RANDOM NUMBER GENERATION: THE DEATH OF THE CRITICAL HIT

Posts

author=Fallen-Griever
...
Fighting games do not require more strategy than RTS games. It is entirely possible to beat decent fighting-game players by button mashing, you will not defeat a decent RTS player by just building a bunch of random shit.


I don't know, I've actually won that way several times. It depends on how well the game is designed, really. Button mashing goes nowhere in Brawl and I've played RTSs where I just build a bunch of random shit and somehow won.
Fighting games do not require more strategy than RTS games. It is entirely possible to beat decent fighting-game players by button mashing

Not going to happen. Sorry. I know people like to say that about fighting games, but if you can consistently (or even one or two matches, really, outside of a crazy ass lucky shot) beat someone in a well made/balancing fighting game by button mashing, that's not a decent player.

Certainly RTS games require a different kind of strategy (and hell, in a lot of cases more strategy, period) than the fighting genre, but not by a wider margin than most other genres, and everything I said still applies and you're pretty much never going to win a fighting game match with a player who can call themselves decent with a straight face with button mashing. Some games in particular (Street Fighter IIIrd Strike, Tekken) it's a particularly useless strategy due to the range of options available to skilled players to counter and punish dumbshit moves and whiffed attacks.
On the other hand name a genre where you can beat a player who can call themselves decent with a straight face consistently using a random strategy.

EDIT: (except for chutes and ladders)
author=LockeZ
"Players with higher rankings win a vast majority of the time"? What does that mean? Chess doesn't have XP or levels. The only rankings in chess are your win/loss statistics. Basically you're just saying "the people who have won more often tend to win more often." Well, duh.


http://en.wikipedia.org/wiki/Elo_rating_system

If you are going to say something asinine, at least be informed. It' a systematic measure of skill level, and has little correlation with win/loss ratios.

Anyway, chess is a PVP game. In PVP, the game has to be perfectly balanced. In a single-player game, that doesn't have to be true. The rate at which the player is supposed to win needs to be balanced against your game's intended difficulty, against how often the player has access to save points, and possbily against other factors as well. If every single battle in Final Fantasy Tactics feels like an even match, then ok, no problem - I can save after each battle. It can even be harder than that, and I'll still enjoy attempting each battle 3-5 times before winning. But if every single battle in Final Fantasy 7 feels like an even match, then whoa holy shit suddenly I have less than a 0.1% chance to make it from one save point to the next without a game over.


Three of the four genres that anaryu mentioned - RTSes, FPSes and fighting games - are PVP-centric games. This does not make anything stated in this topic less valid. Additionally, the comparison to FFT and standard RPGs is odd, because you are assuming RPGs are only about skill whereas that is pretty much never the case (levels/equipment/primary and secondary knowledge). It also assumes that the player and AI have an equal skill level, which is, of course, silly.

Really, I don't even know what point you are trying to make.

Fighting games do not require more strategy than RTS games. It is entirely possible to beat decent fighting-game players by button mashing, you will not defeat a decent RTS player by just building a bunch of random shit.


Hahahaha no. I will admit that I am not a fan of the elitist fighting game bullshit, but this is incredibly misinformed.

Fighting games are disgustingly elitist/competitive and it requires some months/years of practice with them to stand a fighting chance against even mediocre tourney players. It goes hand in hand with them being the slowest to evolve of any videogame genre, even slower than FPSes/RTSes.
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=Karsuman
Three of the four genres that anaryu mentioned - RTSes, FPSes and fighting games - are PVP-centric games. This does not make anything stated in this topic less valid. Additionally, the comparison to FFT and standard RPGs is odd, because you are assuming RPGs are only about skill whereas that is pretty much never the case (levels/equipment/primary and secondary knowledge). It also assumes that the player and AI have an equal skill level, which is, of course, silly.

Really, I don't even know what point you are trying to make.


Bringing those genres into the argument is equally asinine, if you're considering the PVP portions of them. The PVP parts do not apply to our discussion about critical hits and random chances for the same reason that chess doesn't apply.

Comparing PVP fairness to PVE fairness is irrelevant. They're fundamentally different concepts of fairness, because in one you have to make sure both sides have fun equally, and in the other your goal is to maximize the fun of one side at the expense of the other (because the computer doesn't need to have fun). We're making RPGs, so PVE is the type of balance we care about. So stop talking about PVP games! None of the points you make with them are going to translate meaningfully to RPG design.
If you have a complaint about genres that emphasize PvP being discussed, take it up with ana, as he was the one that mentioned PvP games in force. The topic evolved from there tangentially and the discussion has evolved a little. What a surprise! Never seen that happen before.

Additionally, even the OP mentions non-RPGs in regards to RNG, and not everyone here just makes RPGs (Anaryu, for example). I mentioned Chess as an example of a game that entertains many, does not rely on reflexes, and elliminates RNG as much as it possibly can. And even then, you are ignoring the fact that many chess programs come with highly intelligent and competitive AIs that can put up a good fight, even against masters. And yes I know people who play against chess AIs for fun. So there's your omgPVE.

Stop trying to arbitrarily one up me and get over yourself.
LouisCyphre
can't make a bad game if you don't finish any games
4523
I've often toyed with the idea of rates in the place of chances - having results that occur with a fixed distribution. For example, inflicting Burn every third use for a fire-damage spell.

One way to handle this would be to have some sort of counter - when it hits some arbitrary amount, the result occurs. Maybe you add 40 out of 100 to a Burn counter. If the enemy resists Burn, you might have to hit 200 to inflict it. At that point it does get very technical and that's usually the point I drop it - invisible counters for every normally randomized effect isn't sleek or intuitive at all.

It's still a thought, though.
author=ChaosProductions
I've often toyed with the idea of rates in the place of chances - having results that occur with a fixed distribution. For example, inflicting Burn every third use for a fire-damage spell.

One way to handle this would be to have some sort of counter - when it hits some arbitrary amount, the result occurs. Maybe you add 40 out of 100 to a Burn counter. If the enemy resists Burn, you might have to hit 200 to inflict it. At that point it does get very technical and that's usually the point I drop it - invisible counters for every normally randomized effect isn't sleek or intuitive at all.

It's still a thought, though.

City of Heroes used something like this - your status resist was based on a status resist factor (based on your skills and defenses) - each status effect had a rating that would sum with other status effects to try to beat your resist.

When someone hit you with a status effect, as long as the sum of your existing status effects + the new status effect was less than your resist it wouldn't affect you. If that one was equal to or greater, that status effect would be in effect on you (but not the others still.)

When a status effect duration wore off it would be removed (even if it wasn't affecting you) - so if you had 3 status effects with a rating of 1, and your resist was 3, you'd be affected by the last status effect until the first wore off.

If you had 1 status effect from a boss that had a rating of 2, then even a rating 1 status effect would take effect when it hit you, so more status effects could be more powerful separate from their duration and what the actual status effect was (speed, poison, etc.)
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
I still don't think chess against a computer is a relevant comparison to RPGs, because even if you're playing against the computer, you still have a PVP mindset - you're okay with the idea of losing even if you did your best. This is because chess is really a PVP game, and any PVE versions of it are just a gross imitation designed to mimic the PVP, and the player knows and understands this.

In a true single player game, if you lose, it should be your own fault. If you used utterly optimal strategy and did everything perfectly, you should not still sometimes lose. People will put up with it in PVP, because they understand that both people have to win sometimes. And they'll put up with it in games that *feel* like PVP, because it's kind of a mind trick. But they quite reasonably won't put up with it at all in an RPG.

City of Heroes used something like this - your status resist was based on a status resist factor (based on your skills and defenses) - each status effect had a rating that would sum with other status effects to try to beat your resist.

When someone hit you with a status effect, as long as the sum of your existing status effects + the new status effect was less than your resist it wouldn't affect you. If that one was equal to or greater, that status effect would be in effect on you (but not the others still.)


This is actually a pretty good system. Far better than ordinary resistances that cut the status hit rate down from 20% to 10%. Letting the player obtain old-fashioned status resistance doesn't really help at all, unless they have a way to obtain complete immunity, or unless it's still always possible (if much harder) to win even if you get afflicted with every status. CoH's system sounds like it makes sure you can always avoid dying to these deadly effects if you prepare properly.

Randomly getting a game over 1% of the time is not really better game design than randomly getting a game over 5% of the time. It still has the exact same game-crippling problem, that problem just shows up a little less often. I mean, ultimately, in either case, you could just make it so that when the battle starts, it has a random chance to not even let you attempt to fight, and just instead instantly gives you a game over for no reason. And your game would be exactly the same, except with less animated graphics.
slash
APATHY IS FOR COWARDS
4158
This reminds me of how Blizzard designed the matchmaking & rating of the Arena system in WoW:

"You should always be winning about 50% of the time. Yes, everyone would like to win 70% of the time, but that means someone else is losing 70% of the time, and that sucks."
So you go up in Rating when you win and fight better people, where it boils down to:
"You should always be winning about 50% of the time, but you'll be winning against better opponents."

That being said, if PVP is being set up right (in Chess, WoW, or anything) both players should win about 50% of the time (assuming around equal skill level). And out of 100 or 1000 matches, it will probably be close to 50% each, even with random elements (applied equally to both players).

However, when it comes down to a singular chess match and it's intense on both sides and then suddenly Black picks up a Blue Shell, you're boned and you feel kind of cheated.
author=Feldschlacht IV
Everything doesn't have to be measured exactly. Variety, luck, and unexpected outcomes is the spice of life. What are you, a robot? Why does luck and skill have to be mutually exclusive?
I agree with this post. I suppose it's a thing of taste, but I think it's interesting to have a bit of uncertainty going into battle.
While RNG may not work for say, Street Fighter, certain games do benefit from randomness.

Would Yu Gi Oh or Poker be the same without randomness?

EDIT: Like with anything, it's a thing of balance.
I once played a game of Risk where, despite vastly superior strategy and diplomacy, I lost because I lost 15 battles in a row, despite having an extra die. 15. In a row. Four and a half hours of trickery and masterful strategy and I lose because, and only because, of luck.

Fuck Risk, I'll never play it again. I have a big, big problem with games based on a roll of the dice. I've never been a gambler, but that game of Risk meant I will never trust anything to a roll of the die ever again.

And yes, a person who randomly smashes buttons in a Fighting game can win against decent and even, sometimes, good players. I know MOG likes to think that they are some masterwork of skill and the TRUE WARRIOR SPIRIT (and as soon as I read someone saying luck could overcome skill in a fighting game I knew MOG was going to pop in), but lots of fighting games aren't balanced as well as Street Fighter.

In PVP matches, luck is already present: there is no need for the game to add it. If my strategy in any PVP game is good and my enemies is good as well, then it comes down to whoever happens to step into the wrong move or use the wrong attack just once to turn the tides of battle.

Don't get me wrong, a little bit of luck is fine in player vs enemy battles. Doing 80 damage sometimes and 120 other times is alright, it means you can have a good round and a bad round. But to be able to lose a battle or an entire game because someone has rolled some dice after you meticulously build up your strategy is unacceptable.

moral of the story: fuck luck
moral of the story: fuck luck

More like fuck Mario Kart.

And yes, a person who randomly smashes buttons in a Fighting game can win against decent and even, sometimes, good players.

Then decent players of which you have seen not. I know decent fighting game players, you can't win this way unless the fighting game in question is horribly balanced. And even then, I've still seen actual decent players win most of the time even in broken fighting games.

And I've won RTSs by building random shit probably because I still kind of use actual tactics and kind of build what I need at the time even if I do spam buildings. Just because luck sucks in Risk doesn't mean it will suck in every game except Mario Kart I hate those damn item boxes. The only way to win all the time in that game is to pretty much just hack it up like 80% of everyone I played online did.
rabitZ
amusing tassadar, your taste in companionship grows ever more inexplicable
1349
kaempfer you lost because you didn't do this before rolling the dice:

I love how in New Vegas your luck stat actually lets you win at the casino. I can just click random cards at Blackjack and still get kicked out of the casino for maxing out the jackpot if your Luck is near 8 or 9. You can play stupid hands and you'll just win somehow.
And yes, a person who randomly smashes buttons in a Fighting game can win against decent and even, sometimes, good players. I know MOG likes to think that they are some masterwork of skill and the TRUE WARRIOR SPIRIT (and as soon as I read someone saying luck could overcome skill in a fighting game I knew MOG was going to pop in), but lots of fighting games aren't balanced as well as Street Fighter.

Of course fighting games aren't the 'true bastion of skill' on a fundamental level; that's why I applied my argument to decently designed fighting games. If the game isn't designed well, it caters itself to poor playing. Sure you might be able to button mash it up in like, Eternal Champions or the old Mortal Kombats or whatever, but like I was trying to illustrate before, the grand majority of fighting games that are actually taken seriously by the community at a competitive level (King of Fighters, Tekken, SF, Guilty Gear, etc), no, like I said before, you can't button mash against a truly skilled player. The games where you can, generally don't matter. Skill rules all.

But yeah I'm just being anal again; you know how I get about dem fightin gams
author=Feldschlacht IV
Sure you might be able to button mash it up in like, Eternal Champions or the old Mortal Kombats or whatever, but like I was trying to illustrate before, the grand majority of fighting games that are actually taken seriously by the community at a competitive level (King of Fighters, Tekken, SF, Guilty Gear, etc), no, like I said before, you can't button mash against a truly skilled player. The games where you can, generally don't matter. Skill rules all.

You forgot Virtua Fighter, my friend. The series is far more balanced than most fighters, even Street Fighter. It's such a shame it's so buried under the Tekkens of the world, however. You'd be hard pressed to find a button masher succeed at Virtua Fighter.
Yeah, Virtua Fighter is insane. I'm pretty sure that's pretty much the Chess of fighting games. Good luck button mashing there, or even winning by virtue of luck; if you don't have skill on your side, you'd have a better chance of praying for victory from an Aztec god or your opponent being buried under a rockslide in the middle of the match or something.
author=LockeZ
Randomly getting a game over 1% of the time is not really better game design than randomly getting a game over 5% of the time. It still has the exact same game-crippling problem, that problem just shows up a little less often. I mean, ultimately, in either case, you could just make it so that when the battle starts, it has a random chance to not even let you attempt to fight, and just instead instantly gives you a game over for no reason. And your game would be exactly the same, except with less animated graphics.


Let me talk for a minute about Freecell. Its credentials as both single-player and depending entirely on randomness are, I think, impeccable. For a stretch I played it pretty often, several hundred hands with a win rate of around 98-99%. And one of the hands I encountered was Microsoft's random seed #11982, which is unbeatable - the only one of the 32000 randomly generated hands early Microsoft Freecell shipped with that is.

So mostly, I would say randomly getting a game over 1% of the time is better game design than randomly getting a game over 5% of the time. Randomness is a powerful tool to make things play out in interesting ways, if it's used right - even though that can mean every once in a while, someone just can't catch a break.