 |
|  |
 |
|
MrBaggins
|
|
Ok... here are the quotes from the article...
quote: From general negative game experiences...
Bang, You're Dead. Sudden and arbitrary setbacks don't go over well. It is simply human nature that people don't like to have things taken away from them, especially when the decision seems arbitrary and capricious. Players don't want to spend hours building a beautiful city to have it suddenly and arbitrarily disappear into a crevasse. This is particularly true of major setbacks, but is true of setbacks in general.
|
quote: From 'steady progress'...
Emphasize "rewards" and minimize "punishments." It is a simple fact of human nature that players love to claim credit for their good decisions but they are considerably less fond of having to take responsibility for their mistakes. This is, after all, the entertainment industry-playing a computer game is supposed to provide an escape from whatever real-world unpleasantness we'd prefer not to think about. Who wants to boot up a game just to be punished?
In my years of observing game players, I've found that the harsher a "punishment," the more likely players are to resort to cheating-or to quasi-cheating such as repeated saving-and-restoring-in order to end-run the game system. This further reduces player enjoyment by forcing them to drop their suspension of belief and use a Windows dialog box. If our game tries to "resist" save/restore by making it difficult to do through the interface, it simply increases the torture as the player drops to a DOS prompt and copies files around. On the other hand players hardly ever resort to cheating when they've just received a reward. So a game which uses a range of light, moderate, and lavish payoffs can distinguish between different degrees of successful play while discouraging players from breaking the continuity of the game by cheating. Whereas a game which distinguishes between those same degrees of play through a system of punishments runs a much higher risk that players will simply reject the outcome.
Bear in mind the distinction between punishments and "plot reversals," the latter being pre-planned events (e.g. the princess is kidnapped) which are not directly the player's fault and serve to advance the storyline. Plot reversals serve a valid role. Whereas a punishment, in which something is taken away from a player because of a poor decision, risks alienating the player from the game experience.
|
|
|
|  |
 |
|
Gilgamensch
|
|
France
Jun 2002 time: 06:34
|
|
Yup, that is some more reasoning against it.
|
|
|  |
 |
|
Locutus
|
|
ACS CTP1/2 Manager & Civ4 Co-Manager
|
 |
Hengelo, The Netherlands
Nov 1999 time: 06:34
|
|
Random events (both positive and negative ones) can be a real asset to the game as they can add much atmosphere, but extreme caution is required. They should be fairly infrequent, their effects minor and it should be possible to counter them and/or their effects. In other words, you have to very carefully balance them to prevent players from getting annoying by them.
It should be possible to set the frequency and/or intensity of random events at the start of the game (where the default value would be very low), including the option to turn them off altogether. There's no rush in implementing random events, this can be done at any time.
Most of the implementation can actually be done in SLIC. The only changes to the code I can think of would be a slider or two at the start of the game which set one or two variables accessible from SLIC (g.eventchance and possibly g.eventintensity), which can then be used by SLIC coders to control the events. Of course, it would be useful to add some new SLIC functions that can influence to a city's food, production, commerce income and maybe some other things that might be useful for event effects. The only other code change I could think of would be to control the number and placement of map features such as volcanos, but that can probably largely be controlled in the tilefile already.
|
|
|  |
 |
|
MrBaggins
|
|
Many, if not most single players would be tempted to just reload a negative random occurance.
Perhaps we should have an "iron-man" option.
|
|
|  |
 |
|  |
 |
|
MrBaggins
|
|
If we have anything like this... preordain the timing.
You could randomly determine a number of events (probably disease... gelogical or astronomic events aren't IMO good 'civ' concepts) at the start of the game... thus reloading would make no difference.
I can't imagine how you'd prepare or avoid a meteor strike, and a volcano seems too terribly arbitary, and open to powergaming.
|
|
|  |
 |
|
Locutus
|
|
ACS CTP1/2 Manager & Civ4 Co-Manager
|
 |
Hengelo, The Netherlands
Nov 1999 time: 06:34
|
|
I agree things should be set up such that reloading is useless, I actually thought the current system already worked that way (with the seed for the randomizer being stored in the savegames), but I could be wrong about that.
The thing about Volcanos that's so interesting that it's up to the player to decide if he settles next to one or not: do you want to take the risk of having a city there for the bonus resources or do you stay away from it to avoid damage? That's hardly arbitrary if you ask me, that's entirely free will.
Some random thoughts on ideas for random events:
Negative
Flood: every turn x tiles (where x depends on mapsize and g.eventchance, default on normal map could be 1) are flooded. Floods can only occur on low land tiles (Plains, Grassland, Swamp, etc) bordering the sea or with a river on it. If a flooded tile has no city on it, there will be no effect. Otherwise, resource collection is reduced by y% (where y depends on g.eventintensity, default could be 20). Floods can be countered by City Walls (offers say 50% reduced chance of being affected by a flood if it strikes and reduce damage done by 50% when it does) and Battlements (same as City Walls, only 25% (cumulative)). Floods can often been predicted and people can evacuate and board up in time, so the only effect of floods would be in lost productivity.
Fire: every inland and non-river city has a x% chance of being hit by fire (where x depends on city size and g.eventchance, default could be something like 0.2 per pop). Fire can kill at most y% of the pop in a city and destroy at most y% of the buildings (where y depends on g.eventintensity, default could be 5). The chance of a fire hitting and the damage done could be prevented by an Aquaduct and something like a Sewer System (each 50% less chance of being hit, 50% less damage (cumulative)). Great city-wide fires are rare in present times so 100% prevention rate should be possible in modern times. Also, cities near rivers or coastal areas have a natural protection against fire, though in reality this isn't always 100% as I'm proposing here (notable example: London).
Earthquake: every turn there's an x% chance of an earthquake occuring somewhere on the map (where x depends on mapsize and g.eventchance, default on normal map could be 20). A random mountain/hill square is picked for the earthquake: the tile itself (epicenter) and all surrounding tiles (radius of 1) are affected. Units in the epicenter would get at most 4*y% damage, units in the radius at most 2*y% (where no units should ever be killed and y depends on g.eventintensity, default could be 5). If there's a city at epicenter, at most 2*y% of the pop can be killed and at most 2*y% of the buildings destroyed. For a city in the radius the same applies, only with y%. The effects of an earthquake can be fought with technology: for every advance in the Construction branch, the damage in cities is reduced by 10%. Hospitals reduce the population damage by another 20%. Units inside fortresses and cities with Barracks (and/or City Walls) are get 50% less damage.
Drought: every turn there's a x% chance of a civ being struck by a drought (where x depends on g.eventchance, default could be 0.5). If a drought occurs, food production will be reduced empire-wide by y% for at most z turns (where y and z depend on g.eventintensity, defaults could be 10 and 4). Droughts can be fought with technology: for every advance in the Agriculture branch, there's a 0.05% less chance of a drought occuring. Granaries and Food Silos reduce the effects of the drought for individual cities by 5% each.
Mixed
Meteor strike: every turn there's an x% chance of a meteor strike occuring somewhere on the map (where x depends on mapsize and g.eventchance, default on normal map could be 10). A random square is picked for this. If it's a Beach or Shallow Water tile, all tiles within a radius of 2 experience floods. If it's a land tile, the tile undergoes the same effects as the epicenter of an earthquake (surrounding tiles remain unaffected though). Regardless of location, the tile gets a bonus production of +y for z turns (where y and z depend on g.eventintensity, defaults could be 50 and 3), which comes from the minerals of the meteor.
Volcano: pretty much as The Big Mc described, there would be a small chance of a volcano erupting (say 0.5% on default -- and always at least 20 turns apart) but the surrounding terrain would be extra fertile (say +5 for all resources). The effects would be double that of an earthquake (except that units *can* be killed), where the volcano tile itself is the epicenter. The number of volcanos per map should always be rare and AIs should never settle on the volcano tile itself.
Positive
Resource Discovery: every turn, there's a x% chance of the possibility that new trade good is discovered (where x depends on map size and g.eventchance, default on a normal map could be 2). A random tile is picked, if it already has a trade good than nothing happens. Otherwise, if it falls within someone's border, there's a 2*y% chance of a new trade good appearing, if it falls outside everyone's border threre's a y% chance (where y depends on g.eventintensity, default could be 25). If the good is indeed created and it falls within either the borders or the line of sight of a human player, the player is notified of the discovery.
Ruin Discovery: same as Resource Discovery, except it always take place on land tiles and they create a goody hut rather than a trade good.
Baby Boom: every turn, every city has a x% chance of being struck by a baby boom (where x depends on g.eventchance, default could be 0.01%). If this happens, the city gets a 1 free extra pop point per turn for y turns (where y depends on g.eventintensity, default could be 2), as long as a city has a positive growth rate.
Rally: every turn, every city has a x% chance of witnessing an inspiring speach by a powerful leader figure (where x depends on g.eventchance and happiness, default could be 0.01% + 0.005% for every point above neutral happiness). If this happens, the resource collection in that city rises by y% for z turns (where y and z depend on g.eventintensity, default could be 30 and 3).
Years of Plenty (for lack of a better name): same as a Drought, except that the food production is increased rather than decreased. Is unaffected by technology and city improvements.
There are plenty of other ideas that could be added here (most notably Disease, but MrBaggins already provided a nice example on how to do that). Note that the chances of events happening are always low and the effects small (will of course need lots of playtesting).
Note that it would probably be best to disable random events for the first 20-50 turns altogether, as a setback so early on would be a definite game-killer (just as with Barbarians).
|
|
|  |
 |
|
MrBaggins
|
|
quote: Originally posted by Locutus
I agree things should be set up such that reloading is useless, I actually thought the current system already worked that way (with the seed for the randomizer being stored in the savegames), but I could be wrong about that.
The thing about Volcanos that's so interesting that it's up to the player to decide if he settles next to one or not: do you want to take the risk of having a city there for the bonus resources or do you stay away from it to avoid damage? That's hardly arbitrary if you ask me, that's entirely free will.
*SNIP*
|
I see a couple of issues
1) A lot of these are completely random and thus you have no chance to avoid them (unless you wish to connect churches/religion with avoiding catastrophy which would be statistically incorrect, from a historical perspective.)
A player shouldn't feel at the whim of a random event that he can do nothing about, unless its a story component. Are you trying to say that we should make disasters a narative part of civ?
2) The random number generator is problematic in that adding an additional choice changes subsequent random picks... thus a player could avert an event by generating an additional situation forcing a random roll.
Even if you avoided this, by seperating in-turn and between turn random seeds, you still have the problem of a variable number of cities, which again, could be manipulated to avoid an event.
You couldn't pre-ordain a number of these events at the beginning of the game, as you've stated, due to a variable number of cities... although this event in this location might work, and it would only happen if a city was actually there...
|
|
|  |
 |
|
child of Thor
|
|
quote: Originally posted by MrBaggins
I see a couple of issues
1) A lot of these are completely random and thus you have no chance to avoid them (unless you wish to connect churches/religion with avoiding catastrophy which would be statistically incorrect, from a historical perspective.)
A player shouldn't feel at the whim of a random event that he can do nothing about, unless its a story component. Are you trying to say that we should make disasters a narative part of civ? |
Well I think there is evidence that natural disasters have been(and will likely continue to be) part of the rise and fall of civilisations at various points in history. The fall of the High Egyptian Dynasty has been atributed to years of devestating drought - caused maybe by an El Ninio event elsewhere in the world(possibly caused by a huge volcanic erruption!).
And there have been many others(The burning of Rome/the last Ice Age etc).
So the RL precidents exist - BUT we dont have to make them so devestating with a sliding scale of severity and the choice to have them on/off, its really just a way to add a little more life to the build walls/build warriors/attack other civ type play.
If done well IMHO it would add to the whole civ experience, cause sometimes s**t happens - and triumph over adversity is part of all our histories, its why we are here today
quote:
2) The random number generator is problematic in that adding an additional choice changes subsequent random picks... thus a player could avert an event by generating an additional situation forcing a random roll.
Even if you avoided this, by seperating in-turn and between turn random seeds, you still have the problem of a variable number of cities, which again, could be manipulated to avoid an event.
You couldn't pre-ordain a number of these events at the beginning of the game, as you've stated, due to a variable number of cities... although this event in this location might work, and it would only happen if a city was actually there... |
Yes i'm not sure the 'every turn' thing would be the best way to go, sure its just probability - but i have rolled five sixes in a row on a six sided dice before 
Maybe we could store a variable at game start, based on the amount/severity of the random events you want, that would trigger the random roll every X amount of turns. X being a randon number of years based around those start up choices.
So it would have to count the years/turns in the game.
e.g.
I want to experience a huge meteor strike, so i select maximum settings for random events(maybe two choices for: Occurance and Severity?). This gives me that seed(lets call it Raneventtrigger) based on a random call between 10-50(years), so every 10-50 years the roll is made to see if i get an event and what it is and how severe it will be?
I guess the years could be turns if easier to count(so 1-5 turns in this example kinda thing).
I need a bit more explanation on how you could avoid an event dependant on the amount of cities you have?
I would like it not to matter what the player does or doesn't do unless it was a specific event like disease which the player could do something about(like build apothacaries/and or keep a low population?etc).
But for most events like Floods/earthquakes/volcanoes/meteor strike/Drought etc, these are natural disasters that just happen - the player just has to deal with the aftermath(but could lesson the effects of certain ones like Locutus mentioned).
@ Big MC........Moon metal? Well i'd prefer Moon Cheese i think - to feed my population when times are hard
Maybe for the Fantasy version? 
|
|
|  |
 |
|  |
 |
|
Gilgamensch
|
|
France
Jun 2002 time: 06:34
|
|
quote:
Yes i'm not sure the 'every turn' thing would be the best way to go, sure its just probability - but i have rolled five sixes in a row on a six sided dice before
|
That is exactly what I am afraid of as well. I just don't want to have my empire distroyed because of a serious of bad luck. Then I could play CIV 3 with it's combat model (no insult meant).
|
|
|  |
 |
|
Locutus
|
|
ACS CTP1/2 Manager & Civ4 Co-Manager
|
 |
Hengelo, The Netherlands
Nov 1999 time: 06:34
|
|
quote: Originally posted by MrBaggins
1) A lot of these are completely random and thus you have no chance to avoid them (unless you wish to connect churches/religion with avoiding catastrophy which would be statistically incorrect, from a historical perspective.) |
Well, that's why we call them random events, isn't it? That's the whole point But you most certainly can do something to stop them, or rather to limit (and in time eliminate) their effects: build the city improvements and research the advances I mentioned. In the examples I listed, the effects of all negative events except Drought can be reduced by 50% in the first Age. And since the effects are already fairly minor to begin with, I doubt many players would be seriously annoyed by them. Personally, I would find a medium or large stack of Barbarians appearing inside my borders a lot more annoying than most of these effects. They are smaller than the effects of random events in Civ1 (and Civ1 only had negative events), and personally I liked the events of Civ1 a lot, they added atmosphere to the game -- something CtP2 could do a lot better in.
As far as the exact generation of events goes, I don't really mind myself. Having an event every x turns works just as well for me. And if we would indeed roll all the dices at the start of the game, that would avoid any kind of cheating through reloading and the like. Of course, in that case it would be necessary to do more of the events coding in the source code then rather than in SLIC (since SLIC can be reloaded). But that works for me.
|
|
|  |
 |
|
Gilgamensch
|
|
France
Jun 2002 time: 06:34
|
|
quote:
(and Civ1 only had negative events),
|
Are you sure? The goodie huts had both and except the usual babarian plague can't remember anything else.
One thing which disturbs me with those random events: Can the AI handle it? No doubt that the normal player can handle it (most of it, at least) For MP games it could be fun
|
|
|  |
All times are GMT. The time now is 05:34. Apolyton Time is 00:34. |
top of page
|
|
|
Forum Rules:
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
|
HTML code is ON
vB code is ON
Smilies are ON
[IMG] code is ON
|
|
|
|
|
|