Game Design, Programming and running a one-man games business…

Unsure trade-offs in game design

Here’s a thought.

Good games are ones where we make unsure trade-offs. Most games are either about reflexes or decisions. Decisions are more common in the kind of games I make, such as strategy and sim games. I think the two basic approaches to strategy/sim games are plate-spinning and trade offs.

Plate-spinning is where tons of stuff is happening at once and you are trying to stay on top of everything and keep everything from falling apart. Democracy 2 is very big on this aspect of design.

Trade-offs are much more common. Even games that are conventionally reflex ones, such as First person Shooters have a lot of trade-offs. You choose to be a medic, trading ammunition capacity for the ability to heal. You choose to be a scout, trading everything for the ability to move fast. Choosing to have more of X, means less of Y.

Where this system goes wrong in games, is where it is too clear, too obvious, too analytical, to decide exactly what the trade-off is. In other words, the number are a bit too explicit. If I *know* the details of every variable in the trade-off, then it simply becomes a matter of Vulcan logic. It’s when there is a suitable amount of fuzziness around the numbers, that the trade-off becomes one filled with uncertainty, anticipation, risk and excitement. You *think* the best choice is to risk building a new factory in the city, trading off increased pollution against lower unemployment…but you can’t really be *sure* that the numbers will go your way…

To me… that makes for a fun game. I don’t always need to know the numbers. Sometimes, just a hunch makes for more fun.

11 thoughts on Unsure trade-offs in game design

  1. I for one agree, I often find myself quitting a game after I got all of the mechanics figured out.

    adding some randomness (a weapon deals 18-22 damage as opposed to 20 all the time for example) may help

  2. Pingback: |
  3. I’ve got to agree, any game where it is possible to analytically solve for an optimal solution immediately stops being interesting, because sucks any creativity out of it.

    I enjoyed GSB for the fact that I could mess around with ship designs and try crazy stuff and sometimes that crazy stuff worked out well. Or I could fiddle with whether it was better to have that one larger power plant and more powerful weapons, or if the trade-off wasn’t favourable. In either case, there were far too many variables for it to be realistically analysed analytically, and at least some of the relevant information couldn’t possibly be known, such as the exact setup of the enemy ships. There was never an obvious ‘correct’ design, or even one or two strategies that were the best, meaning there was always room for and reason to experiment and get creative. GSB is hardly the only game with these kinds of decisions, but I think it is a good example.

    A game like WoW (or even chess to an extent), on the other hand, has be analysed by so many people over so much time that optimal strategies are just known, and any variation on those becomes only possible in small degrees or for people who have completely mastered the games. That’s just something that holds no interest for me, because to get to the stuff I find interesting I would have to dedicate myself to doing something that just isn’t fun for quite a long time. It becomes like playing tic-tac-toe.

    Possibly it’s just the sort of person that I am, but if you give me a problem that I can solve by doing a simple sum, I will just do it. And at that point, the problem is solved, and a solved problem isn’t interesting anymore.

  4. I somewhat disagree. Yes, if there is a clear, obvious optimal strategy, it becomes more a difficulty adjustment option than a true choice. However, what you really want are uncertain tradeoffs- a higher DPS but it overkills more often? Splash damage, or DPS? Slowing some opponents so that splash damage is even more effective? All these depend on the situation, and how your opponent adapts to the situation.

    Numbers are important for some people, and they’ll get them somehow. I think an excellent example of this is in Desktop Dungeons (ignoring the whole god system, which has been ‘cracked’ by players anyway): you know the exact outcome of every action you could take, but there are so many choices it becomes impractical to find the most efficient one, compounded by the randomness of exploration.

    Adding random values is always annoying to me when done in a single-player game, which makes save-scumming a valid strategy. In a multiplayer game it can balance things out between players so that everyone has a chance to win, but I just don’t find it very fun in single player. Not to mention that it can be mathematically solved anyway for the average case, which means that it doesn’t help prevent an optimal solution from emerging anyway- it just means that rock beats paper 10% of the time.

  5. I think I disagree completely. I think that if there are definite effects of your choice in the game mechanics, those effects should be made clear to the player.

    The trick is to still make the player’s choices interesting ones, even when they know the details of the effects. For instance in your FPS reference, you say how different classes of characters get different gear or skills. That makes for interesting gameplay because there’s not a single, obviously best choice. Even if I know exactly how much healing a medic can do, or exactly how much faster a scout is, it’s still an interesting choice because it changes the way I play the game. Furthermore, I can pick the class based on my preferred kind of gameplay (e.g. slugging it out on the front line, making flanking attacks, providing support, etc).

    What is bad is when one choice is better than the others, but because the game designer wanted to make it “ambiguous” you can’t tell which one. Don’t do that! That’s the cheap way of making the player’s choices hard. Players (or hardcore players, at least) will eventually figure out what the effects are, and then always make the same choice. Then the game gets boring.

    I have to say that this was one of the problems I had with GSB, back when I was playing it in the early days (beta and for a while after the release). (Note, I’m not sure how much of what I’m saying here still applies after the many patches and expansions that have come out since I last played.) The effects of some of the weapons and equipment were opaque to the player. You could try watching your ships in battle so see how they performed, but it was often very hard to know what was going on (e.g. why is one class of my fighters dying, but not the other?).

    I learned a lot by opening up the data files to peer at the actual numbers. I wrote a script to extract the values into a spreadsheet so I could do DPS and shield/armor penetration calculations on them. Once I’d done that I found that there were far fewer interesting decisions to make. Some weapons were simply terrible. Others were devastating (cruiser lasers were my biggest discovery, with DPS that was double any other gun).

    That’s not to say that the game immediately ceased to be interesting once I had the data at hand. There were still tradeoffs to be made. Sure my cruiser lasers had a huge DPS, but they had terrible range. To use them I needed to drop armor or shields from my ships in favor of engines, so they’d be fast enough to get in close where their guns would function. It was usually worth while to mix in some ships with other weapons so that I could deal damage at greater ranges too.

    But a lot of other things had no such tradeoff. The Quantum Blaster (if I remember correctly) was worse than the Cruiser Laser in almost every way (including both DPS and range). So when faced by a choice of which weapon to pick, the QB was never going to get my choice. That’s was bad, since it meant that that part of the game was just a little less interesting.

    I know balancing games is extremely hard. Valve undoubtedly analyzes thousands of hours of gameplay to get the balance right in their FPS games, and I don’t expect the same amount of labor to be possible for a small indie studio. But hiding imbalances, even small ones, from the player is not a good design strategy. Maybe more casual players will eventually get “a hunch” that weapon A is better than weapon B. Hardcore players will look up the stats in the data files and know for sure. But once they know, why will they ever choose weapon B again?

    Its important that there be some kind of benefit to both options, when you have a choice. So weapon A could deal more damage, but weapon B has better range, or can deal its damage effectively to both fighters and cruisers (or infantry and mechs if we’re talking about GTB rather than GSB).

    There’s no need to hide the details of such a tradeoff. Let the players see what they’re giving up and what they get in return, and then they can make up their own mind. Avoid decisions where there’s only one right answer (which you need to obscure) and instead aim for there to be multiple interesting options (that you can explain fully).

  6. I disagree, it irks me when a game hides numbers. I was playing Space Marine the other day, and you can equip Combat Stims that grant ‘Temporary damage resistance and increased close-combat damage’. Sounds pitiful when you have to trade in your grenades to get them. What they actually do is reduce incoming ranged damage by 25%, incoming melee damage by 50%, and increasing delivered melee damage by 35%. Thats massive. The increased damage delivered alone means you can usually kill enemies in one hit instead of 2. In a single player game, perhaps its less of an issue, but in a competitive environment, I want to have all the information I can get.

  7. To clarify. none of the weapons in GTB hide any of the numbers, you can even turn on real-time damage indicators just like in GSB. My point is a wider one about a lack of certainty in game outcomes. GTB has this anyway, because you have no idea what the composition of the enemy force will be, or where they place the next unit.

    I think the problem is most evident in turn based games, where thinsg are way more clear-cut in terms of ideal counter-moves etc. The reason Democracy 2 works well, is that it has a huge number of interdependencies. You simply cannot think that many moves and effects ahead, so you *have* to go more by feel and instinct.

  8. “I’ve got to agree, any game where it is possible to analytically solve for an optimal solution immediately stops being interesting, because sucks any creativity out of it.”

    This, this so much! When you can analytically discover the best and optimal setup to solve anything, the game just becomes boring. Very big part for me in games is to discover and analyze the game mechanic and try to find different kind of solutions which I think are the best ones but with a little bit of creativity I can discover something even better and which works better than my previous solutions.

    The best game mechanic is that you cannot have the most optimal setup for every scenario but rather have huge variety of creative options which all work if you just study their basic ways of functions.

    In games I want to discover and I don’t want anyone to tell me that they have studied or played the game so long that they can straight out say that this and this is the best possible option for anything. No, I really do not want that. I want to see huge amount of variable options which all work and I don’t want anyone to come at me and say this and this are the best possible options for any scenario. I want to discover myself that and especially come up with something new no one has ever thought of and which still works.

    The best kind of game mechanic is that it can be analyzed but it still “stays hidden” so no one can’t ever find the best optimal setup for every scenario. There can be powerful setups which beat most of the other setups but there always is another kind of setup which beats that on some other scenarios.

    There is a reason why I have played Hostile Waters 64 or 65 times through.

  9. I agree with you Cliffski. I hate games where you can cookie cut everything. After you got the optimum it just loses all of its fun and I stop playing. Once I figure out how to beat a game effortless it’s just not a game anymore, it becomes a bothersome duty.

    Besically the gameplay is shallow because there is always just a right and a wrong choice. Because of that I like dices. Sure some people say dice rolls destroy a game because you can’t predict the outcome. But thats life. Don’t ever listen to people who want to take away our precious dice rolls!

  10. I ran across something quite interesting looking up articles on GTB, one from IGN claimed to be THE source of all GTB info, :) Lots of places like to say that about anything in particular that we are looking like we are even half interested in.

    Anyway, so i read the little blurb, it says:

    About this Game

    In Gratuitous Tank Battles, players take control of tanks of various sizes and shapes and either pit them against the other side’s defenses or must defend their own territory from an oncoming invasion. In addition players can download various customized attacks or defences created by other players in the community.

    Genre: Strategy
    Publisher: Positech Games
    Developer: Positech Games

    Release Date: November 16, 2009
    Exclusively on: PC

    And then i saw the date. And i thought to myself, hmmm…. either this was sorta one of those editorial slips of course, except the article itself was posted November of 2009.

    Which means of course that you have a time machine and are pre-testing your ideas years in advance, peppering us with your game ideas and then going back several years to hit us with them to sort of get us ready and then you pop back into the present and release the game and we are instantly pre-sold because of the prior subconscious mental conditioning of the several years before that we are at least consciously unware of.



  11. I think its perhaps more a choice between micromanagement or macromanagement.

    Compare two RTS games. Starcraft vs Sins of a Solar Empire. Both are RTS games, same gameplay. Collect resources, expand, work up the tech tree, blow up enemies.

    The difference is that in order to play Starcraft well you need to spin a whole lot of plates. An immense amount of micromanagement is needed, or you will lose horribly if every unit isn’t doing exactly what it should be doing at all times. This is immensely stressful, and honestly I don’t have the reflexes to be able to keep up with that for more than a few moments at a time. Clicks per minute need to be very high, such that you’re clicking many times per second continually the entire game.

    Sins, on the other hand, is a very slow game. Units are smart enough to take care of themselves. Get the right combination of ships in the same gravity well together and they should be able to tend to themselves well. Get the wrong combination of ships there, and you’re in trouble. But your fleet will not be instantly destroyed, you have plenty of time to withdraw your fleet. You could probably click just once every 10 seconds and still do perfectly fine in the game. Micromanagement can improve things, but only to a limited degree. Its all about the macro. Strategy and tactics are important, not just who has the fastest reflexes.

    I personally find the macromangement games much more enjoyable, because then my strategy and tactics mean something. The winner of a macromanagement game is who has the better plan for their army on the “big map” or who uses firing positions, ambush, and terrain better, or is clever with the R&D. Not who had consumed the most caffeine lately.

Comments are currently closed.