/sci/ - Science & Math

green sword green cuirass quick hitting with some dodge

Is this some sort of logic puzzle?

>>1835093
heres a hint, and no its more of a math puzzle

green sword is best by atleast 1000

Sword with highest rate of damage is green. Assuming everyone picks best sword, best armor is yellow. But I suspect that assuming everyone picks yellow armor, the best sword may be something different, which may in turn make another armor best, etc. etc.

fuck it, ill just solve this myself

Blue vs blue = 6336
Blue vs red = 6256
Blue vs yellow = 6400
Blue vs green = 6080
Blue wepaon avg = 6,268

Red vs blue = 6120
Red vs red = 6120
Red vs yellow = 6000
Red vs green = 6080
Red weapon avg = 6,080

Yellow vs blue = 6210
Yellow vs red = 6035
Yellow vs yellow = 6500
Yellow vs green = 5700
Yellow weapon avg = 6,111

Green vs blue = 6156
Green vs red = 6426
Green vs yellow = 5400
Green vs green = 6840
Green weapon avg = 6,205

Blue armor avg = 6,205
Red armor avg = 6,209
Yellow armor avg = 6,075
Green armor avg = 6,175

For weapons
Blue > Green > Yellow > Red

For armors:
Yellow > Green > Blue > Red


:. Blue weapon, yellow armor.

Full green. You do 9000 damage a minute, and with green armor you dodge 1/4th of all attacks.

Weapon choice depends on the probability of hitting with an attack. If the hit probability is 100%, I'd just go with the higher damage percent, otherwise I'd get the fastest one.

As for the armor, I'd go for green.

Is the subtraction of damage a percentage too, or are you literally just subtracting like 12/8/20 points from their attack? that seems tiny as fuck

blue sword and yellow armor.

green sword has highest DPS and blue has same DPS as red sword but armor makes green and red worthless.

yellow armor is best against everything except blue and yellow sword but no1 uses yellow sword cuz it has lowest DPS and u know all the noobs gonna spam the green sword.

depends on the incoming damage if less than 20 then the yellow is perfect armor if substantially greater than 20 then the dodge is better

ITT: /sci/ is horrible at math, and this failure of logic reminds me of the runescape game engines

>>1835120
(80-20)/80 = 20% dmg reduction

>>1835132

label your equation, how'd you get that?

>>1835132
25% i suck at math

Yellow sword/blue armor.

Blue armor because hell it seems the most balanced, yellow sword because fuck you I'm gonna run around in circles jumping the whole time and you aren't going to get your 180 attacks per minute in on me by a long shot.

this isn't purely a numbers game

In PvP burst damage > all else because you and your opponent won't just be standing around swinging at each other.

yellow sword

green armor

against green/green (probably the most picked option) the best option is
Green / Yellow

against blue/yellow (2nd most picked i assume) the best option is
Yellow/Green

againts red/yellow best option is
Yellow/Yellow

OVERALL BEST AVERAGE AGAINST ALL

Blue/Yellow

Green sword, red armour.

Hey guys, if you were playing rock paper scissors, which would be best to choose?

>>1835154
fucking finally

this doesn't really work..
what if all the creeps only have 40hp?
what if all the creeps only have 19 damage?

DPS =/= ATK (attacks to kill)

e) not enough info

>>1835162
ignore this didn't read PvP

>>1835162
>PvP

blue sword, yellow armor

>>1835154
Rock. You can readily punch the shit out of the other guy if he has the guts to pick paper instead of scissors.

Hey guys, post your faces when this is really just a sociology question

>>1835177
...?

>>1835181
All of this is situational. Each weapon is better against a different type of armor. So all you are doing is assuming what noobs are gonna pick. tracking human habits = sociology

guys, in PVP try, just TRY getting 180 hits in a minute, infact, with the yellow sword youd be VERY LUCKY to land 50 hits in a miniute

YELLOW SWORD, GREEN ARMOR

Depends on what kind of PvP it is. Is it like wow where you can physically dodge around them? Or is it like in runescape where if he can hit you, you can hit him, no way around it

Micky: Alright! Rock beats paper!
Kramer: I thought paper covered rock.
Micky: Nah, rock flies right through paper.
Kramer: Well... what beats rock?
Micky: ... Nothing beats rock.

>>1835110
this!!!
but this is based on the assumption that blue/red/yellow/green weapons are all equally owned.

I'd probably go green sword/ blue armor
{most people would probably go green/green}

Blue/green combination is the best. It doesn't matter about the average damage you deal to all types of armor. What matters is how many of the 16 combinations you can beat with your combination of armor. Blue/green wins 10, ties 2, and loses 4.

Post host is toast?

>>1835107
did you take in account that if you hit against yellow armor you deal 40% less damage while with the yellow one for example you only lose 12% of you total damage?

All of this is assuming that the skill of all warriors is on the same level.

This is never true. How do we account for this?

>>1836478
woah

Interesting

>>1836519

This is irrelevant towards the best combination to choose, is it not?

Are you referring to play-style?

>>1836519
You have enough information to make an educated choice of weapons/armor. Knowing that other factors might change the answer is important, but if we assume all is equal otherwise we can make a good choice.

For instance, if you only have 100hp then you would always choose yellow sword as you have a 76% chance of KO'ing your opponent on the first hit.

>>1836478
Doing it right. Someone should call /tg/, lots of them are in to character optimization.

Green because it's a katana and Katana can cut bullets

>>1836633
and tanks.

>>1836478
What am I looking at

So many fucking variables.
Just fucking choose.

I'd go green armour and blue sword.

I want to be killing lots of dudes so I really want to be taking as few hits as possible, definitely pick dodge over damage mitigation. And I want a sword that packs a punch for burst damage, but not the yellow because it's just too slow, I won't be able to use it in rapid succession to kill a bunch of dudes and if they manage to dodge my swing I'm fucked. Blue packs a punch and the fact that it's a little slower than the green and the red won't matter too much if it's being used effectively in the hands of a good player. I'd also carry grenades.

>>1836478

It's not the percentage of combinations you can beat, but the percentage of matchups you win. If everyone chose according to your chart, then everyone would pick Blue/Green and every matchup would be a tie. But if one person chose a Green/X combination, they would then win a matchup with any other player.

WHERE IS THE NASH EQUILIBRIUM??

Green sword is 9000 damage per minute.

Is the 24% dodge rate a CHANCE or is it constant? I doubt it can be constant, so it's most likely just each attack done on green armor has a 24% chance to be dodged

Which means, it's possible, that the green armor can dodge every single attack in a fight, making it the best.

But this also means it might not dodge often at all, making it the worst.

It just depends how often the dodge occurs.

>>1836745

It occurs 24% of the time, obviously. You're working with probabilities here, so everything you're doing is comparing curves, so your question isn't valid here.

Too many variables aren't covered. Is it 1v1, 5v1, 10v10? Do you get points based on getting the last hit on someone or just doing damage? How much "health" do you have? How long does the fight last?

stupid question is stupid and narrow-minded.

>>1836745
This is probably the stupidest fucking shit I've read in my entire life. My fucking mind melted three times while reading your post it's so fucking retarded I wish you long and painful testicular cancer.

Sincerely yours, a concerned citizen.

>>1836764
Actually, if you're going for a gamble, and the fight will only be once (OP never specified how many trials it will be), then green armor would be the best, but that's a gamble. It could end up being the worst.

>>1835078
Not this shit again

>>1836772

>>1836772
11/10

get blue sword and blue or yellow armor, kite all the faggot noobs who went for green

>>1836478

mfw I choose green/green because everyone else chooses blue/green

>>1836772
This

>>1836764
Hes right you asshole
Someone with a 25% dodge would be more likely to dodge all of the attacks making it the best armor, but at the same time you cant rely on that because there is a chance you will not dodge any of the attacks.

Red sword and green armor

Faster attacks reduces the effect of missing but green sword is too much of a drop off in damage.

Green armor means ~1 in every 4 attacks should miss which is more damage reduction than 4 successful hits against the other armors. (except green or redS versus yellowA)

>>1836772

That's why you do statistics. You don't know who you're facing or the exact sequences of dodges/non-dodges, so you take an average.

Feel free to do a set of Monte Carlo simulations and report back if you're really that curious about variations in performance. Make assumptions about hit points (more HP --> the average winner will win close to 100% of the time; fewer HP --> wildcard green armor may turn the tide against a more powerful opponent in a substantial number of trials) and let us know what you find.

Enjoy your weekend.

>mfw scrubs go for the RNG dodge armor instead of the consistent -dmg armor
Have fun praying for good luck, baddies.

>>1836803
Have fun when I dodge 10 of your yellow swings in a row faggot.

Elitist Jerks

In my /sci/?

I would wear the blue armor and use the green sword, but I would also keep all the other armors and swords in my inventory.
/thread

Swords:

Green = 9000 d/m
Blue = 7500 d/m
Red = 8000 d/m
Yellow = 8000 d/m

Armor:
(Damage: 50 Speed: 100 = 5000 d/m)

Green = 3800 d/m
Blue = 3420 d/m
Red = 3570 d/m
Yellow = 3000 d/m

Yellow armor and green sword for me.

>>1836825
>mfw
Are you... me?

>>1836724

WTF?
Multiply the speed with damage and see how much you can damage by swinging. But the there comes the variable of how often will it miss or be dodged. So green sword and blue armor in this case.

>>1836834
you didn't calculate dodge in either situation

>>1836847
You didn't even calculate it.

ok, so i put it in a spreadsheed...

the sword that deals most damage on average is GREEN at 9589.5 dpm
the armor that receives least damage on average is GREEN at 6176 dpm

>>1836950
How much damage a sword deals depends on which armor you're up against.

And your numbers are wrong.

And this thread is dildos.

Green sword and armor obviously. Even at just 100dps the green armor still negates more overall damage than anything else, and you have to assume the dps would be higher as opponents gain more attack power, so the flat damage mitigation is pretty useless.

stick the yellow sword up your ass, not even the red armor can help yoyu babies.

>>1836964
my numbers are fine:

sword x armor dpm
bb 8064
br 7344
by 9600
bg 6080
rb 8280
rr 7480
ry 10000
rg 6080
yb 7290
yr 6715
yy 8500
yg 5700
gb 10044
gr 8874
gy 12600
gg 6840

average sword dpm
b 7772
r 7960
y 7051.25
g 9589.5

average armor dpm
b 8419.5
r 7603.25
y 10175
g 6175

I are mmo nerd.
It depend on average points of health

>>1836478
this is why you're dumb.
As soon as a large number of people pick the same equipment combo, the smart players will pick whatever is good against that combo. The game is staying ahead of trends, not picking some optimal configuration (there IS no optimal configuration).

>>1837002
>average sword dpm
>g 9589.5
50x180 = 9000 dpm. That's the most it can do against 0/0 armor. The rest of them are wrong for similar reasons.

People on this board are morons.

>>1836478
Fuck year! I want red sword blue armor!

I would choose blue sword and green armor

to answer the question red sword yellow armor look the best out of those

average damage of weapons = 95
therefore dodge > dmg.reduction
common sense dictates green armor is better in this case, which (if played by not stupid people) most of the population will choose. considering this, then also the green sword would be considered the best alternative.
if not counting in the choices of other people, just averaging, then i'd go for blue sword- green armor.

damnit, this thread again?

BLUE SWORD, YELLOW ARMOR

>>1835078
Yellow armor, Green Weap.
DPS and Best Mitigation, Dodge is RNG.

>mfw 180 attacks per minute

RNG says that 15% is hit capped... /trollface

Obvious rpg rules:
- Slow/Hard-hitters perform best against damage soaking
- Fast/Weak-hitters perform best against evading chance
Math provides further proof of this, as green sword sucks against yellow armor and yellow sword sucks against green armor.

Further calculations demonstrate that RED SWORD has the best average damage when confronting anyone without a YELLOW SWORD, leaving the top place to GREEN SWORD in that scenario. YELLOW SWORD has the worst average damage against all kinds of weapons and armors, even when adding the parry bonus as extra effectivenes.
That said, it's worth noting that YELLOW ARMOR has always the best reduction rate regardless of the sword chosen, meaning that the vast majority of players will wear it. This means that, at this point, the average damage against yellow armor weights far more than average damage vs other armors, making the YELLOW SWORD the best choice in this case. But we said before that the GREEN ARMOR is best against YELLOW SWORD.

In the end, players will split into three categories: Ys/Ya (main strategy), Ys/Ga (counter strategy), Gs/Ga (double counter) and Gs/Ya (triple counter). Main strategy is also a counter to triple counter. Rs/Ba has the most consistent dps and damage reduction, doesn't counter anything and can't be countered. Any other combination is statistically worse than these.

Calculations in this post are assuming the following postulates:
- 1v1 pvp scenario
- All players have the same choices;
- No movement involved;
- Infinite amount of HP, the highest damage dealt after 1 minute wins;
- Parry rolls apply after dodge rolls and don't stack with each others.

>>1840244
What are the rules? Are we to assume our opponents are all completely rational players capable of making logical decisions or that there will be a random distribution of swords and armors among opponents?

This reminds me of my Diablo 2 days.

I think I just vomited.

>>1841729
start with the assumption that everyone starts with blue/blue and changes his strategy rationally to counter that and the following distribution. assume that for everyone.

game theory.

This so HEAVILY depends on the EXACT NATURE of the MMO that the question is COMPLETELY MEANINGLESS.

If this were Old-School EverQuest-style (read: 3D MUD), it'd be Green Armor and Green Sword.

Imagine two weapons with the same DPS. One attacks 15 times per minute. One attacks 60 times per minute. They both have the same DPS. If you have the same chance of missing with each weapon, THEN YOU CHOOSE THE ONE THAT HITS MORE OFTEN. Missing with a weapon that attacks more frequently and has the same damage per second as a weapon that attacks far more infrequently hurts you less. Missing roughly half of your attacks with each weapon means that, ON AVERAGE, you will deal MORE DAMAGE with the weapon that attacks FASTER.

I am not going to do a proof of this. I have played enough EverQuest and enough MUDs to know this for a fact.

Secondly, armor. The Green Armor is the Best, as you will AVOID ALL DAMAGE ROUGHLY 25% of the time. 24% reduction in TOTAL DAMAGE taken is higher REDUCTION IN TOTAL DAMAGE TAKEN than with the other armors.

>>1841761
Same DPS + Same Accuracy + Different Hit-rate = Same damage

Every time

The only reason to pick one that hits more often with less power is if the enemy has an HP that would be wasted with more power. (Doing 10,000 damage with 1% accuracy is pointless when the enemy only has 100 HP)

green sword yellow armor

There are a lot of retards in this thread. >>1841711 is the only one to get it right.

>>1841777

No, it doesn't. It only evens to same damage ON AVERAGE OVERALL.

In SPECIFIC instances, you will have more or less misses with one.

Missing 15% of your hits with the 60APS weapon gives you 51 hits. Missing 85% with the 15APS weapon gives you 12 hits. Assume that each weapon deals 60 damage per second. The 60 APS weapon therefore deals 1 damage per hit. The 15APS weapon therefore deals 1 damage per hit.

<span class="math">51\times1=51>48=12\times4[/spoiler]

You fucking retard.

>>1841796

Should read Missing 15% with both.

and 4 damage per hit for the 15APS weapon.

Need more amphetamines.

>>1841796
Same DPS + Same Accuracy = Same damage

Every time, math-illiterate troll.

Every time.

>>1841814

Actually, I seem to understand AVERAGES and STATISTICS better than you.

The FACTS are that MISSING with a HIGH DAMAGE/LOW ATTACKS PER MINUTE WEAPON is WORSE than missing with a LOW DAMAGE/HIGH ATTACKS PER MINUTE WEAPON that has the SAME DPS.

red armor, green sword

>>1841828
Yeah, except you're missing less often with a high power weapon, thus same end damage is being dealt.

Even a child could understand this. FFS, by definition of Damage Per Second.

>>1841841

You don't seem to understand the concept of AVERAGES. Or that you can only have INTEGER hits.

>>1841849
>Frank and Marsha both eat 1,000 calories, but Frank took more pauses between bites, therefore Frank has consumed more calories.

O Math Illiterate Troll, U SO CRAZYY

>>1841864
A and B eat 1000 cal over a day
A has 10 meals of 100 cal each
B has 2 meals of 500 cal each

They each miss 1 meal

>>1841869
Then they apparently had different accuracies.

It depends on how much health players have.

Assuming it's very large:

>>1836478

/thread

>>1836791
Law of large numbers. Learn statistics.

>>1841864

You are a fucking moron.

And yes, you are trolling.

AVG DPM of HIGH DAMAGE WEAPON: 28.039603960396
AVG DPM of HIGH ATTACK RATE WEAPON: AVG 29.5742574257426

Go die in a fire in the Math Department of whatever University failed to teach you about real world applications.

>>1841935
>Everquest
>Real World

>>1841942

It's a real world application. My MATH proves you WRONG.

I hope you enjoy posting what appears from the thumbnail to be bestiality. Does your asshattery make you feel better about being proved a fool?

Let's spice things up a bit

>>1841935
Same DPS + Same Accuracy = Same damage

Every time, math-illiterate troll.

Every time.

>DURRR, different DPM means different DPM
Thanks for the newsflash.

>>1841954
Crimson armor green sword

Fucking powercreep

>>1841955

No, no it doesn't.

Learn to INTEGERS.

>>1841954

Define the effects of a Critical Hit and a Parry.

>>1841970
If going by tradition:

Critical hit: Double damage
Parry: Identical to dodge, but found on weapons.

>>1841954
OP here
Critical hit: 2x damage
Parry: full damage avoidance, same as dodge

>>1841954
Crimson sword, crimson armor. Dat objective damage reduction + crit reflection if you're lucky.

>>1841967
10/10 great troll is great. I am enjoying this conversation.

>>1841987

Keep coming. I love a Sunday morning doing Math.

>>1841989
Well then, I'll bite. Until you understand what "double precision" means, you should check your calculations for rounding errors.

>Missing 15% with the 15APS weapon gives you 12 hits.
> 15*0.85 = 12

>>1841994

Duh. I am rounding down.

What are you failing to understand here?

>>1842021
That how the pros do it or something?? Truncate everything?

>>1842021
I'm failing to understand why you round down.

>>1842026

YOU CANNOT HAVE PARTIAL HITS.

blue and blue

>>1842045
Wait! Wait. I think I got this. So if I hit twice with a 50% chance to land it, I will actually never hit because 2*0.5 = 0.

Makes perfect sense now.

>>1842054
lol'd

>>1842045
Wait! Wait. I think I got this. So if I hit twice with a 40% chance to land it, I will actually never hit because 2*0.4 = 0.

Makes perfect sense now.

>>1842045
The question was why are you truncating instead of rounding. Lose the cruise control and pay attention, trip-fag.

>>1842026
>>1842044

Take a look at this.

You see those STEPS on the graph?

You either HIT or you MISS. Number of Hits is an INTEGER.

>>1842066
>>1842061

I am ROUNDING. Not TRUNCATING.

<span class="math">15\times .75=11.25[/spoiler]

However, Hits are Integer. THERE ARE NO PARTIAL HITS.

<span class="math">15\times .75=\mathbb{Z}11[/spoiler]

>>1842083
No, you're truncating. It was 15*0.85 = 12.75 which in that case rounds to 13. Then by your own example in >>1841796 the slower sword wins because

51*1 = 51 < 52 =13 * 4

Osnap. Do you understand how you're doing this wrong yet?

>>1842103

It rounds down.

TWELVE HITS + three quarters of a hit is TWELVE HITS.

YOU CANNOT HAVE A PARTIAL HIT.

>>1842111
You're aware that rounding down = truncating... right?

>>1842123

You're aware that when dealing with a binary outcome, you cannot have fractions, right?

Also, truncating is not the same as rounding, otherwise

<div class="math">37.45\underline{921}=37.46
37.45\underline{921}=37.45</div>

>>1842139
>>1842139
<span class="math">37.45\underline{921}=37.46[/spoiler]
<span class="math">37.45\underline{921}=37.45[/spoiler]

Can't newline.

> mfw I visit /sci/ for the very first time in my life and I stumble upon this thread: a thread nerdier than anything I've ever encountered on the entire fucking internets

>>1842111
Your blabbering about partial hits is completely irrelevant to the problem. I am hitting 15 times. The probability to hit is 85%. I repeat this as many times as I want to. Sometimes I will hit 12 times, sometimes I will hit 13 times. Sometimes I will hit 0 times, and sometimes I will hit all 15 times. On average, I will hit 12.75 times.

Same DPS + Same Accuracy = Same damage.

Every time.

>>1842150

You cannot have a partial hit.

You can have a glancing blow.

You can have the impact of a hit be absorbed by your armor.

BUT YOU CANNOT HAVE A PARTIAL HIT. IT EITHER HITS OR IT DOESN'T.

Lol noobs
Get 'em all and equip accordingly to counter your opponent.

>>1842156
You're aware that taking the average of a binary outcome can still yield a fraction, right?

>>1842148
Get used to it. This is /sci/

As for weapons/arms, assuming your opponent has the same choices, I'd take green sword, red armor.

Somebody had the right idea earlier, this is a nash equilibrium problem, but it's pretty big to pencil & paper so I'm not solving it.

>>1842261
You should pick red sword/blue armor 0% of the time since it never wins. That's my contribution.

This is actually interesting, I lol'd

>>1842261

only works if everyone makes rational choices. This is an MMO, nobody is rational

>choose green armor

lol ow ow ow ow ow ow

"crappy RNG system..."

>>1842261
has it got anything to do with the guy from a beautiful mind?

Bring the player, not the gear

>>1842358
>implying mmos take skill

Your face when you realize the green sword hits 3 times each second

>>1841954
>>1841976
According to these rules:
Running a simulation with each sword/armor combination 1000 times against each combination, this is the top 10:

Dark Red/Red: 19966 wins
Dark Red/Blue: 19716 wins
Dark Red/Green: 19504 wins
Dark Red/Yellow: 18563 wins
Dark Red/Dark Red: 17698 wins
Blue/Green: 13196 wins
Green/Green: 12796 wins
Red/Green: 12679 wins
Green/Blue: 12176 wins
Blue/Blue: 12172 wins

Now if we were to assume that the shitty combinations are picked less often and are therefore less important, we could multiply each win with the opponent's win ratio, getting the following top 10:

Dark Red/Green: 9344 points
Dark Red/Red: 9293 points
Dark Red/Blue: 9068 points
Dark Red/Yellow: 8210 points
Dark Red/Dark Red: 7887 points
Blue/Green: 6118 points
Green/Green: 6034 points
Red/Green: 5888 points
Green/Red: 5443 points
Yellow/Green: 5418 points

So Dark Red/Green is most likely the best pick, although you won't do much wrong if you pick Dark Red/Red.

>mfw dodge is random making all these charts useless

>>1842394

>he thinks the game utilizes a true random number generator

>he thinks the charts wouldn't still be valid indicators of probability

>>1842394
>mfw I passed 8th grade math and know how averages work

>>1842394
>doesn't know how probability works

>>1835078
yellow armor, green sword.

Yellow armor removes the most dmg.
Green sword causes the most dmg.

>>1842406
How can you apply averages to randomness?
You'd get different averages each time

>>1842430
Ever heard of.. probability funciton? Expected value? Probability theory in general?

Inb4 geuss.

>>1842436
So I can tell what the probability of me getting a certain number when I roll dice?
cool brb I'm going to go play craps

>>1842446
8/10.

>>1842446
>>1842446

Yes you can, you fuck-brain.

Though, environmental effects would all factor in, ranging from size of the dice to the neurological activity within your brain.

>>1842430
>>1842430
I've read a lot of retarded things on /sci/, but this is the mos retarded thing i've heard by far

>>1842454
>>1842452

Damnit, I was just trolled wasn't I?

>>1842446
...yes you can, rather easily

>>1842458
I think so. I fell for it too.

Guys, if I were playing Rock, Paper, Scissors, which one is the best one to pick?

>>1842527
scissors, and FUCK any retards who can't see how obvious it is

>Green armor
>-0 damage
What kind of God would allow for an armor to not protect at all

>>1842552
Think of it as a green t-shirt. Very light, but doesn't exactly protect against swords at all.

>>1835154
HEY. I ALREADY SAID THAT YESTERDAY:

>>1842527

>>1842527
My rock can break scissors and rip through paper.

>>1842552
God 0
Atheists 1

>>1842527

green weapon, most damage per minute
and yellow armor, because if anybody uses the rationale I did for their weapon, it cuts damage to 30 instead of 50 for each blow

>>1842527

It depends how big your hands are.

This is basically how it works, except the green preventing any parrying...

>>1842375
Are you a wizard?

>>1843121
Just a programmer.

>>1836727

I think there's too many Nash Equilibrium to safely determine which is the best sword.

>>1836727

I think there's too many Nash Equilibria to safely determine which is the best combination.

http://codepad.org/YBqF6WVU

Hey guize. Guess what I spent all day hack-and-slashing with what I remember of C from college.

>>1845092
That's the spirit.
Are you sure this is right, though? When comparing it with my results, there are some notable differences.

Blue/Yellow: 10391 wins
Blue/Blue: 10324 wins
Blue/Red: 9704 wins
Blue/Green: 9452 wins
Green/Blue: 8822 wins
Green/Yellow: 8786 wins
Green/Red: 8429 wins
Green/Green: 8368 wins
Yellow/Green: 6977 wins
Red/Green: 6968 wins
Yellow/Blue: 6803 wins
Yellow/Yellow: 6575 wins
Red/Yellow: 6500 wins
Yellow/Red: 6331 wins
Red/Red: 5624 wins
Red/Blue: 5386 wins

Blue/Yellow: 5506 points
Blue/Blue: 5417 points
Blue/Red: 5134 points
Blue/Green: 5033 points
Green/Blue: 4639 points
Green/Yellow: 4619 points
Green/Red: 4462 points
Green/Green: 4455 points
Red/Green: 3758 points
Yellow/Green: 3722 points
Yellow/Blue: 3572 points
Yellow/Yellow: 3433 points
Yellow/Red: 3354 points
Red/Yellow: 3349 points
Red/Red: 3073 points
Red/Blue: 2910 points

The green sword fares especially badly in your simulation, so is it possible that the speed isn't handled correctly? Haven't had a close look at your source yet, but will do so tomorrow if the thread still exists then.

>>1845092
Just curious. How did you calculate these results?

>>1845257

I increment a timer and allow each player one attack every (3600/speed) ticks. I'm not sure how you did it, but if you opted for the damage*speed*opponent's dodge route, we might get different results.

If the HP was lower, the green sword would probably fare significantly better, as the first hit would end it.

Please lemme know if anything strikes you as wrong about the code.

I'll use a green armor and sword because it looks cool, and I'll just attack people I know I can defeat (people with yellow swords/armor above all, I hunt them yellow motherfuckers) and run away from those who can kick my ass.

>>1845312
OK, more in-depth answer:

Each possible pairing of equipment (sword/armor) is sequentially pitted against each other possible pairing of equipment (excepting itself; it's pretty obvious that the first one to attack will probably win in an even matchup). Each match is fought 100 times, 50 times with one set going first and 50 times with the other set going first. Whenever one set wins, one point is added to the final win count for that set.

The actual battle is played out in a turn-based format, where each player gets one attack every (speed/3600) turns. Both players start with 1000 HP, and as soon as one runs out, the other set is declared the victor.

>>1845092

This would suggest that, against a random opponent, red/red or yellow/red would perform best, and green/anything or blue/anything would perform worst. This is only against a random opponent, however; as has been noted, the better stats equipment has, the more players will choose it. (This also gives more weight to combos that trump specific combos all the time; if Y/G kills against R/B, for example.)

Blue.

Always blue.

Blue blue blue.

Damage: 280
Speed: 300

Id probably start out with max damage as a rule of thumb, however I will adjust according to gameplay. The most important stats are extremely different depending on gameplay. Ive played plenty of games where lots of health was always the most important, or lots of damage, or lots of speed, it jsut depends.

I adapt.

Here's the kicker, for the original problem:

-/sci/ SWORD / ARMOR CHOICES-
Green/Green - 6
Blue/Yellow - 6
Green/Yellow - 5
Green/Blue - 4
Yellow/Green - 3
Blue/Green - 3
Green/Red - 2
Red/Blue - 2
Yellow/Blue - 1
Red/Green - 1
Red/Yellow - 1
Red/Red - 1
Blue/Blue - 1

According to the matchup sheet from >>1836478 , if someone had chosen yellow/yellow, they would have beaten 27 of the players.
Next best would have been blue/yellow or green/yellow, with 22 players.

Those who picked green/blue would have won against 17, while green/green would have won against only 12.

Considering their performance against other random enemies, the best choice was actually green/yellow.

Which, if you think about it, it basically a tank with maximum DPS.

gg

Who does best against a random enemy is easy. The hard part is which do you actually pick, if we were all going to play.

Perhaps I would use something like this >>1848064 as the weighting of enemies I would expect to actually come up against, and then I would get a completely different list of who does best against those enemies.

But then I should figure others will do the same so I should use THAT list for who I expect to go up against? And so on and so forth. Pic related.

So what do?

It's just a matter of countering the actual best setup.
Many games have worked like that for long time. Best example coming to mind is magic the gathering: while it allows for a far greater choice, in the end someone will set the standard for "best deck ever" and from then on, anyone else will try to counter that until you complete the circle with a deck that can be countered by the initial one.

>>1848064
Blue/Yellow is good, however Green/Green does not compare and shouldn't be listed at the top.

The chances of Green/Green winning against each opponent are:
Blue/Blue: 52%
Blue/Red: 75%
Blue/Yellow: 4%
Blue/Green: 93%
Red/Blue: 62%
Red/Red: 81%
Red/Yellow: 3%
Red/Green: 97%
Yellow/Blue: 88%
Yellow/Red: 95%
Yellow/Yellow: 27%
Yellow/Green: 100%
Green/Blue: 1%
Green/Red: 12%
Green/Yellow: 0%
Green/Green: 50%

If you compare this with Blue/Yellow:
Blue/Blue: 47%
Blue/Red: 35%
Blue/Yellow: 50%
Blue/Green: 19%
Red/Blue: 96%
Red/Red: 86%
Red/Yellow: 100%
Red/Green: 64%
Yellow/Blue: 44%
Yellow/Red: 32%
Yellow/Yellow: 100%
Yellow/Green: 23%
Green/Blue: 100%
Green/Red: 99%
Green/Yellow: 100%
Green/Green: 97%

then you can see it's better. Blue/Yellow wins in about 68% of all cases and Green/Green just in 52%.

>>1849639
To clarify: a battle is assumed to last exactly one minute and the winner is the fighter with the highest HP left.

And here's a chart that shows the chances.

While I'm at it...

>>1849639

...those are the results.

From this thread.

Of what people chose.

>>1850110
Ah, I see now.

You can ignore the red sword, since it is dominated by the blue sword. The stats that matter are raw DPS and damage per packet, since (DPP - armor reduction) / DPP is what determines percentage damage reduction from the -20 damage stat (for example).