21 damage isn't high. (I'm aware that I'm extremely biased by the nature of my playstyle.) I don't bother with warlocks because they don't do enough damage. Adamantium veteran beastmen halberdiers, with no other buffs, do more damage than the great drake fire breath. Yes, the halberdiers are reduced more by armor, but... 60 production, with a prereq of 700 production, vs 900 casting skill and a prereq of 8000 research.
Its fine to say that putting more into one stack means it should be more expensive but not at the rate it currently is. I'll take my 15 adamantium beastmen halberdiers against your great drake. In fact, I'll take 135 beastmen halberdiers against your doomstack of 9 great drakes. And mine still costs less. Let's assume I'm right, that production is worth 20% less than casting skill. Then, I can actually get literally thousands of beastmen halberdiers for the same cost as 9 great drakes.
Obviously that's completely ridiculous - nothing can be converted nearly so neatly. You can't actually change research into production. You can't get that many adamantium cities pouring out halberdiers. Etc.
But the fact that the costs are THAT skewed, yet I can still even remotely even imagine the damage output from a single unit is comparable? Its a problem. Arguably the best way to win, is to focus 100% on city troops. Literally, ignore nodes, use alchemy on all mana you get, never build power buildings, and just go straight for fighters guild, alchemists guild, and war college everywhere and start pumping out troops, and swarm your enemy. In fact, this is what makes the AI so dangerous. This is the basis for their strategy.
One of your big concerns is that the AI can't build doomstacks like the human can. So reducing the cost on the higher end units will help the human. I disagree. Imagine how many great drakes the AI would get, if it only cost 400 mana. (I'm not saying that's a price I think is good, I'm just using it to make a point.) The AI is already great, GREAT at pumping out huge armies. (I recently fought an AI who had 86 hell hounds within 4 squares of his capital, in 1404.) If the units are cheap enough the AI will simply swarm the human with them. That's what makes lizardmen and common summons so dangerous. The human can make better doomstacks of low end troops than the AI can, but it doesn't matter because the AI has half a dozen stacks for every stack the human has.
I'd be interested to see what kind of costs would be found if units were compared against different tiers.
I don't necessarily think its a solution, but I think it would be interesting to have that information. We could see if there is a general trend for how much more expensive things should be since you can put more combat power into one stack.
Its fine to say that putting more into one stack means it should be more expensive but not at the rate it currently is. I'll take my 15 adamantium beastmen halberdiers against your great drake. In fact, I'll take 135 beastmen halberdiers against your doomstack of 9 great drakes. And mine still costs less. Let's assume I'm right, that production is worth 20% less than casting skill. Then, I can actually get literally thousands of beastmen halberdiers for the same cost as 9 great drakes.
Obviously that's completely ridiculous - nothing can be converted nearly so neatly. You can't actually change research into production. You can't get that many adamantium cities pouring out halberdiers. Etc.
But the fact that the costs are THAT skewed, yet I can still even remotely even imagine the damage output from a single unit is comparable? Its a problem. Arguably the best way to win, is to focus 100% on city troops. Literally, ignore nodes, use alchemy on all mana you get, never build power buildings, and just go straight for fighters guild, alchemists guild, and war college everywhere and start pumping out troops, and swarm your enemy. In fact, this is what makes the AI so dangerous. This is the basis for their strategy.
One of your big concerns is that the AI can't build doomstacks like the human can. So reducing the cost on the higher end units will help the human. I disagree. Imagine how many great drakes the AI would get, if it only cost 400 mana. (I'm not saying that's a price I think is good, I'm just using it to make a point.) The AI is already great, GREAT at pumping out huge armies. (I recently fought an AI who had 86 hell hounds within 4 squares of his capital, in 1404.) If the units are cheap enough the AI will simply swarm the human with them. That's what makes lizardmen and common summons so dangerous. The human can make better doomstacks of low end troops than the AI can, but it doesn't matter because the AI has half a dozen stacks for every stack the human has.
I'd be interested to see what kind of costs would be found if units were compared against different tiers.
I don't necessarily think its a solution, but I think it would be interesting to have that information. We could see if there is a general trend for how much more expensive things should be since you can put more combat power into one stack.
