harmonius_ harmonius_

Trained units are useless

Trained units are useless

Trained units are useless. Sovereigns/champions are much more versatile, strong and cheap, than trained units. In early-mid game you can't train useful units, because you haven't enough technologies and resources. Trained units lvling is poor. When trained units gain a level, they get more health and a bit accuracy. Health is useful for melee units, but ranged units need more attack rather than more health. Really all trained units need some attack bonus per level. Weapons and armors in early technologies should be more powerful. Now the only useful unit in early game is pioneer. Trained units should be improved dramatically. It should be a hard choice between training pioneer and training militia in first turn.

10,973 views 39 replies
Reply #26 Top

Uselessness of trained units leads to uselessness of cities. Why do you need to develop economy? To build an army. If army is useless, ecomony is useless too. Now cities are only for mana income and for researching several technologies for champions.

Reply #27 Top

Here's a weird solution-alternative to this problem:

Heroes (and sov) gain XP to the amount of XP trained units in the battle gained together. Thus the bigger the army hero/sov commands, the more xp he gains. Fighting solo or only with other heroes would give no XP at all, but might still give you a victory. For every friendly unit died in the battle there's less XP to gain so you have more incentive to keep the trained ones alive.

And start the game with one trained unit + sov.   

This would have instant affect of making trained units desirable and heroes to level slower.  It would require quite a lot of balance fixing to make it work, especially to avoid making all trained units to be tanks.

Reply #28 Top

Quoting Trojasmic, reply 25
I had a problem with immortal champions because I would have to kill the same 5 AI champions over and over again.  In the latest beta, I think they've put a lot less champions on the map.  That solved my problem of having to kill too many AI champions over and over, but sadly, I don't have as many champions either.  I think that was their way of solving the uber-champion vs. normal trained units problem = put less champions on the map to begin with.
End of Trojasmic's quote

I just finished a game as Magnar.  I actually did pretty well.  I finished at Ereogg level.  Pretty good for me.  I probably had about 10 - 12 champions by the end.  Some I picked up towards the very end, becasue I happened upon them.  I really only had about 4 or 5 that I used.  The rest were just sitting in my cities.

It was a medium map on normal with 4 opponents.  Two of my champions were just stellar.  They are just way to powerful.  Either of the two champions could just level armies, Dragons, enemy cities, you name it.  He was a level 7 or 9 champion that also came with a mount.  He had three magic schools.  He got to level 15 or 16, and was just awesome.  Fireball was devastating when he attacked a city, it just leveled entire sections of troops.  Flame Dart any Sovereign or Champion, usually one shot them, and them fireball them, then flame wave.  Burning Hands if they happen to get close.  I think I 6 fire shards, so they could just do crazy amounts of damage.

Champions like him need nerfed.  He could have easily been any other Sovereign in the game.  Was clearly more powerful than a couple that I ran into.

Reply #29 Top

Guess it would be usefull if people also mentioned the difficulty level they play on. I'm always on expert, dense monsters, large map, 8 players, rest normal. I need units, that's for sure. But I don't need them desperatly too early on, because there are lots of monsters/cratures that are very easy to take out with 2 low level heroes combined.

Reply #30 Top

I play on expert and ridiculous and I find I use troops every game and they help out in several ways.

 

In the early game it's often helpful to train a couple of spearmen groups to escort your sovereign - club militia are OK too. They help fledgling sovs alot in the early game although I can get by without them with a tank sov build - with any other build they're good to have early game. Slaves are especially good as you can rush out loads of these in no time and they're totally expendable - once they get low on life, you just sac them for mana.

 

They are also good to garrison bordering towns and outposts, which frees up your sov to explore the wilds. They are especially good in this role if you also have spells like freeze or pillar of fire to weaken any attackers.

 

Once you have access to a few techs and resources you can also train some pretty good units to support your champs too, especially if you've also picked perks like trainer and tactician. Mage units are particularly good if you've levelled up in tactician as it helps to overcome the initiative penalties and they pretty much bypass armour. You can train up some pretty decent units using the early techs but the late game techs allow some very strong units with high attack and defence. Normally by the mid-game I'll have champs and units working together. OK you could have multiple champs in a group but then they start getting penalties to xp so I don't like doing that much.

 

Units are really bad against things like overpower and fireball but they are good in other situations and benefit much more from spells/items like courage than a champ would.

 

Also champs require a fair amount of investment too, firstly you have to hire them, then buy them equipment to top up what you've scrounged, and also you have to spend mana enchanting them and using their spells - so the resources they consume are comparable to the time, metal, and crystal invested in units.

 

My view is that units are quite handy but on easier settings you can get by without them.

 

Reply #31 Top

Trained troops are just weak until very late in the game.  Unfortunately by then the Champions can cast just devestating spells.  Does any else think it is odd that a flame dart does 120 points of damage to and entire squad of people, and they all die.  Shouldn't  a "dart" just hit one person in the group.  Fireball, okay, i get it.  Its big and does area effect, but a dart.

If you have two other champions slinging flame dart, fireballs, flame wave--trained troops are pretty worthless.  You can attack a city with 6 garrison troops, and 4 and 5 troops stationed there, and just level them in 3 or 4 turns.

Reply #32 Top

@ Kinghobbit

You do realise that units can be equipped with fire resistant gear right? You also realise there's spells like nature's cloak and protection from fire which are specific counters to fire magic strategies too? The tools are there but the problem is that the AI is only as good as it's programming and so it doesn't take advantage of these things. Over time Brad will programme in more advanced AI strategies and you won't be able to get away with just flamedarting everything, however Elemental is a complicated game and you will only see gradual improvements over time - Brad's likely to update the AI often post-release too.

In GalCiv2 the AI looked at what weapons you were using and invested in specific counters to it however GalCiv's combat mechanics are much more simplistic so I imagine it was much easier to code - there are many, many more factors involved in Elemental's combat system so it must be much much harder to programme. There are many strategies I use that the AI never does; it's effectively playing with a severe handicap ATM so on the harder settings it gets research, resources,hp etc bonuses. I should imagine that over time Brad will reduce the bonuses as he is able to programme more strategies in. Also as the AI programmer, the AI is dependant on Brad's own ability to play the game; there may be tactics that he is unaware of so he hasn't programmed the AI to counter them.

With Galciv it was so much simpler - human is investing in lasers so the AI invests in shields - Elemental has about 7 different damage types, plus loads of different abilities, defences, and spells.

However I'm sure one day you'll be playing as Magnar and the AI will respond by giving it's invasion forces dragon cloaks and fire rings plus champs with counterspell - it will be a good day when that happens but I think it's some way off yet.

Reply #33 Top

I've found that maintenance cost is another reason trained units are prohibitive.  Perhaps if they cost a little bit less, we'd be more likely to generate large armies of trained troops.

Reply #34 Top

Quoting DGB246, reply 33
@ Kinghobbit

You do realise that units can be equipped with fire resistant gear right? You also realise there's spells like nature's cloak and protection from fire which are specific counters to fire magic strategies too? The tools are there but the problem is that the AI is only as good as it's programming and so it doesn't take advantage of these things. Over time Brad will programme in more advanced AI strategies and you won't be able to get away with just flamedarting everything, however Elemental is a complicated game and you will only see gradual improvements over time - Brad's likely to update the AI often post-release too.
End of DGB246's quote

I do not disagree with that it is a complicated game.  My point about flame dart is it should only hit one thing.  Not everyone in a group 5 or 7.  It is a dart.  I actually want to make it harder to kill groups with it.  I thought Nature's Cloak was only used on hero's.  Protection from fire is great.  I even had it cast a couple times.  Didn't matter, it took two fireballs instead of one.  They still couldn't reach me.  One of my champions was level 5 in three different schools of Magic--Fire, Water, and I think Death.  My sovereign was level 20 something, and the other two were in the upper teens.  With 6 fire shards, and equipment that increased damage and reduced mana cost.  I had 4 or 5 water shards so Blizzard would have worked just as well.  The troops were getting vaporized or i could have just as easily frozen them all to death.  If I thought a city was going to be a problem, I had a Shooting Star.  I just never used Blizzard because Magnar was a fire guy.

I truly hope and believe that the AI will drastically improve.  That is not the point I was making.  The problem with trained units, is they are minimized by how powerful the Champions have become.  I am fine with the Sovereign being super powerful, they are supposed to be.  But most, not all, but most of the Champions in the game can become just as powerful as any of the Sovereign's.  That is the part I disagree with.  I didn't need my Sovereign to do anything.  The two Champions I had with him were brutal, all by themselves--at least versus trained troops.  Champions need nerfed to make trained troops more useful.

Reply #35 Top

Quoting Trojasmic, reply 34
I've found that maintenance cost is another reason trained units are prohibitive.  Perhaps if they cost a little bit less, we'd be more likely to generate large armies of trained troops.
End of Trojasmic's quote

 

I agree with this statement entirely. It is bordering on a poor idea to build troops at all until very close to mid game simply because of how much they cut into production and income.

 

Unit construction costs fairly accurately represent the time spent training and equipping a group of soldiers.

 

Unit maintenance costs should represent the wages that these soldiers earn, with a tiny bit of equipment upkeep tacked on top. I mean something like 95% unit wages (per "model" cost) and 5% gear upkeep costs.

 

For example, a unit of 3 should cost .6 gold per turn, and a unit of 5 should cost 1 gold per turn, and a unit of 5 mounted soldiers should cost 2 gold per turn (horses gotta eat too!), then tack on something to the tune of 1% unit production as "equipment maintenance" costs and call it good.

 

Of course, those numbers are all "for example", but the idea that putting 5 men in chain is fully twice as expensive in terms of wages as five men with sharpened sticks is a little silly. 

Reply #36 Top

Well unit maintenance has gone down alot really and it's quite easy to get a decent gold income in the early game due to easy access to markets, bell towers and town halls. Also each champ now costs 1 gold to maintain so this seems OK to me.

 

@Kinghobbit

Getting your dude to level 20 plus having other tough champs too would be really hard to beat and this is where the AI needs to support it's units with it's own high level champs and invest points in the counterspells and spell mastery - currently the AI seems to level up all its champs as assassins or warriors but I think the AI should look to get a counterspell either through path of the mage, fire 3, or water 3. I think good champs/sovs are key and the AI is getting better at grabbing xp for it's guys though not necessarily choosing the best perks or obtaining the right equipment (it seems to have trouble using the store). Even so units can really help champs and I typically have mage units escorting my champs if nothing else. The two supplement each other really

 

BTW cloak of nature is castable on units but the AI never takes advantage of this.

 

I'm hoping that at some point the AI will look at your best champs/sov, identify their strengths, and then actively work towards countering them.

If your Sov is leveled up as a swordsman, the AIs should use chainmail. Otherwise, maybe they should just stick with leather armour, which is more economical.  

If your Sov is running around with stoneskin and a bunch of earth shards then the AI should put together a force that uses hammers.

If your Sov has high elemental resistance, then the AI should avoid training troops that use those damage types

If your sov is an air mage, the AI should place a lower priority on ranged units.

If your sov is good with damage based spells/debuffs then the AI should look at picking up elemental resistances, magic resistance, and poison resistance - whichever counters the specific spells. It should also try to train up/obtain a dude with countermagic and good spell mastery (that guy accompanies the strike force and is given alot of defence e.g. fire resitance if needed)

If your sov keeps using strategic spells when enemies enter controlled territory then AI should increase army speed (e.g. stacking tireless march) and attack outposts & cities from outside borders in one turn

If your sov is far away questing and leveling, then the AI should level up it's own guys with a sudden, surprise alpha strike on your cities - it should place it's invasion force just out of sight and then attack mercilessly - at the moment it gives you plenty of warning. The AI should not let you just go questing and leveling up - it should either do the same/alpha strike you/both

If you're attacking the AI, it should use strategic spells like freeze on your invaders followed up by a strong attack when your guys can only move 1 (ranged units are really good for this)

 

Currently the AI doesnt do these things so it's playing at a severe disadvantage and this is one of the reasons units feel underpowered - the AI doesnt really think about checks and balances or combined arms it seems

Reply #37 Top

A lot of the problems with trained troops will go away once tech trees and progression are more balanced.  The reason that champion can completely destroy the massive army of trained troops is because champions have access to much better weapons and armor in the early and mid game, while troops don't get chain until mid-late game.

I might have a champion with a broadsword, 1 piece of plate, and 2 pieces of chain by turn 30.  It's turn 150* before my trained units start to get the same equipment.  Stick a non-caster champion with a spear and leather armor and see how well they do against an army of trained units.  Caster champions, particularly fire / death caster champions, are awesome in strength, so no comments there.  The previous poster commented on the AI using resistances, that's probably the solution.  Maybe add to it the ability to re-equip units so it can rapidly change their equipment rather than have to train whole new units.

 

*numbers are estimates - I'm going to start paying attention to when I get those techs and when my champions get powerful equipment.

Reply #38 Top


I assume you mean when you are down to just one opponent.  Unless you set up all for opponents on the same team.

Quoting Trojasmic, reply 34
I've found that maintenance cost is another reason trained units are prohibitive.  Perhaps if they cost a little bit less, we'd be more likely to generate large armies of trained troops.
End of Trojasmic's quote
  This is also an awesome point.  Trained units are expensive.  They also stop your build queue from produces another building.  So they hurt a lot for what you get.  With the number of Champions running around and how powerful they become.  A better tactic for the AI would to do the same thing we are doing.  Make a stack of Champions and Sovereigns and level them up.

Two mid level champions can defeat the militia of even a very large "Fortress" city.  Trained troops may help, but usually not enough.  The game is Champion based right now.  They become very powerful, and do so much more quickly than trained troops.  They are cheaper to get, cheaper to maintain, and cheaper to upgrade.  I understand what you are saying, but right now, the way the game has set up Champions, is completely the other way.  And from Frogboys post about Champions, I don't think this is going to change.  He wrote a journal about how Champions need to be powerful.  He could change his mind but it doesn't seem to be trending that way.  I can't find it, but it was a while back.

Reply #39 Top

I think alot of this will be fixed in the balance phase and with further AI work. Yep, champions can become godly but it takes time and you can do good things with units too - in previous beta versions units were OP (clubman rush?) and I'm much happier with the balance now than before