Full Games List
Featured Games
New Games
Reviews
Media
Images
Blogs
Game Updates
Images Feedback Summary
Game Design Highlights
Tutorials
Articles
Scripts
Utilities / Plugins
Resources
Engines
Engine Bulletins
What is Makerscore?
Past Events
Achievements
Misaos
Forums
Post History
Chat
New to RMN?
Full Store List
RMN Music Pack (free)
RANDOMNESS
Posts
LockeZ
I'd really like to get rid of LockeZ. His play style is way too unpredictable. He's always like this too. If he ran a country, he'd just kill and imprison people at random until crime stopped.
5958
LockeZ
View playlists
CloseI've gone on record as generally hating, in RPGs, when my abilities randomly fail or have very random effects. I usually don't use those abilities, even if they're extremely strong. Not just because they're useless, but also because they're not fun. I think most people don't use those abilities, unless they have to. And if they have to, and the ability fails, it sucks balls, because they had no control over their failure. When the player fails, it should be his or her own fault, not because the game rolled dice and decided to make you lose this time.
But, are there ways to do randomness right?
Let's look at enemy abilities. They're usually random. And even if they're actually a set algorithm instead of being random, they're unpredictable, because the player doesn't know the algorithm. This is basically the same end result as randomness: the player knows something's going to happen, but doesn't know what. No one (to my knowledge) has ever accused any RPGs of sucking because you don't know ahead of time everything the enemies are going to do, though. What's the difference here? Why is this kind of unpredictability okay, but it's a problem when the player casts Spell Roulette and causes a random magical effect?
Or as another example, critical hits. The difference between a critical hit and a regular hit is probably the same as the difference between a regular hit and a miss, but people cry out for games to remove miss chances. Critical hits cause problems in a few games, but usually they add fun to the game and no one minds them.
In one of my games, I have been working on classes and ability sets, and have been toying with randomness in a few of them, trying to come up with ways to make it work. Here are some of my ideas so far:
* A class that starts out extremely powerful but extremely random, but as the player uses the class, it gains new abilities that make it less random.
Goal: make the randomness feel like a problem that can be overcome, instead of one that can't.
-- Main ability: steal an extremely powerful combat item from the enemy and then toss it, dealing massive damage or providing party-wide healing/buffs. can only carry one item at a time, so if you steal an item you don't really need, you wasted a round.
-- First upgrade: a powerful ability on a long cooldown, giving you some predictable damage once in a while
-- Second upgrade: another powerful ability on a long cooldown, also restores MP
-- Third upgrade: you deal minor damage any time you try to steal
-- Fourth upgrade: your steal ability no longer has a chance to fail
-- Fifth upgrade: free extra damage after every ten steals
* A class that causes random effects to continue for several rounds, allowing you to respond to them.
Goal: Make the randomness be something you can build off of by using different abilities strategically. The rest of the battle is influenced meaningfully, and the player's response is "make use of what I got" instead of "keep trying the same thing until it randomly succeeds."
-- Main ability: start dancing and causing random effects. The dance continues for the whole battle. Effects include damage, healing, buffs, and ailments. You don't lose control of the character; you can use other abilities (from your secondary class) while you keep dancing.
-- At the minimum, this at least allows the party to decide to focus on healing because the dance caused damage this round, or vice-versa.
-- The player can choose different dances, which have different possible effects.
-- Many effects are very short duration buffs or ailments, allowing the player to exploit them by strategically choosing when to use the party's other abilities. For example, a buff that triples one party member's healing for one round, or a debuff that makes all enemies weak to fire for one round. This hopefully causes the player to sometimes consider using skills he'd never otherwise use, because they're briefly much better.
Do these ideas sound like they'd work? Do you guys have other ideas of how to make randomness fun and compelling instead of just a nuisance?
But, are there ways to do randomness right?
Let's look at enemy abilities. They're usually random. And even if they're actually a set algorithm instead of being random, they're unpredictable, because the player doesn't know the algorithm. This is basically the same end result as randomness: the player knows something's going to happen, but doesn't know what. No one (to my knowledge) has ever accused any RPGs of sucking because you don't know ahead of time everything the enemies are going to do, though. What's the difference here? Why is this kind of unpredictability okay, but it's a problem when the player casts Spell Roulette and causes a random magical effect?
Or as another example, critical hits. The difference between a critical hit and a regular hit is probably the same as the difference between a regular hit and a miss, but people cry out for games to remove miss chances. Critical hits cause problems in a few games, but usually they add fun to the game and no one minds them.
In one of my games, I have been working on classes and ability sets, and have been toying with randomness in a few of them, trying to come up with ways to make it work. Here are some of my ideas so far:
* A class that starts out extremely powerful but extremely random, but as the player uses the class, it gains new abilities that make it less random.
Goal: make the randomness feel like a problem that can be overcome, instead of one that can't.
-- Main ability: steal an extremely powerful combat item from the enemy and then toss it, dealing massive damage or providing party-wide healing/buffs. can only carry one item at a time, so if you steal an item you don't really need, you wasted a round.
-- First upgrade: a powerful ability on a long cooldown, giving you some predictable damage once in a while
-- Second upgrade: another powerful ability on a long cooldown, also restores MP
-- Third upgrade: you deal minor damage any time you try to steal
-- Fourth upgrade: your steal ability no longer has a chance to fail
-- Fifth upgrade: free extra damage after every ten steals
* A class that causes random effects to continue for several rounds, allowing you to respond to them.
Goal: Make the randomness be something you can build off of by using different abilities strategically. The rest of the battle is influenced meaningfully, and the player's response is "make use of what I got" instead of "keep trying the same thing until it randomly succeeds."
-- Main ability: start dancing and causing random effects. The dance continues for the whole battle. Effects include damage, healing, buffs, and ailments. You don't lose control of the character; you can use other abilities (from your secondary class) while you keep dancing.
-- At the minimum, this at least allows the party to decide to focus on healing because the dance caused damage this round, or vice-versa.
-- The player can choose different dances, which have different possible effects.
-- Many effects are very short duration buffs or ailments, allowing the player to exploit them by strategically choosing when to use the party's other abilities. For example, a buff that triples one party member's healing for one round, or a debuff that makes all enemies weak to fire for one round. This hopefully causes the player to sometimes consider using skills he'd never otherwise use, because they're briefly much better.
Do these ideas sound like they'd work? Do you guys have other ideas of how to make randomness fun and compelling instead of just a nuisance?
Milennin
2060
author=LockeZ
* A class that starts out extremely powerful but extremely random, but as the player uses the class, it gains new abilities that make it less random.Goal: make the randomness feel like a problem that can be overcome, instead of one that can't.
-- Main ability: steal an extremely powerful combat item from the enemy and then toss it, dealing massive damage or providing party-wide healing/buffs. can only carry one item at a time, so if you steal an item you don't really need, you wasted a round.
I would say it depends on the chance of wasting a turn and how punishing battles can be. If the chance is fairly low, like 25% or something, it's not really an attractive option in my opinion, I'd rather have a skill similar to the 1st upgrade.
Also, if the combat is difficult up to the point where it punishes the player for making mistakes, missing out on a turn due to a bad roll of the dice sucks, and could cause game overs. Especially sucky when the RNG is against you and you get several bad rolls in a row.
LockeZ
I'd really like to get rid of LockeZ. His play style is way too unpredictable. He's always like this too. If he ran a country, he'd just kill and imprison people at random until crime stopped.
5958
Well, obviously everyone would rather have a skill similar to the first upgrade. That's why it's an upgrade. Once you get it, the class has two attack skills; you keep using the main random ability when the second non-random ability is cooling down.
In this case, stealing an item has a 75% success rate, but you might get an item you don't need, like you might get a healing item when you have full health or a haste item when you're already hasted. You also might get a damage + silence item when the enemy doesn't cast magic, or a damage + blind item when the enemy is immune to blind, or a healing item when you're at 75% health; these are useful, but not ideal, which makes it difficult for me to pinpoint the exact chance of getting a "good item." 50% maybe? If all you want is damage, you will actually be doing, on average, about 50% more damage per round than other jobs can do with their first abilities. It's just going to be a lot more unpredictable.
I should note that the game usually isn't particularly difficult, mostly because running away not only has a 100% success rate, but can even be done mid-round after you take your turn but before the enemy takes theirs. Even against bosses. So you always have at least one tool to let you survive bad dice rolls, and if you actually die, it's probably for a different reason than this.
In this case, stealing an item has a 75% success rate, but you might get an item you don't need, like you might get a healing item when you have full health or a haste item when you're already hasted. You also might get a damage + silence item when the enemy doesn't cast magic, or a damage + blind item when the enemy is immune to blind, or a healing item when you're at 75% health; these are useful, but not ideal, which makes it difficult for me to pinpoint the exact chance of getting a "good item." 50% maybe? If all you want is damage, you will actually be doing, on average, about 50% more damage per round than other jobs can do with their first abilities. It's just going to be a lot more unpredictable.
I should note that the game usually isn't particularly difficult, mostly because running away not only has a 100% success rate, but can even be done mid-round after you take your turn but before the enemy takes theirs. Even against bosses. So you always have at least one tool to let you survive bad dice rolls, and if you actually die, it's probably for a different reason than this.
Addit
6394
I suppose one thing that could probably help randomness not be so…randomness is probably with a growing proficiency level that the party / characters can gain through the use of using those particular skills in battle and on the field and can better upgrade the chances of certain instances not being so random by levelling up that skill or using certain equipment attributes to make it work more often and with much more success (like a Luck stat).
I guess as the proficiency grows, you would probably have to accommodate it, such as stealing a better particular item that you’re actually going to need or don’t quite have yet, or using a random spell ability with the spells being more powerful and actually going to help you more in a fight than not.
I guess some people really don’t like a lot of luck based events. I guess I can kinda agree there; nobody really wants things to be left up to chance. I suppose it’s all fun and good in something like a board game, but for everything else it helps to have more control over these things.
I guess as the proficiency grows, you would probably have to accommodate it, such as stealing a better particular item that you’re actually going to need or don’t quite have yet, or using a random spell ability with the spells being more powerful and actually going to help you more in a fight than not.
I guess some people really don’t like a lot of luck based events. I guess I can kinda agree there; nobody really wants things to be left up to chance. I suppose it’s all fun and good in something like a board game, but for everything else it helps to have more control over these things.
Avee
5794
author=LockeZ
...What's the difference here? Why is this kind of unpredictability okay, but it's a problem when the player casts Spell Roulette and causes a random magical effect?
Most games provide a description of the selected ability's effects. The player knows what to expect here, what the result of said action should be. If the ability fails I don't get what I expected to. That outcome + the fact that success is luck-based rather than skill- or strategy-based can be frustrating. The event that just occurred, me selecting an action and seeing it fail to give me the expected result, involves me by having me make a choice and take action. And the action I picked and the turn I spent were wasted.
When I don't know what to expect, there's no "logical" reason to be frustrated because I couldn't possibly know what the end results will be. The event here doesn't involve me as a player: it's just discovery. And when I know what an enemy's skillset is, I can devise a strategy to fight him wisely. This knowledge breeds better choices and therefore reduces any frustration that might arise from the enemy actions' randomness.
Also, using an ability that can lead to one of several random effects and not getting the one I think would be most suitable at the time is less frustrating than scoring a miss with, say, a status ailment spell. At least I get something. My choice and turn were not entirely wasted.
author=LockeZ
-- Many effects are very short duration buffs or ailments, allowing the player to exploit them by strategically choosing when to use the party's other abilities. For example, a buff that triples one party member's healing for one round, or a debuff that makes all enemies weak to fire for one round. This hopefully causes the player to sometimes consider using skills he'd never otherwise use, because they're briefly much better.
If you need to cast a spell or use an ability to apply the fire weakness, than I would probably never use it, as I would prefer to just spend two turns casting the fire spell. Some of the Final Fantasy games did this right with the Oil status, where it was a long-lasting effect and allowed you to cut battle time by a significant amount in some cases. Other games have done this poorly, where the debuff lasts only one turn, and it ends up being worse than just casting Fire twice. It's one of those things where it depends on how much the damage increase is, and whether or not it would be worth using up a turn.
As for "Random Spell" skills, you could have it semi-random. Final Fantasy III had a Geomancer class that used different skills based on the terrain, with the chance of casting a 9999 damage spell or auto-kill spell regardless of the current terrain. Or you could have tiered random spells, with each tier having better spells. You'd have a general idea of how good the spell will be, but it might not be the element you were hoping for, or could just end up being a status effect targeting all enemies.
Byah
1097
author=LockeZ
In this case, stealing an item has a 75% success rate, but you might get an item you don't need, like you might get a healing item when you have full health or a haste item when you're already hasted. You also might get a damage + silence item when the enemy doesn't cast magic, or a damage + blind item when the enemy is immune to blind, or a healing item when you're at 75% health; these are useful, but not ideal, which makes it difficult for me to pinpoint the exact chance of getting a "good item." 50% maybe? If all you want is damage, you will actually be doing, on average, about 50% more damage per round than other jobs can do with their first abilities. It's just going to be a lot more unpredictable.
You're basically giving this ability another chance to fail, just in case it dares to hit. :P
I'd at least make it so that the stolen item depends on the creature you steal from. That way, the player has to decide who to steal from, and plan ahead. Much more fun than praying to the RNGods, twice. Then I'd remove missing altogether!
I find there's two kinds of RNG: bad (failing) and good (a nice bonus, but not required to win), and missing definitely belongs to the former.
Something I did to make more fun rng, was to have a character inflict poison on critical hits. The player can then build towards more crit, or leave it as is and be satisfied with the one ability that makes him auto crit every once in a while.
LockeZ
I'd really like to get rid of LockeZ. His play style is way too unpredictable. He's always like this too. If he ran a country, he'd just kill and imprison people at random until crime stopped.
5958
author=ByahThere's no actual difference between these two things, though! The only difference is what the player expects. If you expect to get the stronger effect, it's "bad," and if you don't expect to, it's "good."
I find there's two kinds of RNG: bad (failing) and good (a nice bonus, but not required to win), and missing definitely belongs to the former.
Is the key, then, in simply controlling player expectations? Make them somehow expect skills to be worse than they actually are on average? Gambling does the exact opposite of this and is extremely fun for a lot of people. I think there's more to it.
author=ByahI'm more interested in ways to make randomness work than in removing it. I'm well-acquainted in all the reasons to remove every last shred of randomness from every video game in existance. I want to hear the other side.
You're basically giving this ability another chance to fail, just in case it dares to hit. :P
I'd at least make it so that the stolen item depends on the creature you steal from. That way, the player has to decide who to steal from, and plan ahead. Much more fun than praying to the RNGods, twice. Then I'd remove missing altogether!
author=Backwards_CowboyI'd like to hear your reasoning for why this would solve the problems that other random skills have? I mean, FF3/FF5 geomancer and the slot machine skills in FF6/FF7 are basically perfect examples of the problem cases that I'm trying to solve. They're as unpredictable as any skill I've ever seen in a video game, there's nothing you can really do to improve or control or respond to them at all, and they lack any compelling reason for players to use them once they realize how the skills actually work.
As for "Random Spell" skills, you could have it semi-random. Final Fantasy III had a Geomancer class that used different skills based on the terrain, with the chance of casting a 9999 damage spell or auto-kill spell regardless of the current terrain. Or you could have tiered random spells, with each tier having better spells. You'd have a general idea of how good the spell will be, but it might not be the element you were hoping for, or could just end up being a status effect targeting all enemies.
Edit: this would probably be a more constructive topic if more of you were inside my brain and I didn't have to communicate, because I think I'm failing hardcore
Avee
5794
author=LockeZ
Make them somehow expect skills to be worse than they actually are on average?
I'd think of it as a nice surprise when I see it. It would add excitement at best, but I'm not sure it would add incentive to rely on that skill as part of one's strategy. In order to do that you really have to reduce the randomness in one way or another.
As long as I get something positive out of it, even if it ain't the optimal result I could be hoping for, I would keep on using the skill.
My suggestions:
1) the Steal skill already has a 100% chance of success,
2) add one nearly useless item in the list of those I can steal from the enemy,
3) and make the 4th upgrade an increased chance of stealing any item but the useless one.
It could work fine if each enemy has its own list of items (and that would let the developer make at least one item optimal against said enemy, which adds incentive to use the Steal skill).
The first and second upgrade add other means to achieve the goal of "dealing damage", but cannot be spammed because that would defeat the idea behind this character's design. Also, this character seems more inclined towards offensiveness than defensiveness (unless most stealable items have a defensive effect).
I guess you'd have to carefully design the stealable items list and effects so that, for a single enemy, some are less and some are more suitable than the character's unpredictable abilities.
The more punishing enemies could hold more ill-suited items and less good ones, etc. Against those I would probably rely on my other allies to deal damage, but the thief guy would still have a chance of being useful in any situation.
The dancer character could work fine just like that, as long as most of the effects can come in handy in one situation or another, and that no other character designed for support can be more efficient despite having a smaller array of effects he can cause.
Gambling isn't fun to me because the chances of getting nothing or poor rewards is often extremely high. If I can break the game with Save States I can spend a few minutes to win big and buy that Metal Babble Shield, but I won't get any fun out of it either -_-
I think randomness can be fun for more people when it isn't the only factor which can affect the outcome. In poker you can bluff: trying to win by shaking the opponent's confidence. In roulette you can only cross fingers.
author=LockeZThe Geomancer's (at least in 3. I didn't use it in 5.) worst possible skill if I remember correctly was a moderately powerful wind spell. Using the random terrain skill basically guaranteed you would at least deal some damage, so taking the chance didn't have a massive downside like a possibility of missing would. The appeal I found to it was that the first few bosses after the class became available were dispatched much more quickly with a Geomancer than with any of the spells available to mages. The Geomancer was pretty much a decently powerful mage who traded the ability to choose which spell you used for infinite MP, as well as being a bit more bulky with physical stats. I didn't play 6 or 7, so I'm not quite sure how the slot machine skills even work, but if you were to make a roulette wheel/slot machine where the player chose when the spin stopped, you could eliminate some randomness.author=Backwards_CowboyI'd like to hear your reasoning for why this would solve the problems that other random skills have? I mean, FF3/FF5 geomancer and the slot machine skills in FF6/FF7 are basically perfect examples of the problem cases that I'm trying to solve. They're as unpredictable as any skill I've ever seen in a video game, there's nothing you can really do to improve or control or respond to them at all, and they lack any compelling reason for players to use them once they realize how the skills actually work.
As for "Random Spell" skills, you could have it semi-random. Final Fantasy III had a Geomancer class that used different skills based on the terrain, with the chance of casting a 9999 damage spell or auto-kill spell regardless of the current terrain. Or you could have tiered random spells, with each tier having better spells. You'd have a general idea of how good the spell will be, but it might not be the element you were hoping for, or could just end up being a status effect targeting all enemies.
RyaReisender
200
If failed randomness means nothing happens then randomness is pretty bad. Status effect and instant death spells are a good example. Also running away when it simple is a % success chance is the same thing. That's why I never use status or instant death spells in games and never run away either (unless it's 100% successful).
Randomness is not always bad, though. Games that do randomness right are SaGa-games in my opinion. What you learn is pretty random, but yet there are so complex mechanisms behind it, that even if players will never understand them, there is always an overall balance given.
Randomness shouldn't be ignored just because it frustrates players, because often it can also make the games more interesting and in particular give them good replay value. It just needs to be more complex than just "50% success / 50% fail". In fact good randomness needs to be so complex that no player can understand it, to prevent him bothering with trying to figure it out.
Edit: On a sidenote. If I make a game for myself, I try to make it as random as possible. From my own personal viewpoint, this makes the game a lot more interesting, because my goal would be to make it so random that the game can even surprise me as developer.
Randomness is not always bad, though. Games that do randomness right are SaGa-games in my opinion. What you learn is pretty random, but yet there are so complex mechanisms behind it, that even if players will never understand them, there is always an overall balance given.
Randomness shouldn't be ignored just because it frustrates players, because often it can also make the games more interesting and in particular give them good replay value. It just needs to be more complex than just "50% success / 50% fail". In fact good randomness needs to be so complex that no player can understand it, to prevent him bothering with trying to figure it out.
Edit: On a sidenote. If I make a game for myself, I try to make it as random as possible. From my own personal viewpoint, this makes the game a lot more interesting, because my goal would be to make it so random that the game can even surprise me as developer.
Byah
1097
author=LockeZ
Edit: this would probably be a more constructive topic if more of you were inside my brain and I didn't have to communicate, because I think I'm failing hardcore
Oh I get it now, you want to add randomness, and make it fun (like you said in your OP!).
Let's see, one random skill I enjoyed playing with (until I got more reliable attacks anyway) was Metronome in Pokemon. If used, it executes a random attack out of all existing ones, making it very unreliable, but still fun since most of the time you'll still do some damage and every now and then, you'll have that big one hit ko.
Which I think is what can make randomness fun: low investment (one turn), low chance to get a game changer (one hit ko) and at least a decent return a lot of the time.
Grayburg
163
Players choose the random option over the consistent one if the random option has these properties:
-The random option has a "much" higher expected value than the consistent option
Players will use a 1~10 attack over a flat 3, but maybe not over a 5. Pretty straightforward.
-The random option's favorable outcome isn't too improbable
Players might use skills like Death or Metronome for fun, but they won't rely on them.
A skill that deals 9999 damage 1% of the time would never be used over a skill that deals a consistent, pitiful 10 damage, despite being "ten times better."
Having this option gives players a way to scumbag some victories, though. 1-gil Zanmato, all day erry day.
-The random option's unfavorable outcome doesn't lead to an unwinnable situation
This is the most interesting reason to still choose a random option. If options have this property, the first two aren't required.
This can't apply to random skills in most RPGs, because if an attack misses or fails to crit at a crucial moment, there's no way to salvage that situation. gg
Attacks in fighting and action games, however, come to mind. Every attack could be described as a skill with a low success rate for a ridiculous reward. You deal damage, stagger the opponent until it's your "turn" again, and take no damage. Ridiculous!
It's much more probable to have an attack be blocked, though. The attacks that leave the player susceptible to damage, a potentially unwinnable situation, are less used.
This doesn't mean unwinnable game states can't exist for the third property. A battle can definitely be unwinnable if the player pits an Ice Pokemon against a Fire type, but not because Blizzard fails to freeze the opponent.
My way of using the third property in an RPG is to include a defensive system that can nullify any damage. Highly game-breaking, of course, in a traditional RPG, but inline with my design philosophy.
As for your examples, since they are better overall than other classes, I'd use them.
The thief sounds bland and random, except for the fifth upgrade for an OT reason.
The dancer's stated goal is much better than the thief's in terms of appealing randomness, though.
-The random option has a "much" higher expected value than the consistent option
Players will use a 1~10 attack over a flat 3, but maybe not over a 5. Pretty straightforward.
-The random option's favorable outcome isn't too improbable
Players might use skills like Death or Metronome for fun, but they won't rely on them.
A skill that deals 9999 damage 1% of the time would never be used over a skill that deals a consistent, pitiful 10 damage, despite being "ten times better."
Having this option gives players a way to scumbag some victories, though. 1-gil Zanmato, all day erry day.
-The random option's unfavorable outcome doesn't lead to an unwinnable situation
This is the most interesting reason to still choose a random option. If options have this property, the first two aren't required.
This can't apply to random skills in most RPGs, because if an attack misses or fails to crit at a crucial moment, there's no way to salvage that situation. gg
Attacks in fighting and action games, however, come to mind. Every attack could be described as a skill with a low success rate for a ridiculous reward. You deal damage, stagger the opponent until it's your "turn" again, and take no damage. Ridiculous!
It's much more probable to have an attack be blocked, though. The attacks that leave the player susceptible to damage, a potentially unwinnable situation, are less used.
This doesn't mean unwinnable game states can't exist for the third property. A battle can definitely be unwinnable if the player pits an Ice Pokemon against a Fire type, but not because Blizzard fails to freeze the opponent.
My way of using the third property in an RPG is to include a defensive system that can nullify any damage. Highly game-breaking, of course, in a traditional RPG, but inline with my design philosophy.
As for your examples, since they are better overall than other classes, I'd use them.
The thief sounds bland and random, except for the fifth upgrade for an OT reason.
The dancer's stated goal is much better than the thief's in terms of appealing randomness, though.
the most important thing, in my opinion, is transparency
if physical attacks usually don't miss, but they do 1/128th of the time, like they might in something like a DQ game, that's kind of annoying. but if you take something like grandia 3, where they hit unless 1) the enemy SWAYs (evades, due to being an enemy that can dodge, like a bird or a rabbit) or 2) they are literally moving out of the way (the character begins their attack animation, but the enemy is starting movement to perform a new action). a big enemy like a golem isn't going to SWAY, because it's a huge hunk of metal. that makes sense.
then there are games like my own Wine & Roses, in which there's an Agility stat that determines accuracy and evasion, and huge buffs for accuracy and evasion themselves. if an enemy uses one of those buffs, you know physical attacks are going to be less effective... it's transparent. it's right there.
i'll go into more detail on other subjects if people are interested i guess??
EDIT: also my more recent projects always have physical attacks be 2-5ish hits each due to the randomness of accuracy. accuracy means MORE TOTAL damage, not a binary. feelsgoodman
if physical attacks usually don't miss, but they do 1/128th of the time, like they might in something like a DQ game, that's kind of annoying. but if you take something like grandia 3, where they hit unless 1) the enemy SWAYs (evades, due to being an enemy that can dodge, like a bird or a rabbit) or 2) they are literally moving out of the way (the character begins their attack animation, but the enemy is starting movement to perform a new action). a big enemy like a golem isn't going to SWAY, because it's a huge hunk of metal. that makes sense.
then there are games like my own Wine & Roses, in which there's an Agility stat that determines accuracy and evasion, and huge buffs for accuracy and evasion themselves. if an enemy uses one of those buffs, you know physical attacks are going to be less effective... it's transparent. it's right there.
i'll go into more detail on other subjects if people are interested i guess??
EDIT: also my more recent projects always have physical attacks be 2-5ish hits each due to the randomness of accuracy. accuracy means MORE TOTAL damage, not a binary. feelsgoodman
LockeZ
I'd really like to get rid of LockeZ. His play style is way too unpredictable. He's always like this too. If he ran a country, he'd just kill and imprison people at random until crime stopped.
5958
author=Crazegogogogogogo
i'll go into more detail on other subjects if people are interested i guess??
Shinan
4314
For some reason I like the transparency in a die roll. It must be years of boardgame conditioning but a visible die roll with a clear target (4+ to succeed for example) somehow makes randomness exciting.
I guess it has something to do with knowing just how much I succeeded or failed even if it makes no actual difference. (rolling a 3 on a 4+ roll feels different than rolling a 1)
I also think that a combination of non-random and random is generally a bad thing. If everything has a chance to fail the game is about risk-management and maybe building up bonuses to make the random go in your favour. While a non-random system is about optimization in a completely different way.
Of course with non-random and random I don't mean completely so. But take a system like Ogre Battle which is essentially non-random. It does have attacks occasionally miss but everyone's actions and setup are completely non-random at the start of a skirmish. It creates a certain kind of strategic planning.
While a more random game like Blood Bowl is all about risk-management. Doing the least risky things first in a turn (or occasionally doing the things that are the most critical that they succeed) and trying to mitigate the damage that the randomness undoubtedly will cause.
I guess I don't know where I'm going with this at all. But I guess transparency. And also maybe some way of pre-planning to assess risks in any encounter. Having randomness means that a battleplan doesn't survive contact with an enemy but it's still always good to be able to create one. Or something.
Also the concept of "rerolls" is a good thing to have in random systems. If a simple, but critical roll fails it's always good to have some kind of limited system of redoing something and hope for a better result. A player is always more likely to try a 50% chance if he will have another 50% chance if the first one fails. And limiting them might mean you wasted one on a failed 75% chance earlier and now you no longer know if the 50% is worth it. Which is all part of the fun of risk-management.
I guess it has something to do with knowing just how much I succeeded or failed even if it makes no actual difference. (rolling a 3 on a 4+ roll feels different than rolling a 1)
I also think that a combination of non-random and random is generally a bad thing. If everything has a chance to fail the game is about risk-management and maybe building up bonuses to make the random go in your favour. While a non-random system is about optimization in a completely different way.
Of course with non-random and random I don't mean completely so. But take a system like Ogre Battle which is essentially non-random. It does have attacks occasionally miss but everyone's actions and setup are completely non-random at the start of a skirmish. It creates a certain kind of strategic planning.
While a more random game like Blood Bowl is all about risk-management. Doing the least risky things first in a turn (or occasionally doing the things that are the most critical that they succeed) and trying to mitigate the damage that the randomness undoubtedly will cause.
I guess I don't know where I'm going with this at all. But I guess transparency. And also maybe some way of pre-planning to assess risks in any encounter. Having randomness means that a battleplan doesn't survive contact with an enemy but it's still always good to be able to create one. Or something.
Also the concept of "rerolls" is a good thing to have in random systems. If a simple, but critical roll fails it's always good to have some kind of limited system of redoing something and hope for a better result. A player is always more likely to try a 50% chance if he will have another 50% chance if the first one fails. And limiting them might mean you wasted one on a failed 75% chance earlier and now you no longer know if the 50% is worth it. Which is all part of the fun of risk-management.
like shinan and i have both said, transparency is just. so important.
basically establish clear rules -- mechanical creatures can't be poisoned. Toxic Burst deals low damage and has a 70% chance to inflict Poison. but it will never poison the robochopper 2050, and that's okay if the player has been told "mechanical creatures can't be poisoned." meanwhile, Toxic Burst is a good choice against those berserker brutes and those flesh golems! it might not always stick, but at least it deals a little damage regardless and usually helps you focus on lesser foes before taking care of a major enemy.
if you want to have one of those sorts of abilities that MIGHT deal incredible damage, just give it a "oops i fucked up" conditional. a lot of daily abilities in D&D 4e do this -- on a miss, they do half damage and/or apply a reduced effect. you could easily do this with an area storm spell that has a 25% chance to inflict Paralysis, and a 100% chance to inflict Slow. makes sense, right? you still get a positive outcome, just one of the outcomes is more positive than the other.
basically establish clear rules -- mechanical creatures can't be poisoned. Toxic Burst deals low damage and has a 70% chance to inflict Poison. but it will never poison the robochopper 2050, and that's okay if the player has been told "mechanical creatures can't be poisoned." meanwhile, Toxic Burst is a good choice against those berserker brutes and those flesh golems! it might not always stick, but at least it deals a little damage regardless and usually helps you focus on lesser foes before taking care of a major enemy.
if you want to have one of those sorts of abilities that MIGHT deal incredible damage, just give it a "oops i fucked up" conditional. a lot of daily abilities in D&D 4e do this -- on a miss, they do half damage and/or apply a reduced effect. you could easily do this with an area storm spell that has a 25% chance to inflict Paralysis, and a 100% chance to inflict Slow. makes sense, right? you still get a positive outcome, just one of the outcomes is more positive than the other.
This is yet another topic that would benefit from some clear critical analysis and examples, with the end result being an article I can feature.
DFalcon
2141
One option is to use randomness to vary a starting state, or change the player's options rather than the results of his decisions (e.g., drawing a hand of cards every turn). The latter in particular can cause that same need for flexibility in a battle plan mentioned above.
If you do want to use straight favorable vs. unfavorable results, besides the strategies mentioned above you could have a side effect that only occurs on an unfavorable result. Maybe if an ability fails enough in a battle, you can make it a guaranteed success at a time of your choosing, something like that. Though you might have trouble coming up with a specific implementation that's not prone to abuse.
This reminds me about something I was really enthusiastic about at one point. Enough to use exclamation points!
In early versions of Aurora Wing I didn't let the player save except at the beginning of a turn. Since that was often 10+ attacks with 40-70% hit rates, even if you were reloading a lot you'd never get everything you wanted, so it was important to prioritize.
I dropped that because it would be incredibly painful if you forgot to save at the right time, but I remember liking the way it played. Though nowadays we'd probably just use a delete-on-load save instead.
If you do want to use straight favorable vs. unfavorable results, besides the strategies mentioned above you could have a side effect that only occurs on an unfavorable result. Maybe if an ability fails enough in a battle, you can make it a guaranteed success at a time of your choosing, something like that. Though you might have trouble coming up with a specific implementation that's not prone to abuse.
author=Shinan
While a more random game like Blood Bowl is all about risk-management. Doing the least risky things first in a turn (or occasionally doing the things that are the most critical that they succeed) and trying to mitigate the damage that the randomness undoubtedly will cause.
This reminds me about something I was really enthusiastic about at one point. Enough to use exclamation points!
In early versions of Aurora Wing I didn't let the player save except at the beginning of a turn. Since that was often 10+ attacks with 40-70% hit rates, even if you were reloading a lot you'd never get everything you wanted, so it was important to prioritize.
I dropped that because it would be incredibly painful if you forgot to save at the right time, but I remember liking the way it played. Though nowadays we'd probably just use a delete-on-load save instead.
