d2jsp
Log InRegister
d2jsp Forums > Diablo II > Diablo 2 Discussion > Strategy & Guides > The Ultimate Soj Finder > A Solution To The Ladder Dclone Rarity
Prev13456715Next
Add Reply New Topic New Poll
Member
Posts: 1,811
Joined: Jun 27 2017
Gold: 0.00
Jul 14 2017 09:47am
Quote (Titus1986 @ Jul 13 2017 11:16pm)
This is true.

But, as good as this can sound you are reeeeeally overestimating the chances to get 3 unique rings in the same run even if you kill every single monster in all lvl39-52 areas in the game in the same run (at least all the nm areas, since you can't go to normal after that in the same game).

Just for fun, yesterday I kill all the monster in act1 in a full game, including cows, all dungeons, level of towers, andariel and anything you can think, it was in hell (more minions = more monster) and with 526% mf, is less than 688, but with that much mf, the difference between 526 and 688 is reaaally small.

This are more monster than if you kill all the lvl39-52 monsters in the game in the same runs, and in a full game.

Loooot of rings, a loooot of drops, unique rings = 0, ofc this is not going to be always that way, you are going to get 1 time to time, sometimes 2, and even sometimes 3, but 3 is not the rule here, is something rare, even if you go with 1000+mf and a horking barb after you.

And if you fail to get a soj in a 1h run is a wasted hour for nothing.

The "100% soj finding every single game" is the biggest bullshit in all the threat and have plenty already.


Yes, the OP's math is off. Individual chances aren't additive, rather the chance of one not dropping is multiplied.

The concept of limiting target areas to A1nm where you've got a decent shot of a drop is fine. I think I'd be inclined to just modify andy runs to start from inner cloister wp and target all bosspacks along the route. A nova sorc should be faster than an auradin, and cheaper.
Member
Posts: 8,685
Joined: Jun 1 2014
Gold: 2,765.00
Jul 14 2017 01:06pm
Quote (Khboer @ Jul 14 2017 11:47am)
Yes, the OP's math is off. Individual chances aren't additive, rather the chance of one not dropping is multiplied.

The concept of limiting target areas to A1nm where you've got a decent shot of a drop is fine. I think I'd be inclined to just modify andy runs to start from inner cloister wp and target all bosspacks along the route. A nova sorc should be faster than an auradin, and cheaper.



Nah the maths are good about the chance to id a soj
Will always be 1/31 if 1 unique is id'd
Then 1/16 on the 2nd if manald/nangel was id'd first (in both case theyre 15/31 so that fraction is just removed)
And obviously 1/1 if no soj on the first 2 id's

About the nova sorc, no matter the build, you spend your time casting and teleporting. Even a 200fcr nova sorc wont do the job quicker (same killing speed but then while the auradin already finished picking items and moved on, the sorc starts picking)
I tried many options like the nova sorc/200fcr barb with item find/javzon/orbsorc on hero editor
After many tests, the auradin was the very best option for maximized mf% vs killing speed

And its not that expensive you know.. the anni is really not mendatory (only used for the +skill anyway)
You should be able to gather the whole perfect setup described in the 1st post for 500-800 fgs
Any defensive stats is 100% useless (monsters die before they see you) so dont spend money about these stats. 10/10 ptorch is perfect here.
And the 7frw mod on the skillers is just a very little upgrade for the few moments you run (town/picking)

This post was edited by SinsOfTheSun on Jul 14 2017 01:24pm
Member
Posts: 1,811
Joined: Jun 27 2017
Gold: 0.00
Jul 14 2017 01:52pm
Quote (SinsOfTheSun @ Jul 11 2017 01:58pm)
Yep, I used to believe this
but trust me, killing nm glitched andy every 2 min vs killing thousands of monsters per min, youll notice a very huge difference
glitched nm andy has 10% chance of droping a ring
every single monster you kill has 0.6% chance in those area, with a better chance for soj (cuz lvl39 instead of lvl60)
so basically, every 16.6 regular monster killed = 1 glitched andy quest (with 1/31 chance instead of 1/52)


Here's where you screw up your math. Killing NM q andy has a 0.00103 chance of dropping . That leaves another monster with a 0.000061 chance of SoJ.

Chance to drop from x kills = 1-(1-chance to drop)^(monsters killed)

Let's go with 1000 andy kill.

1-(1-0.00103^1k=.64.

64% chance per 1000 QAndy kills.

Now, 16600 monster kills

1-(1-0.000061)^16600=.63

It's a little lower, but not by much. Still, there are most certainly not that many monsters in Cata to even get a 50% chance in a game.

"But if three unique rings drop below lvl51 one will be an SoJ"

Alright. So we adjust the math. We multiply every above chance by 31 to get the chance to find a unique ring. We'll exclude andy.

0.000061x31=0.00019. Let's go with the above number of 16600 again.

1-(1-0.00019)^16600=.96

You'd have a 96% chance of getting ONE unique ring in 16600 monster kills. The chances of getting three in 49800 is good. But there are not enough monsters at the specific level to guarantee a 100% shot at SoJ in a game. So your claim is wrong.

So, could an auradin with tele get more SoJs per hour than a char only killing q Andy? Sure seems like it. But I'd leave claims about 1 per game off the table.
Member
Posts: 8,685
Joined: Jun 1 2014
Gold: 2,765.00
Jul 14 2017 02:18pm
Quote (Khboer @ Jul 14 2017 03:52pm)
Here's where you screw up your math. Killing NM q andy has a 0.00103 chance of dropping . That leaves another monster with a 0.000061 chance of SoJ.

Chance to drop from x kills = 1-(1-chance to drop)^(monsters killed)

Let's go with 1000 andy kill.

1-(1-0.00103^1k=.64.

64% chance per 1000 QAndy kills.

Now, 16600 monster kills

1-(1-0.000061)^16600=.63

It's a little lower, but not by much. Still, there are most certainly not that many monsters in Cata to even get a 50% chance in a game.

"But if three unique rings drop below lvl51 one will be an SoJ"

Alright. So we adjust the math. We multiply every above chance by 31 to get the chance to find a unique ring. We'll exclude andy.

0.000061x31=0.00019. Let's go with the above number of 16600 again.

1-(1-0.00019)^16600=.96

You'd have a 96% chance of getting ONE unique ring in 16600 monster kills. The chances of getting three in 49800 is good. But there are not enough monsters at the specific level to guarantee a 100% shot at SoJ in a game. So your claim is wrong.

So, could an auradin with tele get more SoJs per hour than a char only killing q Andy? Sure seems like it. But I'd leave claims about 1 per game off the table.



Hey man, these numbers are 100% right! Only one thing tho...
thats with 0% mf, and thats also concidering only normal monsters, while a lot of them are unique/champions, and also all ghost (tons in all those areas farmed) have very higher chance to drop rings. And you didnt count andy, while I obviously kill her too.
Yeah my claim to drop a soj all game was just out of happiness after I completed my first full clear and got a soj
tho now its been 15 runs, and the average unique rings dropped is around 2 per game
On 15 runs I got 7 soj so far
3 runs with 3 unique rings : 3 soj, 3 manald, 3 nangel
8 runs with 2 unique rings : 3 soj, 6 manald, 7 nangel
4 runs with 1 unique ring : 1 soj, 2 manald, 1 nangel

15 runs was done over 14 hours
7 soj / 14 hours = 0.5 soj per hour
* 24h = 12 soj
which is corresponding with the numbers I was mentioning before (yes my math was wrong at first, and I acknowledged and corrected that long time ago, but my numbers about the soj drop rate have always been true and they're very stable every day since I started running this)

This post was edited by SinsOfTheSun on Jul 14 2017 02:19pm
Member
Posts: 1,811
Joined: Jun 27 2017
Gold: 0.00
Jul 14 2017 02:18pm
The odds of Andy dropping an SoJ are with 200% mf

This post was edited by Khboer on Jul 14 2017 02:20pm
Member
Posts: 1,811
Joined: Jun 27 2017
Gold: 0.00
Jul 14 2017 02:23pm
I'd also say that over 15 runs you're beating the odds by a long shot. I'd expect that to come down
Member
Posts: 8,685
Joined: Jun 1 2014
Gold: 2,765.00
Jul 14 2017 02:25pm
Quote (Khboer @ Jul 14 2017 04:18pm)
The odds of Andy dropping an SoJ are with 200% mf


Yeah thats even worse when you compare andy Q with 200mf VS only regular monsters with 0mf
like I said, all unique/champions (and the monsters around them) + also all ghost type have 1.6% chance instead of the 0.3% you calculated.

Hey whatever dude, use the numbers you want. I'm just trying to expose what is happening about my ridiculous soj drop rate since a few days.
If you don't like it, just go back to regular mfing and let me explore new ways of having fun and making profit :p
Member
Posts: 1,811
Joined: Jun 27 2017
Gold: 0.00
Jul 14 2017 02:32pm
Quote (SinsOfTheSun @ Jul 14 2017 12:25pm)
Yeah thats even worse when you compare andy Q with 200mf VS only regular monsters with 0mf
like I said, all unique/champions (and the monsters around them) + also all ghost type have 1.6% chance instead of the 0.3% you calculated.

Hey whatever dude, use the numbers you want. I'm just trying to expose what is happening about my ridiculous soj drop rate since a few days.
If you don't like it, just go back to regular mfing and let me explore new ways of having fun and making profit :p


Come back with proof after 500 runs and we can chat more. For now, based on the numbers you give vs the odds, you should be playing the lottery, not D2
Member
Posts: 5,643
Joined: Apr 10 2012
Gold: 35,300.00
Jul 14 2017 04:41pm
Quote (Khboer @ 14 Jul 2017 21:32)
Come back with proof after 500 runs and we can chat more. For now, based on the numbers you give vs the odds, you should be playing the lottery, not D2


ill actually give this a try in a few days as soon as im back from my holidays and let you know my results if you want... im curious if it really works that well.. it does also depend on the amount of packs/ghosts, im not sure if you can really calculate that. well see how it goes, wordt that can happen is i spent some fgs for stuff i dont need :P
Member
Posts: 8,685
Joined: Jun 1 2014
Gold: 2,765.00
Jul 14 2017 05:59pm
Quote (nobosy @ Jul 14 2017 06:41pm)
ill actually give this a try in a few days as soon as im back from my holidays and let you know my results if you want... im curious if it really works that well.. it does also depend on the amount of packs/ghosts, im not sure if you can really calculate that. well see how it goes, wordt that can happen is i spent some fgs for stuff i dont need :P



Do it
If you're not convinced I'll bin your setup vs 1k fgs for your time
I need to convert my hdins to this
Go Back To Strategy & Guides Topic List
Prev13456715Next
Add Reply New Topic New Poll