Complexity FTW? » Forum Post by Wintersong

Polistes

Join Date 02/2008

+19

Yes complexity for the win. They announced this game early enough that I THINK that a lot of ideas could still be implmented fairly easily and also dont forget expansions!

What I am thinking atleast this is how i see it is they are kinda in the planning stages, they want to get as many possible ideas and whatnot now and are most open too new ones.

However there will come a point where they will be in more of a developing stage where they will not as readily take ideas and stuff and instead will attempt to implment what they have gathered from this stage. We still have a year and a half before final release so thats is seemingly lots of time to get stuff done and I have faith that SD will do a suburb job, even if I find their art a little cartoony.

Successfully updated karma reason!

Solam

Join Date 03/2007

+16

All ideas when nicely put like yours are welcome here.

I too have laughed at some of the suggestions. However sometimes it's not really the outrageous suggestion that is important but the idea that it brings forth. Some of it can be used even though it might not be the original idea.

I can just picture the Wrights brother telling people their idea of a flying machine. Most people must of thought them nuts. And yet....

I think Stardock is a great team and will use what is logical and leave behind the illogical stuff.

Keep those idea coming. The more the better I say.

Successfully updated karma reason!

Wintersong

Join Date 07/2006

+235

I know it's fine to people say what they think about what they would like to see in the game and/or what they think would improve the game (not just for them, obviously). And altough i understand that Stardock benefits greatly from such input, even if they have to discard most, i feel that sometimes some of those suggestions feel too big (and some ignore that game has other parts that should receive attention too). Innecesarily big.

I don't mind those suggestions, of course. But it worries me that some people can get carried away by the ideas and then get dissapointed and all the usual stuff. It's natural for it to happen, but i would like to think that people can understand that we will get a good game, but that i won't be "the game of our dreams" (aka exactly as we want it to be). And that some of those ideas (some of them quite good), might not get used not because they are bad but because it cannot be for one reason or another. Not a matter of just logical/ilogical but of resources/no-resources. Hmmm Hey Frogboy, i'll make the spanish translation of the game (a good one, mind you, and "for free") in exchange of amazons and tigers in the game.

Successfully updated karma reason!

Philocthetes

Join Date 10/2003

+153

I used to tell my students that what I demanded from them was reasonable, but that didn't keep me from having boundless expectations for every stupid little essay assignment. I wanted a genius to make me feel stupid for having misunderestimated her. That's roughly how I feel about 'managing expectations' for Elemental.

On the complexity point, if I really got a cake and could eat the whole thing eventually, I'd get a game that was aimed at a 128-bit future when you looked deep under the hood but had good enough virtual staff for the single player to ensure that you could treat the whole thing like a nice visit to a 5-star resort. In other words, I hope that the major improvement for this new engine will be some sort of (hopefully 'trainable') AI stuff that assists the player.

At the simplest settings, a player would delegate a very large number of decisions to ministers, officers, courtiers, etc., each of whose basic settings are fairly concise and easy to change. But each of those player-side agents would be running decision trees that also had UIs that a different game option and/or difficulty setting would make available. Want speed in any given part of the game? Delegate to your ministers or field commanders. Want a TBS game that lets you think about your Tolkien obssession while you play with an insanely big collection of virtual miniatures? Keep delegation to a minium.

Fantasies of outrageous dev time and talent aside, I gotta give a nod to the complaint beginning "If i fight in the desert, how will the clothes and armor of my troops affect their morale and health?" I'm interested in the notion of adding some RPG 'juice' to MoM's de facto successor, but not at the expense of making it a sickly cross-genre mule instead of another great TBS game from Stardock.

Successfully updated karma reason!

I am generally in favour of simplicity instead of stuffing more and more layers of cool stuff in. Consider that chess is very simple, but no-one ever really masters it completely. By contrast, mastering a very complex game is often a matter of finding a strategy that results in a positive feedback loop (A increases B; B increases A; recurse), and that leads to repetitive play and gentlemen's agreements to ban particular strategies that have been played out too many times before.

On the other hand, I am very much aware that MoM was the very definition of shoving loads of cool stuff in, yet worked extremely well and was fun. Way too broken for multiplayer though. I also quite enjoy having content there to discover.

In the end, my feeling is that adding things should always be done with an eye on how it will enhance the core game mechanics. I'm sure that Stardock have their own vision for the game, and know well enough that if something doesn't really fit with that then they should rip it out and save it for another game.

Successfully updated karma reason!

Good rant

I agree on most here. There are a lot if ideas here on the forums that gets way too complex after a while. But I also think that is completly ok, I enjoy the many different ideas I see here, even if many of them are impossible. What is really important is that we are aware that there are limits to a game, and that we accept that. Hopefully noone will get angry because a certain feature wasn't implemented in the game, but I think most people here on the forums are smart enough to understand limits.

Successfully updated karma reason!

Quoting Solam,

reply 2

I too have laughed at some of the suggestions. However sometimes it's not really the outrageous suggestion that is important but the idea that it brings forth. Some of it can be used even though it might not be the original idea.

Thats the truth of it. At the end of the day no matter how much detail I give you, all I can really say is "things could be better". Now if anything that we talk about here in the forum positive OR negative goes on to inspire a working idea in the mind of someone who can change things. Well then the system works.

If you only sit down and try to think up an idea, you will never be able to come up with anything unless you have a need. Sometimes your idea can fill the need of someone else. In the past a bunch of people came together to try to find a way to measure human body temperature in a convenient manner. And you know what we ended up with? Rectal Thermometers. Undeniably useful, but hell I wouldn't have invented them.

Successfully updated karma reason!

Wintersong

Join Date 07/2006

+235

Quoting GW Swicord,

reply 4

In other words, I hope that the major improvement for this new engine will be some sort of (hopefully 'trainable') AI stuff that assists the player.

At the simplest settings, a player would delegate a very large number of decisions to ministers, officers, courtiers, etc., each of whose basic settings are fairly concise and easy to change. But each of those player-side agents would be running decision trees that also had UIs that a different game option and/or difficulty setting would make available. Want speed in any given part of the game? Delegate to your ministers or field commanders. Want a TBS game that lets you think about your Tolkien obssession while you play with an insanely big collection of virtual miniatures? Keep delegation to a minium.

I find myself never using those assistants we get in games like Civ IV or Gal Civ II. And not because i'm micromanaging all the time (which means that in multiplayer i would suck, hahaha). So if this game can create a system that appeals me and makes me use it, that would be something that no other game has done before.

About ideas that fill the need of someone else, i really hope that many of the ideas mentioned in this forum can be used somehow because they are very good. Although i cannot see all of them used to full detail.

Successfully updated karma reason!

I know its not much info, but I like in Fire Emblem how there are 2 options for viewing combat statistics. A 'simple view" that shows only chance to hit, damage that will be dealt, and critical chance. There there is also a 'complex view" that shows many statistics so you can do the math yourself as well as see what might effect special abilities and such.

+1

Successfully updated karma reason!

Quoting landisaurus,

reply 10

I know its not much info, but I like in Fire Emblem how there are 2 options for viewing combat statistics. A 'simple view" that shows only chance to hit, damage that will be dealt, and critical chance. There there is also a 'complex view" that shows many statistics so you can do the math yourself as well as see what might effect special abilities and such.

That would be NIIIICE.

Successfully updated karma reason!

Complexity can easily equal problems, especially with feature creep, poor documentation, poor coding, etc. Now, there are ways around this. For instance:

Start off small and scale big (either through expansions, product patches/enhancements, or training scenarios)
Slowly introduce complex elements
Allow users to easily customize content they want to see and to the extent they would find helpful
Do a lot of interface testing and mock-ups
Include the ability to mod software (hopefully easily)
Do complex things in a way that "make sense"
Provide information in layers/drilling down, along with contextual help while using the interface
Give good examples of what is possible and good strategies along with them
Base doing stuff on the way things work in the real world, minus un-fun/overly detailed variables

Wintersong, you have a very legitimate complaint. I think the early betas will flesh out different playable options before SD gets too far down the programming path, so that by the time this goes 1.0 that it will find a good balance.

I think it's important right now for ideas to be thrown around since the game isn't even in beta yet and even bad ideas can spur good ones. It will be a lot easier to get major concepts and ideas included early on than trying to incorporate them later in the game.

Good caution, though. It would be really sweet for SD to post a list of "here are the major features/concepts we're thinking about- discuss"...

Successfully updated karma reason!

* Start off small and scale big (either through expansions, product patches/enhancements, or training scenarios)

Which is the definition of feature creep.

It's much better to plan for features from the beginning (when possible) to integrate them in the best way then trying to add them on top of an allready existing system.

Allow users to easily customize content they want to see and to the extent they would find helpful

Which makes a game almost impossible to balance...

I agree on your other solutions.

----------------------

"If i fight in the desert, how will the clothes and armor of my troops affect their morale and health? And how will the weather affect the result? How will the status of the war affect their morale? And the distance to our own territory? And what happens with my supply lines? If my opponent tries to attack one of my soldiers in difficult terrain, which is the chance of the enemy losing balance and...?"

Unnecessary micromanagement is a big no-no. No one should forget the uber mega maps that Frogboy likes to talk so much about. If i have to micromanage the breakfast of each of my soldiers just to make sure that he is at 100% of fighting capacity...

No no. Bad way to think about stuff.

There is no inheritent problem in adding complexity (or better said, detailed computation/simulation of certain aspects). It's all in the presentation and UI of things.

Let's say your units are in a desert. Now the game computes their efficacy in battle on the basis of the time of they day (how high is the sun?), amount of clouds, armor they use, how much water they've left, their race, etc. and as a result presents the player with a simple number (or bar, or any other graphical representation) that say, that unit is at 50% efficacy in this terrain.

It can be a very complex calculation (which in this example is probably a waste of computing resources which could be done faster with only looking at the terrain itself) but it's not a problem for the player in itself.

A complex simulation (or better, a very accurate simulation) of things doesn't make a game hard to for a beginner. A bad and unintuitive UI does and obscure game mechanics does. The best example for this is physics. I don't think many more than two or three dozen people on this forum are able to calculate more then some very simple classical mechanic equations, still nobody would protest that an accurate physics engine in a game adds too much complexity. Of course if physics would be presented to you in form of equations, you would have a point, but that's (hopefully) not what would happen.

Thus I declare your point moot.

You shouldn't worry about complexity (besides it wasting developement ressource, about which I can agree with you) but about its presentation and incorporating it into a good UI.

(Yeah, I get really annoyed about people complaining about too much complexity, since I haven't seen a really complex game in the last two decades coming out. I've seen games with horrendous interfaces and very bad presentation which makes understanding them in an easy way impossible. But no games with reall complexity. Needing to micromanage things isn't complexity anyway, it's bad interface design. Or presenting the player with hundreds of in itself meaningless numbers isn't very complex either, it's again a case of stupid interface design.

I understand people being affraid of the need to have to micromanage too much things, since I don't like micromanaging neither, but that fear of "complexity" is imho misplaced, probably coming from bad experience with games with bad UI and even worse presentation, coupled with the need to micro things which could have been automated.)

Successfully updated karma reason!

The only difference between good and bad complexity is how much of it is relevant at any one time.

Too much overwhelms, too little will make for a shallow game.

Successfully updated karma reason!

Quoting GW Swicord,

reply 4

On the complexity point, if I really got a cake and could eat the whole thing eventually, I'd get a game that was aimed at a 128-bit future when you looked deep under the hood but had good enough virtual staff for the single player to ensure that you could treat the whole thing like a nice visit to a 5-star resort.

Don't get your hopes up on seeing commercial (or any other) 128 bit computing in your lifetime. Not to imply that you're old or anything but that's way off and for now pretty much irrelevant. It's an exponential scale. 64 bit doesn't grant you twice as much maximum memory as 32 bit. The relevant numbers are 2^32 and 2^64 bytes (4 GB and 17,200,000,000 GB respectively). I don't think you would ever be able to finish a 17 billion GB sized map in your whole life, why would you want to anything more?

For posterity, 128 bit would give you a maximum 32,000,000,000,000,000,000,000,000,000 GB. I don't even know the name of that number. All I know is that 2^128 seconds is 1000 times longer than the predominant theoretical prediction of the lifetime of a free proton. (Yes, I am one of Vandenburg's two or three dozen people who can calculate more than some very simple mechanics equations ).

Quoting Tamren,

reply 14

The only difference between good and bad complexity is how much of it is relevant at any one time.

Too much overwhelms, too little will make for a shallow game.

Yes!

Quoting Vandenburg,

reply 13

There is no inheritent problem in adding complexity (or better said, detailed computation/simulation of certain aspects). It's all in the presentation and UI of things.

Let's say your units are in a desert. Now the game computes their efficacy in battle on the basis of the time of they day (how high is the sun?), amount of clouds, armor they use, how much water they've left, their race, etc. and as a result presents the player with a simple number (or bar, or any other graphical representation) that say, that unit is at 50% efficacy in this terrain.

It can be a very complex calculation (which in this example is probably a waste of computing resources which could be done faster with only looking at the terrain itself) but it's not a problem for the player in itself.

I disagree. If I march my army into the desert and all of a sudden my units are all at different combat efficiencies, I want to know why. I want to be able to use terrain and modifying factors to my advantage, or at least to avoid being disadvantaged. If time of day, clouds, armor, water, race, etc. all contribute to their combat efficiency then it would take way too long to diagnose why something is happening, and I would never be able to predict the outcome ahead of time.

Like Tamren said - "Too much overwhelms". Even with a good interface, there is still a limit imo.

Successfully updated karma reason!

I disagree. If I march my army into the desert and all of a sudden my units are all at different combat efficiencies, I want to know why. I want to be able to use terrain and modifying factors to my advantage, or at least to avoid being disadvantaged. If time of day, clouds, armor, water, race, etc. all contribute to their combat efficiency then it would take way too long to diagnose why something is happening, and I would never be able to predict the outcome ahead of time.

Well, if you're a compulsive min-maxer, I guess that could drive you crazy.

But from a normal gameplay perspective, it doesn't matter. You're in a desert and your army of heavy armored Mammoth riders now gets a penalty. That's not really hard to grasp for a beginner of the game. (Mammoths -> Ice Age or at least winter/cold -> in desert -> bad idea. Nice logical chain.)

That you can't completely understand (and compute by hand) the system isn't a disadvantage at all. It's part of the easy to learn, hard to master philosophy. (I don't understand the need for games to be exactly solveable anyway. You don't need to know the details, as long as you know the results (which the game should show you of course) you can react to it and adapt as necessary.)

The computer takes care of the calculations for you after all, I see no need that the player should be able to do so too; as long as the player can easily see the consequences (which is mainly an UI/design problem) and those consequences are consistent.

For an other example, take RTS games. There you usually don't get to see the numbers behind the units and the calculations done. But that doesn't prevent you from using them in efficient ways, since from their describtion, their looks, past performance, etc. you can learn and you'll know how to use them in the best way. The calculations behind the combat engine can be very complex and detailed, but it doesn't create a problem for the player since it only gets the results of those calculations shown.

And if the feedback done to the player is done well, you can simulate almost anything (though computing power, dev resources, etc. still create a rather large limit on it) as accurately as you want without overwhelming the player with information (mainly by graphic/sound feedback).

Successfully updated karma reason!

Quoting Vandenburg,

reply 16

The computer takes care of the calculations for you after all, I see no need that the player should be able to do so too; as long as the player can easily see the consequences (which is mainly an UI/design problem) and those consequences are consistent.

I think that's actually the core of it. You're right, with a sufficiently good UI, complexity can go through the roof without overwhelming the player. To take your Mammoth riders in the desert, for example. It might be obvious to me that they'll get a penalty in the desert, but how big a penalty? If it's a 5% penalty I might not care so much but if it's a 50% penalty then I'd care a lot. Personally I don't care about calculating exact percentages and numbers - but I like to be able to figure out ballpark figures for things in games. If there is too much complexity and the UI doesn't help you figure out hypothetical situations until after the fact, then I feel like I'm playing a game of chance, not a game of strategy.

So basically if there are a lot of modifying factors, especially if any of them are obscure or at all confusing, then there needs to be a good UI to help you roughly figure out the effects of moving your Mammoth riders into the desert.

Successfully updated karma reason!

Philocthetes

Join Date 10/2003

+153

Quoting Vandenburg,

reply 16

Well, if you're a compulsive min-maxer, I guess that could drive you crazy.

But from a normal gameplay perspective, it doesn't matter. You're in a desert and your army of heavy armored Mammoth riders now gets a penalty. That's not really hard to grasp for a beginner of the game. (Mammoths -> Ice Age or at least winter/cold -> in desert -> bad idea. Nice logical chain.)

That you can't completely understand (and compute by hand) the system isn't a disadvantage at all. It's part of the easy to learn, hard to master philosophy. (I don't understand the need for games to be exactly solveable anyway. You don't need to know the details, as long as you know the results (which the game should show you of course) you can react to it and adapt as necessary.)

...

Well, I'm totally behind "easy to learn, hard to master," but I'm a little confused about how you can say things like this while you also appear to want randomness minimized (elminated?) at the strategic level (I'm talking about some things you've said about city founding and research). Is it possible that you have fully compartmentalized your control freak needs and actively embrace randomness on the local battlefield while you want none of it back at court? I don't have nearly that control over my inner control freak.

Successfully updated karma reason!

Quoting GW Swicord,

reply 18

Well, I'm totally behind "easy to learn, hard to master," but I'm a little confused about how you can say things like this while you also appear to want randomness minimized (elminated?) at the strategic level (I'm talking about some things you've said about city founding and research). Is it possible that you have fully compartmentalized your control freak needs and actively embrace randomness on the local battlefield while you want none of it back at court? I don't have nearly that control over my inner control freak.

It may sound like a contradiction, but it isn't. (I think...)

I'm not a fan of randomness, true. But that doesn't mean that I want everything predetermined and I have no need to be able to understand a full causal link between everything.

For example my insistence on a detailed simulation of battles comes exactly from my dislike of random number generators. A sufficient complex battle simulation may not look different from a randomly calculated one on a superficial level, but it has imho some very important differences.

Without too much randomness, you can expect results from similar actions to be similar all the time. Thus you can plan on how things work out. With random tactical battles like we've in GalCivII (or even worse, the older Civilization versions, Tank vs. Spearman for example) you can get very silly/annoying/frustrating/suprising results that aren't explainable by anything than just bad luck (ie. I rolled a low number with my artifical dice). And personally I hate that.

But a system where I can exactly calculate what happens is boring, of course. Thus there needs to be enough options to make results variable again, but in a way that is under my control and not determined only by a random number generator. Of course you can't eliment rand, but lower it's usage to such a level, that it is not primarly responsible for a win or loss.

Take chess as an example. It has no randomness at all and everything is determined (thus you could theoretically solve the game by using computers with very large data storage). But it has thousends of possibilities after only a few turns, making it virtualy unpredictable for us humans. And that's what I would like to have in a game. No randomness (better; as low as possible) and enough possibilities to still make it interesting.

I hope that made sense.

Successfully updated karma reason!

Someone here mentioned the game Fire Emblem and the fact that you could show 'complex' and 'less complex' info on your units. While I haven't played the game myself I actually think that this is a good idea.

There was a discussion about mammoths in the desert or whatnot earlier and one person wanted to know EXACTLY why his mammoths suffocated in the desert while another said that it was all simple logic and could be explained by a 'this unit has a 50% chance of surviving in desert' screen.

Now personally, I hate games that make me watch a screen filled with random numbers and such. I wouldn't call them complex, just games with a bad design choice. When I play games like that i get bored...fast! But I understand that there are gamers out there who actually enjoy taking a look at these kind of stat. screens and learning all the formulas, so an option for min-maxing info. on units would be like a gift from heaven.

Successfully updated karma reason!

That is one problem I often give a lot of thought to. I like games to be realistic because our minds understand realistic things. The less realism you have the more you mind works overtime to understand the minutia of the system that is trying to replace it. In other words, things that make sense are easy to understand.

The 50% rule simply does not make sense because why the mammoths died is not explained. If you quantify that "heat" mechanic then you can implement it in a better way because you understand how it applies to different things. Hotter environments make people drink more water and tire faster. Animals with too much cold protection will overheat and die. You can even go beyond that, hot climates make ice meld and liquids evaporate and so on.

Not all hot places on the Earth are deserts. Much of Australia is hot enough to cook meat and the African savannah in the dry season is a harsh land. It not good enough to simply list these 3 enviroments as a 40-60% "heat rating".

Successfully updated karma reason!

Okay, at the risk of being accused of digressing again...

I don't recall saying that SD should do all of the stuff we're suggesting. Yes, most of what we want will be modded in, possibly by different modders, some better than others. I know I want to get in on the 'heroes' and 'animal handling' expansions, and probably those for Politics, Espionage, and several others. In the meantime, I'm more than willing to discuss anything and everything.

Will the number of mods make the game IMPOSSIBLE to balance? Yes, yes, it will. BUT if we learn which mods to avoid and which to encourage, we as players will be able to keep a decent level of balance while avoiding stupid-win strategies. Besides, there will be purists who will only play with SD issued material, maybe with mods with the OFFICIAL seal of approval on them.

Do I want to micromanage breakfast, making sure that toast is available? Bleh. OTOH, I see no problem with populations growing faster/healthier/happier with a varied diet. Take a look into history and see how much life improved when ancient people realized that having BOTH millet and wheat in a diet was better for them. Consider the economic dilemma 'Guns or Butter' - do I want to micromanage it? Meh. Do I want the automatic resource distribution system worry about it? Absolutely.

Honestly, I'd like to see something akin to Majesty. You can set policy, but you don't control EVERYTHING (unless maybe you research that). You should control your military and MAYBE your heroes, and certainly issue orders to those near your channeler. But there are levels of micro-managing I've seen requested that just shouldn't be available. I'm also a proponent for the 'messenger-envoy diplomacy' and 'imperfect intelligence' mods. So yes, the Elemental that I play will probably be different from the Elemental you play. And that's just fine, as long as we each enjoy the games we play.

Will the different mods make finding like-minded people online more difficult? Yep, count on it. But look at other games that allow modding. If a mod becomes popular and develops a fan base, it usually fills some 'niche need' that the unmodified game doesn't. I'd rather see a game like Elemental, where mods are CHOSEN and OPTIONAL rather than other ways I've seen mods implemented (usually single package, take it all or leave it). I had some point here, but my mind has obviously taken more than one sharp turn.

Successfully updated karma reason!

Think of it this way.

Elemental runs on building blocks. These blocks are used to build towers. Before you know if a tower is balanced or not, you have to build one. Everyone who plays Elemental gets to use the same building blocks. If we try to build a tower out of these blocks and the tower design is not balanced, then it will crumble and fall leaving us with a broken tower.

If Stardock can perfectly balance those blocks then it doesn't matter what we try to build with them. If we don't regulate ourselves and build balanced towers then we don't get to use them because they were broken from the start.

Successfully updated karma reason!

Complexity FTW? sure. As other posters have stated, I'm trusting SD to build a game i want to play and I"m trusting the community to build mods i want to use.

I want complexity and i want it in a fantasy setting. I've enjoyed many other games, some complex, others not but i cannot ever remember playing a fantasy game that i found complex and compelling (sorry, i only i discovered MoM quite recently, maybe if i found it 10 years ago...). I'm hoping that Elemental will fulfill that need. However, i don't want something along the lines of EU or Hearts of Iron, it's far too text based for me. I enjoyed a lot of things in those games but the complete lack of tactical battles used to depress me. And I don't want a <insert expletive> 'Sims' game or sim city builder. But i certainly want aspects of city building.

The problem is i still can't define in concrete terms the game that i want but i do know that i don't want stuff hidden away from me. If something happens, i need to be given some information so as i can figure out why that something happened. And as for micro-management, I want to be able to do it when i chose and automate it when i don't. I want micro-management to be scalable. I want detailed tactical battles and the option to auto-resolve. I want intelligent city governers and a wise advisory council. I want generals that command my armies who can be pre-scripted for auto-resolve battles. I want to be able to spend hours doing one late game turn and then spend only two minutes on the next turn knowing my army of automated councilors are making reasonably good decisions. I want endless tech/magic/research trees that are upgradable. I want...

But at the end of the day, this is only a forum and we're only offering ideas. I'm trying not to get my hopes up too much, as i've seen so many exciting and new ideas here that it would be impossible for SD to implement a fraction of them. But if they did make it into a game... WOW