Yeah, "partial victory with repercussions" is a design idea that has always appealed to me. I''ve never seen it in a game, but I''m almost certain several games have done it to a greater or lesser degree. You lose a particular mission, and you can still proceed, but there''ll be repercussions. Obviouosly a lot of RTSs have "secondary objectives", but these are usually short term in their effects.

You don''t neccessarily have to have a mission-branching type structure as yckx described - that''d be nice, but would add a fair bit of dev+testing time. Just have certain settings in one mission dependant on pervious mission(s). Eg, you fail to destory an enemy convoy in one mission. For the next three missions, the enemy is going to have more equipment. You fail to take out an enemy comand/control/comms post. Next mission, enemy AI is superior than it would otherwise have been - better coordination, more organised/aggressive attacks.

You could also integrate this kind of thing into a dynamic story system. You have an ally in your campaign. Maybe you fail to rescue one of their top brass in a particular mission. Your "approval rating" with your ally goes down. Depending on how you do in other missions related to them, your allies might refuse to send reinforcements/supplies, and so on, dependant on how much they like you. Sounds almost like a form of emergence, but really very simple - just have player achievement of certain missions/objectives increase or decrease their "approval rating". Have other triggered events dependant on the player''s rating. Just needs a solid scripting system, which almost any RTS is going to have anyway.

Obvioulsy, this can also be applied to other genres - I could see it working very well in a mission-based FPS, for example. RPGs effectively do this kind of thing already.

Partial victory altering the later stages of the game is an old idea. The Ocean game Inferno (space combat game) had it about 8 or so years ago. As mentioned the major problem is the extra work load. If you have four levels of success for a mission that spawns four new levels that have to be created, each of which may have their own chain of follow up levels. That would be an extreame example though as you could simply alter the difficulty of some of the later levels or allow the player to access/not access a level that gives them some additional weaponry that helps later on.

One other point to keep in mind is the definition of "victory". Roof Top Pew Wee''s example of a few hundred units defeating or at least damaging a much larger force but ultimately being destroyed might easily be counted as a victory as the Spartan "victory" at Thermopylae was. Although ultimately the 300 Spartans that made their final stand there were killed they succeeded in delaying the advance of the Persian army long enough to allow further Greek forces to prepare defencive positions, thus changing the outcome of the "next mission" in the war. Victory is the meeting of objectives, not necessarily the total destruction of an enemy force.

Dan Marchant
Obscure Productions

Very intriguing!

What if each mission had a very particular objective that affects future missions: even in a base-building, upgrade-as-quick-as-you-can type RTS, the map for the mission could have a very specific offensive or defensive objective -- such as defending a "strong point" or attacking a fuel depot -- and a success in that battle would take the player to a different mission than a failure.

For the macro-players, I think it would be really cool to give the player some kind of War Theatre Map where they can view or even plan a Path-to-Victory; i.e., a visual indication of weak points, strong points, supply depots, primary bases, and so on, each offering certain advantages or disadvantages to the enemy, and let the player figure out which ones to attack next.

Alternatively, the game could be kept much simpler, and still linear, by merely offering a battle advantage to the victor of the previous battle. For instance, if the player''s last challenge was to destroy a supply depot and they failed, then they still progress to the next mission but the enemy will have greater supplies than if the player had succeeded, giving the player a marked advantage.


Brian Lacy
Smoking Monkey Studios

Comments? Questions? Curious?
brian@smoking-monkey.org

"I create. Therefore I am."

Well, while this is certainly a fascinating idea, I think it might be a bit hard to actually implement. It is impossible to let the computer generate as interesting and believable stories as a human can. I guess that''s the reason why almost everyone is using this system: Because it works. A story tree works, too, the problem there is as many have pointed out that you have to create many more missions.
Also, I don''t know if these suggestions would actually make for a better game.. In a war, the more you lose, the harder it gets.. But if I was a player new to a game, of course I will lose a lot of battles. But what happens? They get even harder! (Because it''s realistic). So in the end I''d be frustrated because I just can''t seem to get ahead in the game and I''d quit playing it. As simple as that.

Well, still the thought is a very exciting one, and if done right, it could make for a game that might be very interesting indeed. Because the question wouldn''t be whether you won, but instead, what you accomplished. Maybe what is needed is indeed a combination of pre-created stories and generated content..

quote: Original post by Wuntvor
Maybe what is needed is indeed a combination of pre-created stories and generated content..

how about world war II?
or Vietnam? ect ect..
sit down with some history buffs and try to figure out what would''ve
happened if things had gone differently in particular battles.
my iwo jima reference before comes to mind.. the united states
wanted iwo jima''s air strips to launch attacks on the japanese
mainland directly (as hawaii was too far for their fighters to fly,
and japan''s navy was superior to ours).
Real life wars would make an excellent model for a story tree..
If you created a comprehensive, yet user friendly mission-creation
system, you could get the missions done with no problem. playtesting
shouldnt be a problem either, as i''m sure you''d have a lot of
people who would consider the question.... ''what if?''


-eldee
;another space monkey;
[ Forced Evolution Studios ]

Do NOT let Dr. Mario touch your genitals. He is not a real doctor!

quote: Original post by eldee
Which brings me to my design idea:
Why do we always have to win? Defeat is inevitable (to some degree)
in any war or series of battles. Every win and loss is part of
the intricate web of war. Some battles you will find yourself in
a situation so precarious, a surrender may even be necessary.

Every battle should attribute something to the end game''s results.

in fact THIS is allready included in RTS multiplayer mode. For example Warcraft 3:

loosing some ''minor'' battles in the beginning of the game ( Lost Tempel: gaining control over the fountain of health ), killing a few of the enemys units while not loosing one urself and TPing out(town portal), seeing unwinable battles and retreating without fighting them... this all is VERY important... more then good macromanagement ( = building up base, producing units)

I like the idea you are all coming up with, but it seems to me like you might as well go to the next step. Instead of having 100s of scenarios organized in a massive tree, you could just have one big scenerio? The problems with this idea are somewhat obvious. You would need to have a MASSIVE map which composed of the ENTIRE world. There would be a lot of units. Mostly memory requirements. Which could be overcoe by only dealing with part of the entire map at a time. This way, you don''t need to create a huge list of scenarios and such.

just some thoughts.

tazzel3d ~ dwiel

eldee-
But I wonder how good it is to try to find "choke points" in a games progress? In many ways, victory conditions should be unknown or variable. If the game designers/levelers can deduce what the weak points are, then the players can too.

Sometimes the weaknesses are obvious, sometimes they are not. Sometimes strength''s are illusory. I think that in many ways, play balance is overrated in game design. In some ways I feel that balance should happen through the playing of the game, not predetermined at the beginning of the game. Look at the American Revolution for example. If we played this as a "balanced" game, America surely would have lost. The truth is that wars are fought when an imbalance exists, and its up to one side to redress that balance as quickly as possible.

So if you try to figure out the turning points, or weak points in a game, the players are sure to figure it out too. I actually think this topic deserves its own thread......be back in a flash

quote: Original post by Wuntvor
Well, while this is certainly a fascinating idea, I think it might be a bit hard to actually implement. It is impossible to let the computer generate as interesting and believable stories as a human can.

I don''t believe thats what I was suggesting. That''s a completely different debate and better left for another thread.

quote: Also, I don''t know if these suggestions would actually make for a better game.. In a war, the more you lose, the harder it gets.. But if I was a player new to a game, of course I will lose a lot of battles. But what happens? They get even harder! (Because it''s realistic). So in the end I''d be frustrated because I just can''t seem to get ahead in the game and I''d quit playing it. As simple as that.

Not necessarily. Every so often the outcome of ONE battle leads to a complete turnaround in the war. Consider the battle of Gettysburg in the American Civil War, or D-Day during World War II -- both massive turning points in two very bitter wars. The right outcome in the right battle ought to allow the player (or the enemy) to change the course of the game''s story.

Again, I''m not talking about the computer automatically creating missions based on the outcome, that would be ridiculous with today''s technology. I''m talking about the pre-existing scenarios of the game being weighted by adjusting the objectives confronted by and resources available to the player in those scenarios -- all of these would be pre-arranged by the designer, but the scenarios themselves need not be developed from scratch. Or, at the most ambitious level, the designer would indeed develop a whole tree of scenarios which, as one person suggested, would allow for incredible replay value, which most RTSs currently lack completely in singleplayer mode.

quote: Well, still the thought is a very exciting one, and if done right, it could make for a game that might be very interesting indeed. Because the question wouldn''t be whether you won, but instead, what you accomplished. Maybe what is needed is indeed a combination of pre-created stories and generated content..

Good idea. Its something to think about. Generated content presents another layer of technical challenges, but can eliminate SOME of the more repetitive aspects of the design process and potentially add replay value.


Brian Lacy
Smoking Monkey Studios

Comments? Questions? Curious?
brian@smoking-monkey.org

"I create. Therefore I am."

If you consider a battle as being part of a larger war, it is possible to win the battle, but if it costs you the war then it is a pretty hollow victory. Persistent campaign systems model this automatically, but for one of skirmishes, it could be modelled as part of a scoring system.

Some of my earlier thoughts on this stuff here