New experimental version

[Arnuz]

Ehhh, anything you get from the ai is something you're profiting from their cheating bonuses from. Realistically, just getting the city, with no buildings at all, is a huge bonus. Don't attack little things so early, and you'll probably get at least one building. Amp towers etc are just the most extreme form of that but even sawmills are hugely important.

Anything that discourages super early attacks, and makes neutrals weaker, is a good game feature in my eyes.

Completely disagree. It might not suit you, but it's a different style, some people just like shorter games. Why should the game only cater to your style? both approaches should be possible.

You don't even consider what I'm saying: the razing of buildings is already delaying the player; giving sale money for the ruined buildings only gives you 1/4 of the amount you need to buy them; so that's a small help to the aggression that doesn't seem to tip the scale.

Try another random wizard, but make sure you have a minimum of 3 picks of retorts that work together

I know but that used to be the requirement to stand a chance at Impossible or completely overwhelm Extreme just a year ago. Something definitely isn't right with difficulties if I need to do that to stand a chance at Expert. Is Expert as strong as Impossible was now?

This is good. It means that your work is making the AIs better. As a consequence, you might want to tone down the resource cheating, rather than preventing good algorithms like the aggressive colonization of myrran.

Razing has definitely increased the difficulty for the human: even in the long term strategy one eventually needs to conquer cities, and when that happens - either sooner or later - the change affects the player. Similarly, if you toned down the cheating then less of the cheat resources would end up in the human's hands: the razing might be toned down a bit then, or the coin for raze idea implemented.

BTW Nagas are the shit now. All over the place. Sorcery lizardmen is even stronger - trying it next. Was the movement change too much?

[Arnuz]

PS in my opinion the early aggression by AI is also a consequence of their resource bonus. I notice this:
- you can't garrison as much as AIs due to the bonus. It's also a bit pointless to garrison cities very far from the frontline.
- other AIs only consider your cities as targets, because they know their garrison despite no scouting
- they immediately act opportunistic, whatever their alignment, and only towards the human given the garrisons on all the other cities

Basically the game is: how do I kill an endless stream of incoming units? Luckily, the silly strategic combat rules mean that if you have archers and walls you can deal a lot of damage without issue, but it seems... A bit off to play like this. It's even exploitable: while all the AI's troops slowly wade towards your back-line empty city, you can prepare for a fortress strike. It also gives an enormous advantage to undead creation, given the garrisons that you get in the AI cities. I can man all the cities and all the nodes with that.

Note that this is much better than before, when their troops remained around their capital. This change is fantastic, I'm just wondering whether we couldn't improve the situation even further

[Nelphine]

Ah, I don't consider the reduction of neutrals and conquering good because it makes a longer game.

I consider it good, because if in game one you find 5 neutral cities near you, and in game 2 there are no neutrals on the map, then any strategy based around neutral cities has a GIANT fluctuation in power. Game 1 you might win due to those 5 neutrals, while playing exactly the same strategy, you lose game 2. See all the discussions about sprites and early conquest with catwalk.

We want strategies to be predictable in how effective they are.

Similarly fighting AI. If you conquer cities from the AI and that is your primary means of getting power, then what you will find is that the higher the difficulty, the easier the game, because the AI cheats more, and so their cities are better. This is backwards. The benefit from conquering AI cities should be roughly the same regardless of the difficulty. The change to increase destroying buildings is based on difficulty, specifically to counteract the increasing AI cheating.

[Domon]

We want strategies to be predictable in how effective they are.

do we? or do we want to reward adaptability?

[Arnuz]

Similarly fighting AI. If you conquer cities from the AI and that is your primary means of getting power, then what you will find is that the higher the difficulty, the easier the game, because the AI cheats more, and so their cities are better. This is backwards. The benefit from conquering AI cities should be roughly the same regardless of the difficulty. The change to increase destroying buildings is based on difficulty, specifically to counteract the increasing AI cheating.

Then reduce the AI cheating: what's backwards is to leave AI cheating to this amount, and to have to raze buildings artificially, after that the game's become more difficult through AI improvement. Conquering, in a game such as this, is a big part of the satisfaction (well, unless one's "playing like a legal perfectionist", I guess ), so it's clear that making it worse detracts from the game's satisfaction.
Again you don't answer my point: sale value for razing (1/4 quick build cost) is not the same as not razing. It's 75% less gold, plus the building delay. How's that equivalent?

We want strategies to be predictable in how effective they are.

do we? or do we want to reward adaptability?

Well put!

[Nelphine]

Did my thing really come across as legal perfectionist? I hate auto correct.

I tried to say peaceful perfectionist, to compare to the AI personality and objective.

So, 2 parts: a) the actual ai is largely the same on all difficulties. The difference is the cheating bonus. Even if you reduce the amount, that won't reduce the need to balance the difficulties.

B) if you gained 1/4 the gold of the buildings cost, if you raze the wizards guild and the amp tower, you still get TONS of money. Far far more than you do currently. Do that on 2-3 cities, funnel all gold to your troop building cities - or do opposite of our draconian friend, and funnel it all into mama, allowing for more skill creation. Suddenly you get a massive boost because you can focus those gains - just as we know its a problem that the human can ignore garrisons and get strong offensive stacks much faster than the ai. This also would mean if you razed 6+ buildings, ypi could be getting over a thousand gold. Razing the whole city doesn't give you that much gold, so since there is meant to be an economic balance between razing a city and conquering a city, you'd also need to increase how much hold you get from razing a city. In late game, this could easily lead to getting thousands, yens of thousands of gold from razing just a few cities. This isn't a good move, considering getting 500 gold from conquering a city is already borderline overpowered ad it is - and razing a city already gets close to a thousand gold.

[Arnuz]

Well, there will be a lot more razing now, I can tell you that... Why keep a city that you can't defend at all close to enemy territory when you could sell all its buildings at once rather than one per turn? Then, the tens of thousands are gained anyhow. Rinse and repeat with the cities re-built from scratch with the always flowing AI resources and what's the difference? So do you suggest removing also the raze gold?

The difficulty is also given by the aggressivity of AIs, which is different at the various difficulty levels. Besides, late game is not an important consideration. You have a lot of resources anyway: ten thousand gold in y13 is still handy, but a lot less meaningful than one thousand in y03. This affects early game a lot more, and the issue of raised difficulty has been observed by others, I think that razing is part of the picture.

Furthermore, a new point came to me: the "using AI resources" argument doesn't hold with neutral cities. Why should neutral cities lose more buildings at raised difficulties?

(I was going by memory, you probably did say peaceful - my bad)

[Nelphine]

I dunno, I've chewed through 13000 mana in 3 months before, and that was only around 1409. I've probably done higher. Mana (and therefore gold) is extremely important late game.

So you choose razing all the cities. In current, you get ~600-800 gold per city for that.

In current, when conquering cities, I average razing 6 buildings and I get ~300 gold for actually conquering the city.

Assuming an average building cost of 250, that means your proposed idea would net me an additional 750 gold. So now I'm getting 1050 for conquering a city. To keep the current balance between razing and conquering cities, razing cities would need to be increased to ~2700 gold.

Right now, I have to raze ~13 cities to get 10000 gold. Under your proposal, I would need to raze 4. 13 is incredibly difficult in one turn. My last war with Merlin, I got 6 on the first turn; so I could easily have made 15000 gold in one turn. I feel that's excessive.

[Nelphine]

Aditinally, there's a realism point of view. Currently when you conquer or raze an enemy city, the gold you win actually comes out of the enemy wizard gold supply. Gold is not created during these actions. However, your proposal would mean that gold would be created. Somehow a destroyed building translates into a pile of gold.

[Seravy]

Why should neutral cities lose more buildings at raised difficulties?

They don't. Neutral cities have a flat 50% base destruction rate as far as I remember.