The Shrine

Seeing Seeker's post on Utopia this is one of my interests that I never was actually able to play, since for a game like Utopia I really don't like the fact that you go to sleep and get attacked by 3 million ninjas that raped your town and you don't even have a chance to defend against it. I think a strategy game needs to have some graphical representation of the outcome, because if it's just 100% text you end up having limited ability to affect the game and then it becomes more of a empire building game as opposed to military oriented.

Of course if there's some kind of graphical interface you can interact that can further amplify the effect of being there versus not, so you got to have a few things going:

1. It should be very hard and time consuming to defeat a well-defended guy even if he is not there at all (assuming equal forces).
2. It should not be possible to defeat a well-defended position if the opponent is there (assuming equal forces).
3. There should be no tactical manuevers advantage in such a game.

No matter how well you design a game there's going to be an issue in that some guys who don't need to sleep or on a different time zone can attack you during 3AM where you'd not be able to actually play the game, so assuming you're not going to have a foolproof AI you've to expect that if you're allowed to move units around there might be some way to beat the AI (and if it's not possible then there wouldn't be a point to playing the game because the AI can do everything). So let's say you got some map with a layout that everyone can see. The player should be able to give the following commands:

1. Send unit to attack objective X, with one possible waypoint so that you're not always trveling the most direct route possible (otherwise it'd be easy to defend).
2. Once a unit is out the only change in objective is retreat. Once the unit is retreating it cannot do anything but retreat, though if it cannot escape it'll fight instead until there's an escape path. You can set to allow the AI to retreat automatically when your unit strength is below some level you set.

A unit should attack the objective and whatever else is needed to accomplish it, and once it's done it will either occupy the new objective or go home (if the objective is say, an opposing unit). Therefore if you choose to specifically attack an enemy unit X your unit will go home after that unit is dead instead of continue fighting. This means as a defender you've more options on how to control your units since even with this restriction you could say put 2 units to attack a range unit, after they're done they go home and then you send it out again. On the other hand for the attacker it'd be pretty stupid to do this because then your unit will walk all the way home after that objective is done. So, the attacker is limited to attack say the location X (and his units will attack whatever's in the way), and defender might be able to finetune at some level. Note that because you don't control the units, that means if the ranged units properly protected by melee units, then just because you told your units to attack the range units doesn't mean they'll somehow try to walk pass the melee units. Here the lack of human control is important, because otherwise you'll always be able to find a way to walk pass the melee units defending and hit the range units.

What this means is that to hit the vulnerable units you most likely need a manuever on a greater scale. You'd either need you or maybe an ally to take a much longer or at least different path to reach the more vulnerable units behind. That is good since you want a game that emphasizes on cooperation (otherwise you can just play single player). Again, all this should be available in some kind of graphical interface, and really you should be able to see the units fighting on it. How far you can see might be a function of say your spying/scouting level and some function of the enemy unit size/type (more of them = easier to spot, but maybe some units are harder to spot).

So such a game will turn into a lot of manuevering, but if you make defense relatively strong or at least time-consuming for the attacker then you should eliminate the case where someone was defeated while all the guys on your side was asleep. You can probably even have preset conditions like 'always assist if we have this many guys and they only have that many guys'. The AI that manages your troops should be able to do okay in your absence but should never be better than you actually playing the game.

Sorry, I don't have the time to reply in full at the moment (but I will).

I do think Utopia would be better if all 25 players in a Kingdom could maneuver attack forces. The way it stands, it is really a game of teams where everyone controls their armies independently; but ultimately it ends up working out a lot better than you would think on paper. The biggest flaw in Utopia is that everything does happen instantly, and it means that there are exact time periods when people have to be online. Good players can get away with logging in about 6-8 times a day for a combined total of 10-15 minutes of gameplay time, and manage to get a top 50 province; it is a matter of when you log on, not how long you log on for.

Tribal Wars, top players essentially have to have multiple people running their accounts, otherwise it would take exponentially longer periods of time to increase ones empire, and run it with the same efficiency that one ran it when it was only one town. This is why I actually much prefer Utopia, despite being a simpler game. Although in Tribal Wars, I once organized a powerful Empire which had people doing what they wanted except that they had to have at least 10 fully active cities (although this number changed) fully dedicated to the Empire. It worked well, and we ended up defeating much larger alliances to become the largest alliance (and growing) for several months before I quit due to the amount of time required to manage things (We had many smaller feeder alliances; people started to get fed up with my not being online 24-7 like some members seemed to be; so I quit on amicable terms - because I don't want to spend more than 20-30 minutes a day on games like these - preferably 5-10 minutes).

You're basically talking about games that have a unit loss-less situation so having one really good player can just dominate the whole game. Usually it's the result of both a lack of transparency on the game engines and the game being poorly designed. It'd be like playing Starcraft and you're the only guy who knows that siege tanks have a siege mode. Almost all strategy games greatly favor the aggressor so you might as well have one guy just playing your whole side doing all the attacking. The only reason people don't actually do that would be due to time limitations. In a game where everything happens instantly (like Utopia) there is far too little information for the average player to digest. It's like Master of Orion 2's strategy mode where the AI just figures out the outcome which always ends in one side being completely annihilated, and it uses a formula that you do not know, but even if you did, you still wouldn't be able to predict the outcome with any certainty. So if you're the one guy who truly understands the formula you'll be able to clownstomp your way to victory easily simply because no one else even has a clue if they're winning or not.

This is why there needs to be some kind of real time representation of what's going on. You should be able to see the units move around in real time like a game of Starcraft and you should be able to see that you're winning or not. Although Starcraft and its related games suffer from the issue of imitation, it is a necessary element if you want a game to actually evolve. You need to be able to see like this guy made this really awesome move and then you died, instead of just being told your kingdom was destroyed by 3 million ninjas. However it's important that you can only see this, as opposed to actually control this, because otherwise you'd have cases where you just lose badly because you were asleep or on vacation or just decided you need a break from the game.

Actually Civ 5 would make a good model of a defense-strong (relatively game). Let's say all you have researched is Oligarchy for both sides (Nationalism slightly harder to predict). A 33% combat strength bonus is good for at least a 3:2 kill ratio (think it'd be closer to 2:1 really). Your units can heal in your territory while the enemy cannot. You can hide units like Great Generals and artillery in cities where they'd not be vulnerable to counterattacks. So if the offense cannot give a new order beyond retreating, you can probably get away with having a computer defend for you. If you're actually defending you can send out a unit to target specific units (they'd have to retreat after that's done but given this is your border you'll quickly be able to send them out again).

Now to make sure people don't just turtle up the offense should be able to disrupt the economy of the defense, at the cost of losing more total resources overall. But defense really should be very easy. Historically most major battles are fought in the field, not near fortifications since it'd be very unlikely to win unless the odds are truly overwhelming.