI. If you haven't and want to get a good idea of how effective thugging can be, try the following: Play Tir na N'og with an e9n4 bless. Try making armies that consist of 2 sidhe thugs, and 3 mages with air magic to at least level 3. Thugs should be equipped with a vine shield, a frost brand, and possibly rainbow armor to help with fatigue. Nothing else. [quote]

I have tried thugging. Normally to keep my thugs alive, I have to put a ridiculous amount (50+gems) of equipment on them, making them not cost effective. In the example you describe the equipment is minimal, so maybe I should give that a try. I'm not convinced it is going to work -- against 100+ troops, mistform never seems to last long enough. Once it fails, my thugs die quickly.

Re: Troops vs magic, what am I doing wrong?
 

It sounds like you're trying to thug the wrong stuff.

And even if a thug does cost 50 gems (which it never should) if it then defeats army after army of the AIs, that isn't cost inefficient at all.

Re: Troops vs magic, what am I doing wrong?
 

One thing to keep in mind is that a lot of advice is geared towards MP, because that's what most of the experienced posters play. Small numbers of elite units is the efficient way to go, so that's what the standard strategies counter.
SP is easier to beat, because the AI only has only has one real strategy and doesn't try to counter yours. You have to figure out how to effectively kill large numbers of troops.
Cheap raiding thugs work in MP, because no one invests in PD or has large armies wandering around in their backfield. And there's no point investing more gems into them, since if they run into a real army with mages, they'll die anyway unless they're a full-fledged SC. Or they'll get hit by something designed specifically to take them out.
In SP, you don't have that problem, but you'll generally be hitting stronger PD and larger armies so the thugs will need to be tougher. Which is OK since they won't be targeted by anti-SC forces.
You were playing C'tis? Have you tried Banelords? (Or Banes for a lesser verion) Maybe, Fire/Frost Brand, Vine Shield, Horror Helm, Luck Pendant, Antimagic Ammy or regen if it works on them. Flying shoes if you can get them. The default armor is good protection, Forge something if you need specific resistance or a special ability.
Wraithlords, with similar gear and casting Soul Vortex are better, but more expensive.

Re: Troops vs magic, what am I doing wrong?
 

The thing about thunderstrike is, used en masse, it can defeat ridiculous amounts of units as long as they aren't immune. Nine e9 tuatha sorceresses, for example, should kill about 27 units per turn, even before accounting for the splash damage (which also causes fatigue, which can allow your normal troops to defeat them more easily and even cause them to pass out altogether) which should allow them to kill significantly more. And because they have e9 in this example, the extra reinvigoration will help them to keep throwing off these thunderstrikes. Aim is not critical, per say, but it does help make each thunderstrike more efficient. Thunderstrike is also more efficient under storms, when you cast summon stormpower. Which can then allow your a2 recruitables to also spam thunderstrike. So a nation like TNN can mass a lot of thunderstrikers if it needs some chaff clearing. Of course, at 6 evo, you get wrathful skies... which will eventually kill any army that's not completely immune to lightning, and would not be hard for just one mage to cast behind a handful of lightning immune thugs.

Here are a few spells to try if you still find magic too inefficient: curse of the desert, stellar cascades, strength of giants, legions of steel, flaming arrows, wailing winds.

Curse and stellar cascades can weaken troops enormously, and even outright kill them with enough casters, but only require a few to have a pretty significant effect on tough troops. Strength of giants through flaming arrows are good army strength multipliers, and will make your own troops even more effective on a unit per unit basis. Wailing winds, well, used appropriately it will let you rout armies with much fewer casualties, which will have the side effect of giving you lots of easy kills in battles.

Also I agree with thejeff with regards to banelord kitting, but if you're using them with skellyspam as I've suggested you can drop the air boots and horror helm, and they'll still work. Banelords are NOT as durable as they might seem though, and should not try to take on large armies by themselves. They work better as raiders and army support. Taking on 100+ man armies single-handedly is the domain of real SCs and upper end thugs like sidhe :P And a final note on thugging: one of the biggest killers of thugs vs conventional armies is fatigue. Having more than a net +1 or 2 fatigue per turn is pretty much suicide.


Edit: Oh yes, I forgot. I was going to say that I once won a game against full on impossible AIs using unequipped sidhe lords as though they were troops :D I won the significant majority of my battles with those, and only started using real thugs in the end because it was taking too long to finish the enemy off. I lost probably less than 10 sidhe lords that game.

Re: Troops vs magic, what am I doing wrong?
 

OK, so I tried thugging with Sidhe Lords, just to see how it would work out. First, I never got a chance to apply vine shields and frost swords to them -- I got sucked into an early war against Ashdod. But I did have alt-3 researched, so with mistform, barkskin, and a E9N4 bless they held up very well against chaff. 4 of them could easily take out an army of 100 light infantry, which is about par for cost effectiveness.

But once I got to Ashdod's capital, the tables got turned. His mages stepped outside, summoned earth elementals, and my Sidhe Lords turned into Sidhe Roadkill. :(

So... I'm not convinced an army of 80 Firbolg javelineers would have fared worse. I would have had more casualties against his regular troops but roughly the same number of wins, and I don't think I would have gotten pasted so badly at the end. Lots of small troops with throwing weapons is just the thing to kill tramplers.

Seeing how painful that was to me, though, gives me an idea, maybe I need to fling some elementals at the AI and see what happens :)

Re: Troops vs magic, what am I doing wrong?
 

Two or three battles will attrition 80 Fir Bolgs to nothing. And typically by year 5 I face armies of 500+, against which thugs work much better.

There are also somethings against which combat thugs don't work well, and tramplers are one of them. The spell Swarm kills life drain thug chasis. Nobody said they are invincible.

Re: Troops vs magic, what am I doing wrong?
 

If you're facing mage-supported armies or mage-supported PD, a few thugs are not going to cut it. You could just as easily been killed by a few fireballs, or skelly spam, etc. To take an enemy cap, you'll need some real SCs, or a mage-supported army of your own (which of course would still do well to include some thugs).

Re: Troops vs magic, what am I doing wrong?
 

Well the point is you take advantage of when the thugs are extremely efficient and avoid situations where they won't be efficient. If you're saying 4 took out an army of 100 light infantry with no casualties, that's not "about par". It would be "about par" for gold-efficiency if you lost 3 of them and the last one was crippled. Sidhe Lords are stealth and cloud trapeze capable and so can pick their fights fairly well.

Re: Troops vs magic, what am I doing wrong?
 

Typically AI troops are poor quality, starving, etc., so if I come at an AI army of 100 troops with my 60 Principes I'll lose about 10. With the Sidhe Lords, I was losing about 1 every 2-3 battles. They're not indestructable until you give them the shield and sword combo, which pumps up their cost a bit.

As long as you attack where the enemy is weak and run away where he is strong, you really don't suffer too many casualties. The trouble is that when you're surrounded by 6 AI opponents all gunning for you there's few places to run and hide from his strength. Maybe I really should take up raiding.

Re: Troops vs magic, what am I doing wrong?
 

OK, so I found a convoluted way of getting some serious firepower at seriously low cost with magic. Is this sort of result typical, or am I taking advantage of an exploit that's likely to get nerfed?

1) EA Ermor
2) Principe meat shield
3) 5 Augurs. No, not Auger Eldurs (250 Au apiece, expensive! 80 Au Augers -- cheap!)
4) 3 astral pearls on the 5th Augur.

Lead Augurs are scripted to Communion Slave, Hold, Hold, Astral Fires, Astral Fires.

Tail Augur is scripted to cast Communion Master, Power of the Spheres, Light of the Northern Star, Astral Fires, Astral Fires.

Cost: 400 Au, 3 pearls
Result: 10x Astral Fires.
Efficiency: Medium. If an Astral Fires does 50hp damage average (depends on if I'm up against good MR troops or not), then it works out about even.

Smashing the AI with a pile of Principes is a whole lot easier, though :)

Re: Troops vs magic, what am I doing wrong?
 

That's not an exploit, it's a small communion. Against anyone bar Abysia, getting the Master to do PoTS and Phoenix Power followed up by everybody spamming Falling Fires would cost less gems and quite possibly do more damage.

Re: Troops vs magic, what am I doing wrong?
 

That's why we play on 300 supplies.

Re: Troops vs magic, what am I doing wrong?
 

I like the extra range on astral fires, it allows me to hit the archers the AI parks in the backfield. The AN damage seems a nice bonus too, along with slightly larger area effect.

You're right, though, that I could get away with one gem instead of three. Am I right in thinking that two gems won't work, because the AI will waste both on the first casting and have none left for the second? I guess I should just give it a try and see what happens :)

Re: Troops vs magic, what am I doing wrong?
 

Try these thugging options with E9N6 Sidhe Lords:

Option 1: Vine shield + Rainbow Armor. That's 20 gems.
Script Blessing, Air Shield, Mistform, Resist Lightning, Soothing Song, Cast spells

With Con-4, Alt-3 and Evo-2, your Sidhe Lords will cast Shock Wave which will wipe out AI chaff. You should have enough reinvig and regen from the bless to survive a long time, but you can also slap on the Ring of Regen and the Amulet of Resilience to help, or put on a Ring of Tamed Lightning and skip the Resist Lightning part, which brings the gem cost up a little. But the above setup can be mass produced and is quite survivable.

Options 2: Snake Bladder Stick + Vine Shield + Rainbow Armor. 25 gems.
Script Blessing, Resist Poison, Air Shield, Mistform, Soothing Song, Attack Closest.

Needs Con-4 and Alt-3, is a little easier on the fatigue, and will do bad things to non-poison immune troops. You can slap on a snake ring and spare yourself Resist Poison for another 5 gems, or additional reinvig/regen in a misc slot.

All these are achievable with relatively little research, can be mass produced, and the vine shield is the only thing that cannot be forged by the Sidhe Lords themselves (and you always have mages who can forge them).

Re: Troops vs magic, what am I doing wrong?
 

Don't forget that Astral Fires is MR-negates, which can really cut down its effectiveness. But yeah, the range is great.

Re: Troops vs magic, what am I doing wrong?
 

I also just discovered that it's precision 100. SWEET!!! :D I think I'm sold.

Oh, BTW, it's official, the AI is an idiot. I decided to try and see what would happen if I gave my communion master only two pearls. Yup, he blows both pearls on casting PotS, and doesn't have a pearl left to cast LotNS. So three pearls is the only way to go with this strategy.

Nice thing about pearls is that they're plentiful -- it's cheap to turn any other gem into them!