Realistic Game Models and the Learning Curve

[Mark_Everson]

Visitors to the forum have occasionally complained about Clash's complex game models (Economy, Government, Technology etc.). I've been saying for quite some time that not only would our decision to "go realistic" result in a richer game, but that it should also help with the learning curve. The reason is that if it works in the real world it should work in Clash. This is opposed to the case for a basically abstract 4X game, where you need to know the rules in detail, because you can't rely on a connection or result that is sensible in the real world to play out the same way in the game. Well, a recent review of Europa Universalis gives credence that at least one game reviewer thinks this premise is reasonable. In the "learning curve" part of the review it says:

quote: The game is not designed for the strategy newbie. However, utter realism makes easier to play. Think of things in real world terms and usually it's right.

Unfortunately, for a variety of reasons, EU doesn't look like exactly the right game for me. But if anyone else on the team, or a visitor cares to comment about its learning curve, I'd be happy to hear about it.

[axi]

Well, I am proud to inform you that I got Europa Universalis about a week ago. This is the last of my new toys, that have in part caused me to drop out of these forums (before that it was Baldur's Gate II, Half-Life and Heroes of Might and Magic III).

EU is really a great game. By concentrating on a specific historical setting, it has managed to provide us with a much deeper and more realistic gameplay than any other strategy game until now. These guys have performed enormous historical research, trying to simulate the historic setting of the era. The game however is lacking in a few major areas:

1. AI. Although the diplomatic AI is very competent at forging alliances and declaring war when and where you're most vulnerable, the military AI sucks, just as bad as the civ AI does. Attrition, supply lines, sieges etc are very good rules to prevent civ-type warmongering from the part of the human player, but the AI just seems to just overlook their existance. It creates monstrous armies and sends them fight on the other side of the continent, into areas with no supplies, losing most of them due to attrition. It uses 200.000 men to siege a fortress and suffer maximum attrition, while only 20.000 could do the job as well and not lose a single soldier. Single-province states, overlooking the fact that if their fortress falls, their enemy has the option to completely annex them, send their armies to siege some far-off province, although the enemy is already marching towards them. Invaluable explorers are sent to sure death across the Atlantic, trying to get as far as they can before their last ship sinks, without ever considering returning to port from time to time.

2. Interface. The game interface is not so well polished out as it could be. There are 8 different aspects of the map that one can see, but not all are accessible in the same way. The Religion map in particular is hidden as deep as it can be. Reorganising your armies is very problematic once you're on foreign land. I never understood why there are only infantry sprites on the map and no cavalry ones. The tech cursors are very inaccurate and fine-tuning is quite hard. The game is also quite a memory hog, since it performs extensive computations for all of the (up to 90) nations in the game.

3. Technology. In an attempt to counter the "steamroller effect" which occurs in civ, where technology is only dependent on spending so that the bigger civs obtain a huge technological edge, thus crushing the smaller ones, EU has introduced a system where tech acquisition is cheaper the less provinces your state has, while other bonuses like neighbor and monarch bonuses remain absolute, thus giving disproportionate advantage to the small states, whose armies and merchants become invincible after a while. The same happens with pirates and rebels which tend to be much more advanced than the armies and navies of the imperialist nations. It is the big AI nations that really get screwed, since the human player, usually finds a way to partially counter his disadvantage.

As for the learning curve, I have to say that it was much easier for me to begin competently playing EU than civ2. This is partly due to the fact that civ2 was the first strategy game I ever played, while now I am a really experienced strategy player. However one surely feels much more at home in EU, since the setting is much more familiar. The game also practically runs itself, specially if you're not a colonialist nation, since there is no micromanagement in the economy and the military units (one has always relatively few armies, because of the bonuses and specialties attributed to leaders and because, against the big AI armies, only equivalently big armies are effective) and there are relatively very few things to build. The really difficult stuff, like the combat system or the trade system, or where is the best place to build a colony can be adequately administered following the common sense (it is common sense f.e. that cavalry can't do sieges, but cannons are way better for the job). The only really steep part in the learning curve is diplomacy, which follows the rules of the era and where one can really do blunders (that's why I always save before sending a diplomat) and perhaps missions, where one needs time to start making valid assessments on his own capabilities (twice I said I would annex Algiers in five years and twice the time passed without ever making a serious attempt). Inflation remains a mystery for all EU players, because their game model is based in some irrational monetarist theory (putting your monthly income to the treasury increases inflation, while doing the same with your yearly income doesn't) which is unfamiliar to us commoners. At least I live in a country where inflation has reached 25% only 10 years ago (now it's less than 4%) and I know some things about it; what would a German with constantly less than 2% inflation know? As for the other stuff, AI will be always AI and there will be always tricks against it, which have to be learned. The manual is a piece of very nice prose, full of history lessons, but it won't help you much because it's disorganised (it even lacks an index) and it won't discuss the inner game mechanics or strategy at any satisfactory degree (What manual would ever do that? Exploring the game's possibilities and discussing your findings with the other gamers is by far the most satisfactory part of the gaming experience).

What a TBS gamer has to learn anew is time management, not in the RTS sense, but synchronisation between units and events, so that there is no dead time and opportunity windows get exploited (fe when multiple sieges take place and one takes some provinces just before losing some of his own, one can get peace with favorable terms). During wars, several events may happen in a single day and things are hard to monitor unless one uses pause and plays at the slowest speed. So there is is a sense time micromanagment; playing slow and pausing often may take forever to play a full game (I played one week for just 20 out of the 300 years of game time) but it has great benefits. This distinguishes the EUgamers in two groups, on that does what I do and another that plays fast, having played the Grand Campaign with many nations many times over already. I guess this is something we'll see in Clash too. Personally I prefer this sort of "stopwatch" games (games of all sorts, like Caesar 3, SimCity, Baldur's Gate) even against TBS games.

The bottom line is that EU is a very bold step for the games industry towards the right direction in strategy gaming (it isn't by chance that this is not an American product), but still it is just a very small step. The gameplay is not nearly as deep as it could have been or as we would like it to be (economy, technology and combat are still too generic) and also a big consession is made, that realism and depth always have to be inversely analogous to the width of the historic game setting. We in Clash should strive to break these limitations, by creating extensive and versatile game models and providing wide possibilities for customisation and scenario creation.

------------------
"In a time of universal deceit, telling the truth is a revolutionary act."
George Orwell

[Mark_Everson]

Thanks for the detailed review Axi!

And here I was thinking you weren't doing those promised refinements on the infra model because you had too much Work to do...

[Lord God Jinnai]

I also got EU and i haveto say it is very good overall (9 out of 10).

Since Axi has given a good picture of it, i'll only elaborate on my points.

The graphics while lacking in some areas, esp as Axi pointed out with Cavalry units, was in general to the point while still appealing. I don't like games that get too detailed if it sacrafises other qualities.

As to the movements of the Units, they did a good job in showing how it takes much longer to explore new areas as opposed to already mapped areas. While there are limitations because of how the game is in that you cannot gen an inaccurate (only incomplete) map, i think they found a good way around it. The map when viewd without most of the special things happening actually looks like a map from that period. One problem with troop movements as I mentioned earlier is that if you accidently forget to deselect the unit and click the wrong mouse button on another province, depending on the scenerio and other factors, the unit will change directions and if you correct it, you must start your march over again!

The use of pirates was done relatively good and the ability to deter them, although the lack of privateers for poorly developed naval countries or landlocked ones can be frustrating, but that's a minor point.

As far as the technology goes, i haveto give that only 5 out of 10. Mainly because although they did a good way of simplifying it for the limited time in the game and based on historical accuaracy in linking with religious regions, this has stifled alternative histories. Although in their behalf, this is from the point of Europe, not the world in general. Still...

Finally the last paragraph leads to their implimentation of religion which IMO they did very well for Christianity and Islam. With historical events and whatnot. I do however, haveto disagree with listing every other religion as Pagan. Although, from the 2 other religions all others (save to some extent Judaism) would be paganistic, the Islamic nations had much interaction with Hiduism and Budism and would have known what they were. This also limits the scenerio possibilities such as if the Moguls had conquered most of Europe, etc.

Finally they did very good with the engine by linking historical events that wouldn't fit the norm. into the engine while still allowing for the fact that EU is a game, not an exact representation of history. This has been in some ways improved by fans in the Improved Grand Campaign Scenerio (IGC).

The thing that made me really like this game is it is not a mainly military conquest game. That is important, but diplomacy and internatal stability are also of great concern. Sure you can expand yourself a few provinces, but if you do, you might have too many revolts to handle as well as make a few enemies by conquering that small inncoent province who just happened to be in an alliance with Russia and Austria.

Again, this is a very great game and extremely difficult, not for a beginner, but those who crave challenges

[axi]

Mark: Back then, I really had work to do. For the past few weeks I've been goofing off. Alas, work awaits me around the corner - in a month or so I may be as busy as one can be. However I don't think that my absense has hurt Clash in any way, did it?

As for the refinements, the infra spreadsheet is revamped (I had done some work back then), but I got stuck in the fact that public investment would require some optimisation method, which is more of a programming issue and can't be tested in a spreadsheet. I can send you what I've got, but it won't do, unless you assign a team (containing a programmer) to start working on infrastructure again.

I am available to provide any explaination or help. What I am lacking right now is enthusiasm and initiative (This is probably my own personal problem and it has to do with other things in my life too, so don't worry about it.)

Lordy: I used my usual civ tactic, of letting them declare war and then crush them ruthlessly, but my badboy got to 39 and all Europe hates me. As long as they daren't declare war I don't mind, but if one does, then everybody will be after me. It's hard to be Spain after all...

------------------
"In a time of universal deceit, telling the truth is a revolutionary act."
George Orwell

[Mark_Everson]

Hi Axi:

No, your absence as it turned out didn't hurt the overall progress in any way! Don't worry, I would have let you know if it was a show-stopper... The reason infra lost priority is that F_Smith, who was doing the graphical stuff, basically couldn't get anything done due to real life issues, so I had to take over that coding also. (Couldn't have a military demo that does much without being able to see and move the TFs)

So although infra coding was my next priority quite a while ago, its still realistically about a month from happening . Hopefully it won't get further away!

On your being held up due to lack of optimization... If you get your enthusiasm back, I'm sure you'll be able to do something very simple in terms of optimization that would allow you to move forward. It sounds like you just need some goof-off time, so have fun with it .

[4VAlien]

if all options would have a reasonable value, so that you can start playing by just doing the basic things (like standard civilizations actions) and if you make that clear in the manual people will eventually look in the deeper options. My idea of an ideal learning curve is that you start playing without reading the manual, and then later you get at a point where you say: i manage the basics, nou let's check the manual to see how i can tweak my strategy. I really hope that we can make this game playable out of the box without reading the manual.

[Mark_Everson]

4VAlien:

Yes, that's my thinking too. And you should Never have to understand areas you aren't interested in, just issue overall orders and have them implemented by the AI.

[Lord God Jinnai]

While I do complain about people who get frustrated at not being able to something and I ask them, "Did you read the manual?" and they say, "No." I do realize many people are very eager when they get a game to jump right into it. We could instead do like most games and have a tutorial scenerio for these people and I personally believe in having the manual all in the help file with hyperlinks for easy access.

[4VAlien]

the learning curve is important to me.. i agree to jinnai that i hate people who don't read manuals (and think i know ALL about windows cause i'm studying computer science) but you can't expect people to study a few hours on the manual before being able to start a decent game.. People will decide to play the game only if the think it is fun, if you HAVE to read a manual it's no fun, so they'll take another game.

[Gary Thomas]

One of the problems with the way programs are written is that the writing of the code and the writing of the help are disassociated. They are done by different people, widely separated in time.

Maybe we could start a new system. It is not impossible that every time a coder writes a user interface routine, involving a JTextField, or a selection in a JList, they could write, in the JavaDoc comment, a draft help message, with a suitable JavaDoc tag. Lacking other options, @todo would work. This means that, right from the start there is an online help basis available.

Just a suggestion...

Cheers

[This message has been edited by Gary Thomas (edited April 13, 2001).]

[Mark_Everson]

Sounds reasonable. Can you crosspost this to the coding discussion and in the Demo 5 interface thread? F_Smith was already making some use of tooltips. Can't those be used directly for what you're talking about?