Archived

This topic is now archived and is closed to further replies.

Generic rules for a task, any task...

This topic is 6031 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

This is so vague I expect only a few replies, but I'll try anyway... If the core element of gameplay is interesting decisions, I wonder if you could come up with generic attributes that embody any given choice. The best I've been able to think of so far is fast, not so careful and slow, but very careful . Time expenditure and chance for success become major factors, and when you're competing in a task against someone else these two become vital factors. Example: Sniping in Counter-Strike. Though base skill is a factor, the player that lines up his shot carefully has a greater chance to hit. The fast/not careful and slow/very careful raise chance of hit while conceptually lowering movement and manueverability. Thus part of the interesting choice becomes whether to have a better chance at a kill while being vulnerable, or less of a chance at hitting or being hit. ?????? I'm chasing this because I've been thinking that one of the main reasons the player always has to either be solo or a manager / leader is because being a pleb isn't very interesting. Either your a solo pilot / gunman / driver / etc, or commander / general / CEO / whatever. You might be able to hold lots of different careers in a game if there were set "plays" that had different risks and costs attached. But before I figure that out, I've got to figure out if there are generic rules that can be applied to a task; or if tasks can be grouped (like management; combat; contests; etc.) and generic rules applied. As usual, any help appreciated... I'm mostly trying to figure out if you could hold different jobs as a player with some abstract interface that gives you choices, shows icons / animations, and lets you make interesting decisions-- without creating a dozen different game types. -------------------- Just waiting for the mothership... Edited by - Wavinator on June 8, 2001 1:36:02 AM

Share this post


Link to post
Share on other sites
I''m always willing to play with these sorts of ideas at 3 am.

First, let''s define a task, shall we? How about, "An action by the player or an AI that restults in an alteration in the game universe." The part of the universe so altered may be the player or AI themselves, or practically anything else. Usually the task is carried out for some motive: the player opens the box because zie wants the treasure that is probably in the box. The player makes pesants so zie can use the peasants to harvest vespene and crystals, so zie can use the vespene and crystals to build technologies and units, so zie can defeat zir opponent.

Your example is of a task where there are multiple strategies that can be chosen from. You want to look at basic variations of strategies that may exist. Your label your example as a choice of speeds, but I think it is more fundamentally about riskiness to payoff ratios, an fundamental theme of game theory. (With, of course, modifying factors for human tendencies to enjoy blowing away enemies, want to show off their abilities, want to earn a special ending, want to mop up uvery last unit for completeness'' sake, or want to get on with the story.)

For a different type of strategy variation, let''s consider something where riskiness is not an issue. You are at the beginning of a plain, ordinary maze with no monsters, no timer, no treasures. There is a right-hand passage, a left-hand passage, and a straight-ahead passage. Which do you take? Well, do you want to pass through the least area of maze, or the most? The efficient choice is to take one of the side paths and follow that side wall until you get to an exit, hoping that the maze doesn''t have any loops in it. The efficient scenic route method is to do this, anot go out the exit, continue until you are back where you started, then take the shortest path back to the exit. You could call this variance minimal vs. thorough.

Another type: you are generating your character at the beginning of the game. What gender do you make your character? Assuming the two genders are strategically equal choices, there are several ways to go about chosing one. There''s random, default (player''s own gender), ideal (gender player would like to be, if different), exotic (gender opposite player''s own), esthetic (what gender character would be easier on the eyes), social (I better play my real gender or my friends will think I''m wierd) or psychological (if I role-play a girl I will get more attention). The difficulty with all of this is that the player''s motivations lie entirely outside the game, and so cannot really be predicted or controlled for.

I think another major factor in choice of strategy is annoyingness/boringness. Of course everyone evaluates this differently, but most people would rather whack 100 different monsters than 100 of the same monsters.

Share this post


Link to post
Share on other sites
skill vs weapon power
- shoot for the head, in tho-three bullet sprees (and risk missing and doing little damage), or empty your magazine
- choose an automated weapon (fail once, keep firing) vs an manual reload weapon (very powerfull first shot, but if you have poor aim, you''ll die alot)

style vs utilitarianism
- fight in CS using only handguns, or going for the glory of a knife kill vs always using the best weapon


specialisation vs versatility
- using shotguns (only good at close range) or Magnum snipers (only good at long range and if you have your back covered) vs a all around good weapon like the Steyr Aug assault rifle (less close range damage than shotgun, less zoom and firepower than Magnum)

team vs self
- will you help the team or will you only care about your frags, your life, your weapons, etc.


good vs evil
- some people play counter or terrorist on general principle

role-play vs win
- while playing a counter, you are faced by a terrorist scum hiding behind a hostage. You dont care about the money you''d lose if you killed the hostage (whatever reason). Do you just shoot through the hostage, or aim carefully, all the while the terrorist just shooting at you (through the hostage) ? Or better, open the door to the terrorist and hostage, throw two grenades inside and let God sort them out

Share this post


Link to post
Share on other sites
thx for the feedback so far!

So to be interesting, it seems that every task should have a risk and reward associated with it, with resources that are gained or lost based on the choice.

Certain factors determine the outcome of the choice (skills, randomness, the mood of the Goddess, whatever).

So, as an example: Sim-Job

At the most elemental level, for every task different strategies will be possible. Each has risks (in terms of resources like time, or safety, or money, or prestige) and rewards. The result, as s&s noted, alters the game world in some way.


SimJob Strategy 1: Sloppy

You''re at your terminal and decide to persue this strategy. The risk is that the work will fail, and you''ll lose peer status and promotional advancement opportunities (which could be a simple linear resource). The reward, though, is more time and less stress.

SimJob Strategy 2: Precise

You''d gain reverse of the above.

SimJob Strategy 3: Innovative

You''d risk spending more time, but possibly make the job less stressful and more efficient, which would impact your sim choices later and elsewhere outside of work.

If the player were playing something like The Sims, then this concept would allow more participation during certain tasks. I notice that when you''re training your sims, often they just repeatedly perform an animation without your input. This isn''t very interactive.

An idea like this might allow a Sim to have various jobs in many different environments, without each job actually having to be a mini-game in and of itself. Thus more career tracks would be available.

As always, comments welcome...

--------------------
Just waiting for the mothership...

Share this post


Link to post
Share on other sites
Interestingly, one genre where it is a common option to be a member of a team and not the centerpiece of the collective is Sports. This option is often called player lock, and while most gamers (in sports games the avatar being controlled is the "player", who must be distinguished from the human playing the game, thus the "gamer") will choose the superstar, it is quite possible to elect to be a role player who doesn''t even get much playing time. The fulfillment of one''s role in terms of being in certain positions at certain times to help complete "plays" is the immediate task the gamer faces on each successive offense and defense. The gamer may do this well or poorly, which affects the game outcome in terms of score.

Analogously, provided the role of the player within a larger context is well defined, a very entertaining and deep experience can be had as a subordinate or pleb. The problem lies in consequences: in the sports genre, bad play by the gamer doesn''t result in a loss of confidence in his/her abilities by teammates (they''ll still pass the ball to you, and the coach won''t bench you until the "regular" parameters say so). This reinforces Wav''s observation about the necessity of the gamer''s actions yielding in-game consequences - additional or fewer opportunites as a result of previous decisions. So in addition to attributes about how the gamer makes a decision, there should be attributes about the results of the decision to enable the game respond accordingly. This has been implied but not explicitly codified.

Excellent discourse. Please continue as I learn much more than I (claim to) contribute.

Share this post


Link to post
Share on other sites
quote:
Original post by Wavinator
So to be interesting, it seems that every task should have a risk and reward associated with it, with resources that are gained or lost based on the choice.



Actually I disagree with this part. A decision whose results are purely esthetic, like choosing the color of your avatar''s hair, can be very interesting.

Share this post


Link to post
Share on other sites
quote:

Original post by sunandshadow
Actually I disagree with this part. A decision whose results are purely esthetic, like choosing the color of your avatar''s hair, can be very interesting.



You''re right sunandshadow. An esthetic decision can be highly interesting. In example, when you find yourself behind an unsuspecting enemy sniper camper and you must choose between carefully aiming a bullet at his head, backstabbing him with the knife or placing a grenade at his feet and watching him jump.

But you must agree that even esthetic decisions have risks and rewards. After such a decision, you may like (reward) or not like (risk) the result. Or more, in a game like the Sims or Roller Coster Tycoon where you can show other peoples your creations, they may like or dislike your choices.

Everything is ultimately risk and reward, but exact definitions of risk and of reward are hard to give. And yes, risk and reward extend out of the game world (if you camp or tank rush people hate you, if you create a beautifull park in RCT the reward is much more than the game money you make, etc.)



More choices :

short term vs long term
Build military units early on or use the resources to build up your economy. Risk either loosing to a rusher or being crushed by a player with superior economy later in the game.

Or, taking a high paying job that can only last for a small time, or taking a less paying job that in time will get better paid.


Share this post


Link to post
Share on other sites
Uhh... vagueness... ick

Ok... the original post seemed to show me something: that often, a decision isn''t between Choice X and Choice Y. In fact, it is about picking a point between Minimum and Maximum. Sometimes the game/system/situation will make the granularity such that there are only 2/3/N choices, but essentially, the player makes a judgment on a tradeoff. Essentially, the choice of how much to invest against the expected rewards.

Example of the sniping concept given:
less time invested = poorer aim + more time remaining.
more time invested = better aim + less time remaining.

The player has to guess how much time it is worth expending on a given target, given that there are external unpredictable variables.

However... Diodor goes on to detail a whole load of other decisions of a different nature.

Good vs Evil is a social concept and can only be generically embodied when the game sets a specific moral framework in place.

Role-play vs Win, Team vs Self, and Style vs Utilitarianism all seem like player preferences (ie. some artificial decision chosen by the human controlling the character). You could also claim that Role-play vs Win is just a specialisation of Style vs Utilitarianism on a game-playing level. I don''t think these choices can be meaningfully measured or accommodated in a generic system.

Skill vs. weapon power - this to me seems like my first point: deciding how much time to invest per shot.

Ok, so after that babble above, I really can''t see any way in which you can possibly make any useful generic system without trivialising it into a "Do It Well" vs "Do It Quickly, With Time To Try Again" system. The interest in a particular role is in the details that vary from role to role... and given that the variance is what makes them interesting, the interesting part is the part that cannot be modelled generically.

Share this post


Link to post
Share on other sites
Hm, guess I missunderstood the topic a bit. I didnt compile past the "interesting decisions" part, and the words CounterStrike just pushed the button. Thanks Kyolatan, you''ve corrected some mistakes.


Good vs Evil
This could be implemented in some carreer simulations. Ie. if you become a lawyer, you start getting an evil aura. Before long, you''ll find that some jobs are out of your reach (super-hero, health care, minister), while other job opportunities open for you (evil overlord, etc.).

quote:

Original post by Kyolatan
Role-play vs Win, Team vs Self, and Style vs Utilitarianism all seem like player preferences (ie. some artificial decision chosen by the human controlling the character). You could also claim that Role-play vs Win is just a specialisation of Style vs Utilitarianism on a game-playing level. I don''t think these choices can be meaningfully measured or accommodated in a generic system.



Role-play vs Win is out. Style vs Utility, I''m not so sure. Different jobs can affect how the world thinks about you. So, some jobs can generate more cash, while other generate less cash, but may change the NPC''s opinion on yourself in one direction or another (being a cool super-hero and rely on robbing goblins for a living vs being the towns merchant).

Team vs Self however can be easily translated in jobs. Do you help a guy on your team if he asks for help ? If you do, the overall team performance improves. You may have a lot to win yourself by being altruist. Or, you can send him to do his own homework, and enjoy his failure as an opportunity to advance faster through the team ranks, or simply have more time to focus on your own projects. And risk that the company goes under, or your mates wont help you either. (Or you can get real dirty, and secretly put bugs in your mates code. Let him get fired, and pretend to work a lot to debug his code )

Skill vs Weapon power cannot be used either. It is an interesting decision though. Do you specialize in a hard to control weapon that requires skill or do you use reliable hardware. AK47 vs Colt M4A1. ADOM : Do you play the "move through monsters" barbarian or the weaker wizard ? The wizard risks dieing a lot, so he needs a lot more care and skill to play, but in the end, he is may end-up more powerfull than a fighter class. Hm, perhaps it can be used after all : do you join Microsoft and do whatever they tell you to, or join a startup - garage game developing company. If you are that good and have luck, you might get much richer with the second option (ID software ), though you''ll probably get paid less than with Microsoft even if you are quite good.

Specialisation vs Versatility is also easily used. Some jobs allow you multiple carreer advance choices, while others, well, dont.

And to add another one :

Health vs Money : This is Ego inspired. The job you choose can affect your attributes. If you are a blacksmith, you''ll get stronger. If you''re a librarian, you''ll get wiser. If you are a gold miner, you''ll get rich, but your health may suffer. Donating blood (money++, stamina--) also fits here. And if you handle toxic wastes at the local nuclear plant, you may grow breasts / moustache .

Share this post


Link to post
Share on other sites
quote:
Original post by Oluseyi

Interestingly, one genre where it is a common option to be a member of a team and not the centerpiece of the collective is Sports. This option is often called player lock, and while most gamers (in sports games the avatar being controlled is the "player", who must be distinguished from the human playing the game, thus the "gamer") will choose the superstar, it is quite possible to elect to be a role player who doesn''t even get much playing time. The fulfillment of one''s role in terms of being in certain positions at certain times to help complete "plays" is the immediate task the gamer faces on each successive offense and defense. The gamer may do this well or poorly, which affects the game outcome in terms of score.



WOW! I never thought to look at sports, but this is a great example. I imagine for something like football or basketball the roles are similar enough in terms of actual actions (like throw the ball, or run/move with it) that you can give the player a subordinate role and let the AI play everyone else. In this example, everything depends on teamwork, so the player is performing a task that will help his/her group.

quote:

So in addition to attributes about how the gamer makes a decision, there should be attributes about the results of the decision to enable the game respond accordingly. This has been implied but not explicitly codified.


Very good point. You show me that an individual task actually doesn''t need to be engrossing as long as the player is working towards some bigger picture. The gameplay in a basketball game, for example, doesn''t include dribbling the ball (in most I''ve seen, pls. correct if I''m wrong). This is a task that''s handled automatically.

quote:

Excellent discourse. Please continue as I learn much more than I (claim to) contribute.


Actually, Oluseyi, this is a great point I hadn''t thought of! Thanks!

--------------------
Just waiting for the mothership...

Share this post


Link to post
Share on other sites
quote:
Original post by Wavinator

WOW! I never thought to look at sports, but this is a great example. I imagine for something like football or basketball the roles are similar enough in terms of actual actions (like throw the ball, or run/move with it) that you can give the player a subordinate role and let the AI play everyone else. In this example, everything depends on teamwork, so the player is performing a task that will help his/her group.



That''s exactly what happens. Properly implemented, the different co-players will act out their different roles on offense and defense, often exhibiting the characteristics/properties of the real-life athletes they''re modeled after. So a Shaquille O''Neal will be dominant in the post, score a lot and grab a lot of rebounds, a Patrick Roy will be phenomenal in goal, leaving your Roberto Carlos to take his patented free kicks (from basketball, hockey and international soccer, respectively).

quote:


Very good point. You show me that an individual task actually doesn''t need to be engrossing as long as the player is working towards some bigger picture. The gameplay in a basketball game, for example, doesn''t include dribbling the ball (in most I''ve seen, pls. correct if I''m wrong). This is a task that''s handled automatically.



Precisely. You''re never required to continuously press a "dribble" key or anything of the sort, though I advocate in my own designs more control, up to the user desired level. What this means is that there is a huge table of actions that the gamer can explicitly perform or have handled by AI, including aiming and timing on a jumpshot (so that the gamer simply presses the "shoot" button). Based on personal style preferences (and on a PC, on input device), various combinations of "responsibilities" can be selected, varying the skill requirements on the gamer. Thus the same game at the same difficulty setting can vary from painstaking detail to arcade run-n''-gun.

This allows the task to be as engrossing as the gamer wishes without detracting a jot from interactivity. So in a strategy sim, I would advocate being able to delegate certain responsibilities to lieutenants and subordinates. Obviously, there has to be some variability to this as well, which brings up the concept of training. In the Sports scenario, if your player has a high shooting rating, for example, then setting the aim and timing to automatic will yield good results; doing the same with a poor shooter will yield frustration. Coming back to the strategy game, the more training in tactics and management you invest in your commanders, the better the decisions they''ll take and the more responsibility you can delegate. This shifts the gamer''s focus from micromangement to the grand scale.

Share this post


Link to post
Share on other sites