I'm back!!

I was on a week-long cruise.

Long story short, I didn't not enjoy it as much as staying at home.

Anyways... onto business of this thread!

[SPOILER=What's going on in MapleStory?]I can get that there's Mu Lung Dojo and Monster Carnival 2 and the rest of the works that come with them.

Then I read up on weekly spirit events from Cassandra or something, and the "let's spell crap" event. I even saw the reward lists and the letter drop lists.

There's one thing that I'm perplexed about: 100% GM scrolls. Where are they coming from? o.O Also, am I missing anything else? =D

 Calling all math geeks!

So I tried figuring out on average how many gems/crystals it would take to create an advanced grade. I'm aiming for a general formula, not necessarily a correct model of MapleStory's Maker systems (although that would be nice). I need help verifying that my math is correct.

I started with three equations:

Code:

100C = 75B + 24I + A
I = 10B + (10/7)B = (80/7)B
A = 10I + (10/3)I = (40/3)I

Where C are ungraded gems/crystals, B are basic crystals, I are intermediates, and A being advanced.
The first equation describes the probabilities of getting B, I, and A from C.
The second equation describes the average amount of B needed to make I. This is assuming a 70% success rate for convertion of 10B into I.
The third equation describes the average amount of I needed to make A. 30% success rate this time.

The basic goal is to crunch the first equation into the form x*C = A, which should tell us how many crystals/gems, on average, it takes to get an advanced grade.

I decided to work this two ways: Determining the answer assuming you can't appraise C into A directly (this changes equation 1 to 99C = 75B + 24I), and determining the answer the regular way.

Method 1:

Code:

99C = 75B + 24I
99C = (105/16)I + 24I
99C = (489/16)I
99C = (1467/640)A
[color=Red]43.190184049079754601226993865031[/color]C = A

Method 2:

Code:

100C = 75B + 24I + A
100C = (105/16)I + 24I + A
100C = (489/16)I + A
100C = (1467/640)A + A
100C = (2107/640)A
[color=Red]30.374940673943996203132415757[/color]C = A

As you can see, the results are radically different, and it demonstrates that a single percent in being able to gain an advanced crystal/gem directly from the initial appraisal affects the averages greatly.

Is the logic sound?

You were gone?

Also, your results make sense because 1/43.19... ~= 0.023 and 1/30.37... ~= 0.033, so it went up about 1%. Which also is the initial appraisal rate for a high grade.

And the new 100% scrolls apparently come from Gach.

You were gone?

What, should I spam threads that were a week old to prove my point? >_>

Just search up my posts, there's a week gap you know.

Anyways just verify my maff plz

welcome back kaj, i just fixed my old comp... itz running like a beast.
Too bad im getting a new one next week haha. Oh kaj tell me the exp rate per hour at the new maps.

Oh and i finished dl'ing MS.
Im gona make a pg mule at first, then a koc and leech it to 120, then lvl hero and ks lolpally<333

Those 100% GM Scrolls come from Mushie Shrine Gach, as do the 2 new chairs. Welcome back.

I think a few months ago Stereo calculated that out of 1000 crystals, it takes 28 unrefined crystals and 5m* to make 1 advanced crystal. Lemme try to find it. Here it is.

Also, your results make sense because 1/43.19... ~= 0.023 and 1/30.37... ~= 0.033, so it went up about 1%. Which also is the initial appraisal rate for a high grade.

What about the other stuff before the C = A + A part?

I think a few months ago Stereo calculated that out of 1000 crystals, it takes 28 unrefined crystals and 5m* to make 1 advanced crystal. Lemme try to find it. Here it is.

I'd rather see a sample size of 1 billion than take his word for it. 1000 is a fking joke for a sample size if we're staring at 1% chances or something.

What about the other stuff before the C = A + A part?



I'd rather see a sample size of 1 billion than take his word for it. 1000 is a fking joke for a sample size if we're staring at 1% chances or something.

Well you calculated it to take 30 crystals, and he calculated it to take 28, I'm guessing that he's pretty close to the exact ratio. Besides, nobody's going to be trying it with over 100 anyways.

Well you calculated it to take 30 crystals, and he calculated it to take 28, I'm guessing that he's pretty close to the exact ratio. Besides, nobody's going to be trying it with over 100 anyways.

No, but the difference is Stereo kept running a simulator until all that was left were Advanced Crystals, while I took an entirely theoretical approach via mathematics.

Actually, I'm pretty sure Stereo did use math, he just started with a fixed number of crystals.

"740 low grade + 250 mid grade + 10 high grade"
Doesn't look like a random simulation to me. That was systematically split.

What about the other stuff before the C = A + A part?

What other stuff? your equations?
-rederives-

Code:

0 = 0.7*(I - 10B) + 0.3*(-B)
0 = 0.7I - 7B - 0.3B
7.3B = 0.7I
I = (7.3/0.7)B = (73/7)B

Hm, it is a little off. Your number (80/7) supposes you lose 11 basic crystals on success, not 10.

By analogy, the third equation should be:

Code:

A = (3.7/0.3)I = (37/3)I

So your actual results:

Code:

99C = 75B + 24I
99C = (525/73)I + 24I
99C = (2277/73)I
99C = (6831/2701)A
[color=red]39.1449275[/color]C = A

Code:

100C = 75B + 24I + A
100C = (525/73)I + 24I + A
100C = (2277/73)I + A
100C = (6831/2701)A + A
100C = (9532/2701)A
[color=red]28.3361309[/color]C = A

Code:

0 = 0.7*(I - 10B) + 0.3*(-B)
0 = 0.7I - 7B - 0.3B
7.3B = 0.7I
I = (7.3/0.7)B = (73/7)B

Hm, it is a little off. Your number (80/7) supposes you lose 11 basic crystals on success, not 10.

By analogy, the third equation should be:

Code:

A = (3.7/0.3)I = (37/3)I

So your actual results:

Code:

99C = 75B + 24I
99C = (525/73)I + 24I
99C = (2277/73)I
99C = (6831/2701)A
[color=red]39.1449275[/color]C = A

(80/7) should be 10 + 1/.7, which should be the cost of success and the average cost of failure. I'm honestly not sure how to represent the cost of success at this point. All I have to say was that it (the 1/.3 part that applies) looked good for converting I into A.

Well, you also have to consider the average cost of success, since it only happens 70% of the time.

0 = 0.7*(I - 10B) + 0.3*(-B)
0 [assumption: no value gained or lost by appraising] = 0.7*(I - 10B) [0.7 chance of gaining I and losing 10B] + 0.3*(-B) [0.3 chance of losing B]

What, should I spam threads that were a week old to prove my point? >_>

Just search up my posts, there's a week gap you know.

Anyways just verify my maff plz

[COLOR="Red"]Wut kind of Southperrian are you? I check SP on vacations.

Oh well...welcome back Kaj. WillDaSnail quit. [/COLOR]

WillDaSnail quit.

NOOOOOOOOOOOOOOOOOOOOOOO!

NOOOOOOOOOOOOOOOOOOOOOOO!

Well, you also have to consider the average cost of success, since it only happens 70% of the time.

0 = 0.7*(I - 10B) + 0.3*(-B)
0 [assumption: no value gained or lost by appraising] = 0.7*(I - 10B) [0.7 chance of gaining I and losing 10B] + 0.3*(-B) [0.3 chance of losing B]

Yes, that makes more sense xD

10.428571428571428571428571428571B = I
12.333333333333333333333333333333I = A

Method 1
39.144927536231884057971014492743C = A

Method 2
28.336130927402433906840117498945C = A

Thanks Russt!

Oh well...welcome back Kaj. WillDaSnail quit.

Cruise ship has shytty internets and it costs 50-100 cents a minute to use >_> I'd rather hek their security than pay such outrageous prices.

And yes, I know Will quit. He needs to get his life back on track.