I would like, for once, just the once, to see a RTS or strategy game, in which you not only save up between two maps, but also add needs and enlarge your vision of the overall combat.

Let me explain.

Why is it that, in, say, StarCraft, I can end one map with a zillion ghost ships and still miss them in the beginning of next scenario? The reason is, mainly, that only your main character gets transported to the next situation.

But why is it that my past achievements, or NON achievements, never have any affect over the next scenarios?

Let's say you RUSH a scenario (still talking about StarCraft...) You have invested most of your resources in creating units, instead of researches. But if your opponent has invested more resources in researches than in units, then his researches are more advanced for the next scenario, even if he lacks units, in the beginning.

But as you go further through the campaign, the buildings you have built through the past scenarios are used to provide you with an additional number of units for the beginning of the next scenarios, but with a modifier due to distance, like it provides 50% of max production in one round for previous map, and only 25% of the one before, and then only 12.5% for the one before. And the researches are kept throughout the campaign. And of course the same still applies for your opponent, which means that, as you go further through the ennemy's ranks, and maps, you'll have to face more bunkerized areas, which will require either more firepower, or more cunning, or more stealth...

This means that the time you would spend building structures before ending the scenario would lack in the next, because that would give more time to the ennemy to create new units in the following map, and get some more troops from the back lines. But it also means that the longer you take to beat them, the more difficult it gets for you. But rushing through a scenario without preparing for the next is almost suicidal, because it means attacking the next map without new trops, or almost.

And I am also wondering what would happen if you failed objective 9, and were destroyed? Would this mean that you would have to replay through the objective 8 to go back to objective 9?

It seems that the general consensus is towards open-ended strategy games, either through having no strict goals (like the Sims), or by having a single big goal (like in Civilization), as that tends to go well with randomness thrown in (in building maps, random situations etc.) to help replayability. I like these too, which is why I will be writing the design of my game with these ends in mind.

Quote:

Original post by Telastyn Randomly generated scenarios suck over time. [think privateer] Privateer was a fine game, but it's not really replayable, and after a time the random missions became repetative and tedious. [edit: but on the other hand, X-Com's missions didn't get so stale, mostly because of the heavy application of random maps...

In your opinion though, was this a fault of the concept of randomly generated scenarios, or was the implementation lacking? Could randomly generated scenarios be done better? You've mentioned that combined with random maps they aren't quite as bad.

Since randomly generated stuff appeals to me as a game designer (partly because I like the replayability and unpredictability of it all, but mostly because I love writing the algorithms) I'll be including as much of this as I can.

Quote:

Original post by The_Good_Doctor For me a good strategy game should have all three types of goals.

Quote:

Original post by RobAU78 With strategy games, replayability is the key. Flexibility does a lot to ensure replayability. So I'd say include as many of those goal types as possible. Making each type (except for implicit goals) optional also makes the gameplay more flexible.

There's a lot of requests for all the goals to be included. This makes the design a little bit trickier, as in my view the scenario based goals and the implict goal design structure are hard to combine well. As has been suggested, if the scenario based goals were optional, rather than mandatory, then it would be possible to combine the two. If anyone can think of some interesting ways to combine both of these, or has seen an example of this in an existing game, then it would be useful to discuss it here. How can we combine all these goals together, to accentuate the good points of each and to downplay their weaknesses?

Perhaps through the choices that the player makes in any scenario based goals, there could be multiple endings. Each ending could tie together into a full conclusion of the storyline (and that doesn't neccesarily mean you need a lot of story in the game, but enough to show the player where they are going throughout the RTS). This would promote replay, as well as build a sort of mandatory effort into the game.

Quote:

In your opinion though, was this a fault of the concept of randomly generated scenarios, or was the implementation lacking? Could randomly generated scenarios be done better? You've mentioned that combined with random maps they aren't quite as bad.

Well, to one degree it is because if you've seen one "go to X, kill the pirates" scenario, you've seen them all. The environment doesn't really change, the decisions you need to make don't change; thus gameplay doesn't change.

Sure they can be done better, but personally, I think that they don't provide enough gameplay for the work and tradeoffs required to make them distinct.

Quote:

Original post by Telastyn Well, to one degree it is because if you've seen one "go to X, kill the pirates" scenario, you've seen them all. The environment doesn't really change, the decisions you need to make don't change; thus gameplay doesn't change. Sure they can be done better, but personally, I think that they don't provide enough gameplay for the work and tradeoffs required to make them distinct.

I know what you mean; the more simple random scenarios tend to be a bit too detached from the rest of the game world to feel like they have any meaning. However, I still think that there's a lot of gameplay potential out of a well-crafted random scenario generator that will make it worth the effort.

What if instead of "go to X, kill the pirates", the scenario was more like "go to X, kill the pirates because they've kidnapped your sister". Then to make things interesting, after the scenario is completed, you learn that your sister has become a pirate and asks you to give up your pirate-hunting ways and join her (adding a meaningful choice to the game)? If a random scenario generator could output things like that, and chain them together so they make sense in regards to a story arc, would that make the game a lot better, given that the system could change the path of the story to fit the decisions that the player makes?

Yes, if you could somehow design the greatest 'build your own story' book known to mankind, that'd be neat.

The problems come that such a thing is quite difficult, quite tedious to write up all those stories, and still somewhat limiting to write a story in such episodic fashion. Basically, it's easier to write an algorithm to generate an infinite number of coherent maps than it is to write an algorithm to generate an infinite number of coherent stories. Let alone good stories...

--A little bit offtopic--

It would be interesting if you had some continually regenerating map. The player will so have to consider not only the enemy but also the dynamics of the environment. (Ofcourse there must be some definite rules of this evolution)

At least I know I want to play such a game
Hope that would be of any value

Quote:

Original post by Telastyn Randomly generated scenarios suck over time. [think privateer] Privateer was a fine game, but it's not really replayable, and after a time the random missions became repetative and tedious. [edit: but on the other hand, X-Com's missions didn't get so stale, mostly because of the heavy application of random maps...]

I would argue that X-com's missions didn't get stale because they were part of something bigger. Even though each mission was, in a way, completely self-contained, the effects of the mission could be far-reaching or part of a larger goal. Alien ships had specific 'jobs' that would affect the game outcome, so if you shot down one alien ship you might prevent a terror mission, an invasion at one of your bases, or an infiltration.

The outcomes of the missions had wider implications too. If you fail a terror mission your standing with that nation would drop; they might reduce funding or pull out alltogether. Early on, losing a Skyranger because your entire team died was pretty harsh - they were expensive to replace and you'd have to wait for a new one to arrive. You might have lost one of your more experienced soldiers, something that could really set you back.

On the other hand, I think most Elite-style games (like Privateer) rely too much on wholly self-contained missions. At most, you might get a boost with one faction, but the overall game isn't affected much. You get a little more money and maybe you have to repair your ship, that's about it.

To stay on-topic, I think with strategy games smaller sub-goals are a good thing when the sub-goals are generated semi-dynamically.

X-com is the perfect example. In a way, the missions are 'generated' based on the actions of a pretty simple strategic enemy AI. The player's adversery has certain goals (or gives the illusion that it has certain goals) and the player is tasked with stopping it from accomplishing them. You're still pushing towards one major goal (destroying the Cydonia base), but along the way you're given numerous sub-goals that appear to be much fuller than madlibs-style randomly generated missions.

I'm a HUGE fan of open ended strategy games, but one thing that I lament is that not having some overarching goal which can validate or provide meaning for the player's game can really sap the life out of the experience. I sometimes wonder if what we fans of open-ended gaming want isn't so much a game that goes on forever, but rather a game that gives us *enough* of whatever experience it is that we crave.

Quote:

Original post by Trapper Zoid What if instead of "go to X, kill the pirates", the scenario was more like "go to X, kill the pirates because they've kidnapped your sister". Then to make things interesting, after the scenario is completed, you learn that your sister has become a pirate and asks you to give up your pirate-hunting ways and join her (adding a meaningful choice to the game)? If a random scenario generator could output things like that, and chain them together so they make sense in regards to a story arc, would that make the game a lot better, given that the system could change the path of the story to fit the decisions that the player makes?

Man, I want to see this done so badly! [grin] The whole idea that of being able to cause a chain reaction of events, sort of a "billiard ball physics" of experience (and storytelling, to a degree) would, I think, really make a world come alive. The devil is just coming up with the heuristics that drive the system in a meaningful way.

Quote:

Original post by Telastyn Yes, if you could somehow design the greatest 'build your own story' book known to mankind, that'd be neat.

Yes, wouldn't it be so neat [grin]? Actually, designing the "greatest build your own story book known to mankind" is my ultimate goal. I don't think I'll achieve it with this game, and I probably will never fully achieve it, but I plan to get as close as I can.

Quote:

The problems come that such a thing is quite difficult, quite tedious to write up all those stories, and still somewhat limiting to write a story in such episodic fashion. Basically, it's easier to write an algorithm to generate an infinite number of coherent maps than it is to write an algorithm to generate an infinite number of coherent stories. Let alone good stories...

I agree it has to be limiting and tedious to write up all the stories in episodic blocks, and that this technique will only work with a certain style of storytelling, but I feel that the trade-off in the loss of story writing precision in return for great immersion will be worth it.

Quote:

Original post by ParadoxishI would argue that X-com's missions didn't get stale because they were part of something bigger. Even though each mission was, in a way, completely self-contained, the effects of the mission could be far-reaching or part of a larger goal. Alien ships had specific 'jobs' that would affect the game outcome, so if you shot down one alien ship you might prevent a terror mission, an invasion at one of your bases, or an infiltration.

Quote:

On the other hand, I think most Elite-style games (like Privateer) rely too much on wholly self-contained missions. At most, you might get a boost with one faction, but the overall game isn't affected much. You get a little more money and maybe you have to repair your ship, that's about it.

That's what I feel about random missions too. If they are obviously a self-contained unit with little connection to the overarching story or situation, then there is little investment by the player in the mission. However, if your actions do influence the world (such as how you described in X-Com), then the experience is better. That's the sort of thing I'd like to try to incorporate into my design.

Quote:

Original post by Wavinator I'm a HUGE fan of open ended strategy games, but one thing that I lament is that not having some overarching goal which can validate or provide meaning for the player's game can really sap the life out of the experience. I sometimes wonder if what we fans of open-ended gaming want isn't so much a game that goes on forever, but rather a game that gives us *enough* of whatever experience it is that we crave.

The problem is figuring out exactly what the experience is that the player craves. But I've been thinking that maybe it's just that the player sometimes needs a nudge towards a possible objective; games that are too open are often a bit aimless. Do you think that their needs to be an overarching ultimate goal, or do you think that a series of smaller subgoals could replace this? I'm thinking a bit about soap operas here, where a character will often have a single big goal which is sometimes achieved (such as getting married) but since the show still goes on they get a new goal to work towards.

Quote:

Man, I want to see this done so badly! [grin] The whole idea that of being able to cause a chain reaction of events, sort of a "billiard ball physics" of experience (and storytelling, to a degree) would, I think, really make a world come alive. The devil is just coming up with the heuristics that drive the system in a meaningful way.

Tell me about it [grin]. I've spent a while thinking about this to come to the conlcusion that it is possible, but the whole entire gameworld needs to be designed in just the right way in order for to click. That's why I'm prepared to spend ages working on my game design until I know it's just right, because this kind of thing will fall apart if I try to slap extra gameplay into it at the last moment.