Personally, I've always loved the Samurai Shodown series. Particularly III and IV (though is it just me or was IV ridiculously easy compared to III?)

What makes the game system unique in this series (starting at SS III) is how you can dodge attacks by leaning to the side rapidly. This can be used against ANY attack, but requires very good timing. Someone with enough skills and reflexes could counter virtually any attack by dodging the instant the opponent moves and countering with something while they recover from their attack. This, of course, requires an unparalleled amount of skill.

An other good series by SNK with a similare system is Last Blade. I've only played LB II, though the first one has a similare system: you can smack your opponent's weapon out of the way if you time it right. This doesn't work against every attack, but it certainly keeps you on your toes and beats the heck out of cheap blocks.

The point here being that neither of these series boil down to a rock-paper-scissors system at all.

Edit: You can't predict your opponent's moves in rock-paper-scissors. ESPECIALLY when your opponent is a computer. It's pure luck because any and all figures have an equal chance of being displayed with no bias (though, with RNGs being what they are, that's highly debatable Seeing as we can't possibly hope to make even an educated guess as to what'll come out next, though, I think calling it random is right). Against someone, they might subconsciously favor, say, paper, but there's still no way you can put the odds in your favor even slightly significantly. Unless your opponent is really stupid.

[edited by - RuneLancer on April 12, 2004 10:28:38 PM]

Apparently, some people are pretty serious about playing
Rock-Papaer-Scissors on the computer.

RoShamBo Programming Competition

Some others argue that it''s a waste of time, because a random
strategy is the best strategy.

---

Kami no Itte ga ore ni zettai naru!

First, I agree that rocks-paper-scissors (RPS) isn''t the way to go. Number one, because as mentioned, it boils down to a guessing game.

Secondly, it''s just not how fighting works. One of the better roleplaying systems I''ve seen (which is turn based) is a indie game called The Riddle of Steel. You can download intro rules from their website at www.theriddleofsteel.net.

In a nutshell, each player decides secretly whether they will attack or defend. They represent their choice by throwing a red die for an attack, OR a white die for defense. When the GM calls out, each player throws their die.

If both players choose to attack, then the character with the highest reflex rating goes first. This is bad news for the slower guy, because damage applied is immediate. In other words, if the slower player takes damage, he must attack with any penalties associated for being wounded.

If both players choose to defend, then they circle each other like hawks.

If they differ, then the attacker and defender secretly decide again how many dice they wish to use in the upcoming exchange (TROS uses a combat pool which is a number of d10 based on your skill proficiency, attributes and other modifers). SO for example, Player A might have a combat pool of 12, and player B might have 14. Player A decided to attack, while player B decided to defend. Player A commits 7 of his dice to attack, while player B commits 8 dice to defense. During the declaration, the attacker must state not only how many dice he is committing, but also the mode of attack and where he is aiming (they use a fencer''s quadrant position).

Now each player rolls their dice pool against a certain target number. The target number and dice pool are also affected by what manuever each combatant chooses. For example, for a defender to attempt a counter, he subtracts 2 dice from his pool. The target numbers are determined by weapon type and type of maneuver chosen (for example a lunge or a slash with either a falchion or a rapier will have different target numbers). The dice are rolled and every die is compared against the target number. For every die that matches or beats this target number, that''s a success. Whomever has the most successes ''wins''. So if the defender rolls more successes than the attacker, he successfully blocks. The more successes an attacker achieves, the more damage he does.

Normally the order of battle is only determined in the first exchange. After that, whomever has the most successes gains the initiative.

Now, I changed the rules a little bit, and here''s how I do it. First off, there''s little incentive to defend. In attacks, multiple successes means more damage, but with few exceptions, multiple successes for defenders mean nothing. So I changed the rules to mean that every success for the defender is an extra die gained if they choose to attack in the next exchange.

I also changed the first round of combat. The player must select two dice...either both red, both white, or one of each. They also must add the numbers they roll. The highest tally goes first. If the defender wins the initiative, he may chose to pre-empt the attack by sacrificing half his dice, but switching to attack mode instead, or he may switch to using a counter. Another thing I changed was if both players attacked, if the margin between both for initiative was 5 or less, then they attacked each other simultaneously (I didn''t buy the designers excuse that there was no such thing as simultaneous attacks...the Japanese coined a term for this very act, Ai Nuke, ''mutual desctruction'')

A system like this would be fairly easy to implement on a computer system. The Riddle of Steel is very much oriented around fencing, but you can figure out your own manuevers.

One thing to remember. Attacks and defense aren''t inherently balanced. Just think of all the factors that go into a manuever:

Speed: How quickly can you execute the technique?
Accuracy: How accurate can this technique be applied?
Power: How much force over time can be exerted?
Openness: Will executing this technique leave you vulnerable?

Also remember, there really are no ''counters''. What are called counters can be applied to many varieties of techniques, not just one or two. So there really is no such thing as Rocks-Papers-Scissors. If you watch anime or hear fencing stories, ("Aha, I see that you are using the Agrippa style...I will easily counter that with my Ballistrade-Fenestra!!") that''s really just fantasy stuff. Also many maneuvers change depending on how they are executed. For example, most people think that the karate ''chop'' is a strike. It''s really meant to be a block that hits the attacker right at a golgi tendon, causing the reflex to be activated. But in a pinch, it works as an offensive strike too.

quote:

---

Original post by tieTYT

quote:

---

Original post by dgaf

quote:

---

Original post by tieTYT
Almost every fighter has a rock-paper-scissor aspect to it. Even with rock-paper-scissor you can anticipate to beat your opponent.

---

But not every fighter is fundamentally a rock-paper-scissors game.

There''s a level of uncertainty inherent in a fighting game (or combat, for generalization purposes), but the core play cannot revolve strictly around the roll of a dice or the flipping of a coin. Such activities are anti-interactive.

---

But a rock paper scissor game is not that way either. I''m sure it is the first time, but if you end up playing long enough you will be able to anticipate your opponent.

I think we may be thinking of different things. The rock paper scissor aspect in Soul calibur 2 is that mids hit people ducking and lows hit people standing. And you have that horizontals track, but verticals can break through horizontals, and that B''s break A''s, but you can avoid them by stepping.

---

Yeh, that was the way I was thinking about it.

Let it be known that I in no way think that the paper-rock-scissors method is the best simply because I chose it, and it was just a random thought at the time.

Perhaps one of the biggest problems with the RPS system is that it invariably puts the characters on equal footing. No matter what the other guy does, you''ve got a 33% chance of victory, a 33% chance of defeat, and a 33% chance of a draw. Take a look at the simplified little fighting system in the InuYasha flash game at Adult Swim''s webpage for a slightly modified approach. Not only do you get to fool around with distance and position for evasion, etc., but you can choose to invest more or less in the individual moves you execute. It''s a neat system, and worth a look. The downside is that it''s highly turn-based, but the action queue makes it go pretty quickly, and forces you to anticipate far enough ahead that some real strategy comes into play.

Every fighting game works off of a rock paper scissor system. Sometimes it may have multiple RPS''s in the same fighting game and almost always one is weighted more heavily than another (it''s better to use rock than the other two for example).

For example, in any 3D fighting game, as soon as i''m in reach, i would theoretically be as likely to go for a mid attack as a low. But since mids are weighted so heavily in favor of lows in every 3D fighter i''ve ever seen, the game does not boil down to random guesses. This adds depth.

If every fighting game were to come down to RPS, then every fighting game would be random.

You can structure the fighting system to work off of the idea that no one path is greater than "all others", but to base the entire experience on RPS (including chosing actions) negates any skill that a player may posses. He is no more likely to beat anyone that he plays.

RPS means randomness. A structure in which no one way is "always" the best way is not the same as randomness. RPS is based on that idea of an "always" condition in result, but the actual act of picking your path is as random as flipping a coin.

There are two elements to RPS: the choice and the result.

The result is based off of "no one way is greater than all others". It is evenly balanced - 33/33/33%. No one option is more enticing.

The act of picking my move is a random choice. You may be able to anticipate someone''s move, but you cannot know until it''s over.

Just because I have two options of punching and one is "more favorable in reward", it does not mean that I will choose that punch. I could choose the less effective punch because I ancitipate that you expect me to punch you in the more effective way.

It''s worth re-iterating that in a R/P/S system, AI opponents can be absolutely infuriating, since they have neither strategies nor habitual actions. When a buddy and I play MK:DA (which 0wz0rz Soul Calibur II), we eventually reach a synchronized state, where our opening attacks can be anticipated with some regularity, and our fight get longer and more intricate. When a third person gets involved, we either win or lose in a big hurry.

Just one more reason that although an RPS system is "easy to implement", you get what you pay for

quote:

---

Original post by dgaf
If every fighting game were to come down to RPS, then every fighting game would be random.

---

I didn't mean literally RPS, my last reply clearly explains that. But the concept of RPS that all 3D fighters implement is the RPS concept of A beats B, B beats C, C beats A. Please keep in mind that this is the RPS aspect that fighters use, not the random factor.

[edited by - tieTYT on April 13, 2004 9:52:23 PM]