ATMA Drop Calc for v1.11

Aerwynd

Diabloii.Net Member
ATMA Drop Calc for v1.11

This is really weird...

I'm creating some Quick Reference Guides for popular MF Targets (Meph, Pindle, etc) and I noticed something weird...

*NOTICE* All numbers come from ATMA Drop Calculator for v1.11...

For Pindle, all numbers are /p1, because according to the Drop Calc, /pX settings doesn't matter with Pindle.

Pindle's chance for Isenhart's Case (MF - 0%, 250%, 500%, 750%, 1000%):
0% ---- 1:1410
250% -- 1:539
500% -- 1:501
750% -- 1:502
1000% - 1:503

Does that make sense? The higher your MF% the LESS it will drop?

With Pindle, it's like that with EVERY Set item...

Let's try Uniques...

Pindle's chance for Venom Ward (MF - 0%, 250%, 500%, 750%, 1000%):
0% ---- 1:3663
250% -- 1:1628
500% -- 1:1377
750% -- 1:1276
1000% - 1:1221

Hmm, that makes more sense... Slowly increases your chance, which is the way MF is SUPPOSE to work...

Let's try Mephy... Mephy will go by /pX Settings AND MF%...

Meph's chance for Isenhart's Case (/p1 - 0%, 250%, 500%, 750%)
0% ---- 1:120
250% -- 1:50
500% -- 1:48
750% -- 1:49

Hmmm, same thing... Are the chances for set items broke? Or is it some sort of /p1 thing as the pattern doesn't fit with /p3 and higher...

Meph's chance for Isenhart's Case (/p3 - 0%, 250%, 500%, 750%)
0% ---- 1:101
250% -- 1:42
500% -- 1:41
750% -- 1:41

Meph's chance for Isenhart's Case (/p5 and Higher - 0%, 250%, 500%, 750%)
0% ---- 1:98
250% -- 1:41
500% -- 1:40
750% -- 1:39

Meph's chance for Venom Ward (/p1 - 0%, 250%, 500%, 750%)
0% ---- 1:292
250% -- 1:130
500% -- 1:110
750% -- 1:102

Meph's chance for Venom Ward (/p3 - 0%, 250%, 500%, 750%)
0% ---- 1:245
250% -- 1:109
500% -- 1:92
750% -- 1:85

Meph's chance for Venom Ward (/p5 and Higher - 0%, 250%, 500%, 750%)
0% ---- 1:228
250% -- 1:106
500% -- 1:89
750% -- 1:83

With the above numbers, for Pindle, the best MF (For those Higher TC SET items only) is ~500%. While the MF for Uniques should be as high as possible... It depends on what you are going for I suppose...

As for Meph, with the numbers above, optimal /pX setting is /p3... Less HP/XP but slightly more runs (seems minimal to me) WHICH ARE FASTER... I think it's worth the cut in /pX setting to get a better chance of getting Uniques from him...

Input is always welcome :thumbsup:

If someone wants to test these numbers, please do, and post them here when you are finished.

Aerwynd
 

scwormy

Diabloii.Net Member
This is really weird...

I'm creating some Quick Reference Guides for popular MF Targets (Meph, Pindle, etc) and I noticed something weird...

*NOTICE* All numbers come from ATMA Drop Calculator for v1.11...

For Pindle, all numbers are /p1, because according to the Drop Calc, /pX settings doesn't matter with Pindle.

Pindle's chance for Isenhart's Case (MF - 0%, 250%, 500%, 750%, 1000%):
0% ---- 1:1410
250% -- 1:539
500% -- 1:501
750% -- 1:502
1000% - 1:503

Does that make sense? The higher your MF% the LESS it will drop?

With Pindle, it's like that with EVERY Set item...

Let's try Uniques...

Pindle's chance for Venom Ward (MF - 0%, 250%, 500%, 750%, 1000%):
0% ---- 1:3663
250% -- 1:1628
500% -- 1:1377
750% -- 1:1276
1000% - 1:1221

...snip...

Aerwynd
Isn't that saying with 0 MF, it will drop once in 3663 times
with 1000% it will drop every 1221 times...therefore the more MF, the more likely it is to drop.

And as Bongo indicated, the more MF, the more likely to get a Unique rather than a Set.



 

Hp_Sauce

Diabloii.Net Member
Isn't that saying with 0 MF, it will drop once in 3663 times
with 1000% it will drop every 1221 times...therefore the more MF, the more likely it is to drop.
Thats how I always read the ATMA numbers too.

Using the Pindle example: 1000% Mf has 1 in 500 odds which is basically saying you should find 3 Isenhart's cases in almost the same number of runs it would take you to find 1 Isenhart's case with 0% Mf.

-hps
 

Aerwynd

Diabloii.Net Member
@scwormy --> That's correct... The more MF you have the more likely it is you will find it...

It shows correct for Uniques, but for sets it's different at /p1... You shouldn't get LESS of a chance to find an item just because you have MORE MF... That doesn't make sense...

Yeah, Bongo's right... You have MORE of a chance for a green item to be gold, but that shouldn't LOWER your chances of getting that GREEN item...

The chart that shows chance for Uniques/Sets with certain MF% would be wrong then... This chart (Taken from Hrus' MF Guide)...

Code:
MF     Unique  Set     Rare    Magic    Real MF multiplikator for Uniques
100     71      83      85     100      1.71
200    111     142     150     200      2.11
300    136     187     200     300      2.36
400    153     222     240     400      2.53   
500    166     250     272     500      2.66
600    176     [B][U]272[/U][/B]     300     600      2.76
700    184     [B][U]291[/U] [/B]    323     700      2.84
800    190     [B][U]307[/U][/B]     342     800      2.90
The Bolded and Underilined text in the code would be wrong if you have LESS of a chance to get Set items... Correct?

Someone set me straight if I am wrong...

Aerwynd
 

Are

Diabloii.Net Member
Since the game checks for unique first, and then for set, you will eventually get to a point where the chance for set remains the same or becomes lower, as your chance to get the item as a unique rather than set becomes higher than the chance to get set rather than rare.

Or something like that.
 

Aerwynd

Diabloii.Net Member
THX CAT :thumbsup: It still doesn't make sense that they would code it that way...

That's like saying if I want an IKSC, It would be better to lower my MF to the cap, thus DECREASING the chance for a very rare Elite Unique...

Bah, at least we have SOME sort of answer now though...

Thx again Cat...

Aerwynd
 

Sir Lister of Smeg

Diabloii.Net Member
What I did not get from Thrugg's post (and that whole thread for that matter), is what that drop cap is. I know there is a drop cap, that follows from the way effective magic find is calculated. It is however, far higher than what is reached at 600% mf.
Effective magic find is given by trunc(a*mf/(a+mf)), which goes to a if mf goes to infinity (a=250 for uniques, 500 for sets, and 600 for rares). This means there is a hard cap in the form of a, but even at 1200% magic find (higher than the maximum reachable, 1167%), the effective magic find is still far lower than that cap. Additionally, the formula above does increase between 600 and 1200, so again what is the hard cap mentioned by Thrugg?
 

jjscud

Diabloii.Net Member
Its a simple math problem, at least when you take away the numbers :grin: .

Lets take Pindle, he drops some item with one of his two item drops. Because he's a SU, that item is guaranteed to be Magical, Rare, Set or Unique. No argument, good.

Now, increasing MF increases the chance of getting a unique, good. But it also increases the chance of getting a Set, Rare and Magic. Now we have a problem. They simply cannot all increase, its impossible.

So, what happens, well, Uniques gets checked first so they never get robbed but everything else does. Every step along the way, the chance of finding a Set item is getting cheated. Now, at low MF rates, the Sets chances are increasing so much faster than the Uniques that the little bit it loses isn't noticed as its still gaining healthily. Then around 600% MF the numbers work out so that Set items are getting cheated more than they can make up.


Perhaps here's an easier way to look at it. Since uniques get rolled first, if you roll a unique you will never perform the Set test. So a 1/20 chance of getting a unique is also a 1/20 chance of not checking for a Set, regardless of how likely a Set item is.

Considering this, the table that Aerwynd quoted makes perfect sense. The chance of rolling a Set on the Set test is always increasing BUT the chance of skipping the set test (rolling a unique) is also always increasing.


Anyways, I'm just glad Doc has given us easy access to the real numbers so I can talk math while waving my arms in the air and not getting any chalk on my hands.
 

Aerwynd

Diabloii.Net Member
Good analysis jjscud :thumbsup:

That makes more sense to me... A little anyway :grin: I just don't agree with the numbers ATMA is using :laugh:
 
Top