Its official lads! The WB randomizer works! Withered Jim second week in a row :D

[Divinexve]

wish it was him forever, love me dem arcano crystals

[Kuntantee]

It's a general misconception that pseudrandom algorithm cannot be used to create real random numbers.
What is true is that with the seed and the current randomizer (usually CPU cycles or timestamp) any value can be (p)(re-)calculated.

But what people often forget is that the randomizer is truly random, as in a multi-threaded multi-user environment with semaphore-tokened registers it is unpredictable and therefor indeed random which randomizer is used to create the random number, it's not like the randomizer is a consecutive row of predefined numbers.

Collecting entropy from your disk, or some other input device is sufficiently random.

[Vaerys]

You cannot predict what comes next. And there's never a situation with only one boss in the pool. We are not talking about bingo here. The system should prevent repeats (in a row) not remove bosses from the pool.

Discussions of randomness always come down to people talking past each other with different ideas about what randomness means. This one is no exception. The scenario in which options are progressively eliminated is LESS random than one in which all options are valid in each selection. "You cannot predict what comes next" would be technically true for a scenario in which the number of possible options had been narrowed down from several million to only two. However, if the final result turned out to be one of these two options time after time, when the expectation is that the result was selected randomly from a large pool, then we would be warranted in criticizing the randomness of the process.

[The One Percent]

Well wither me timbers, look at all this QQ

[Accendor]

It is random if you cannot predict who comes next. You can disagree all you want.

But you can not predict it.

[Vladinator]

Thanks to JolietJerry for the numbers. You can find his comment here on Wowhead.

His log basically results in the following table:

(1) 2016-09-20 Nithogg
(2) 2016-09-27 Levantus
(3) 2016-10-04 Withered J'im
(4) 2016-10-11 Withered J'im
(5) 2016-10-18 Calamir
(6) 2016-10-25 Drugon the Frostblood
(7) 2016-11-01 Flotsam
(8) 2016-11-08 Shar'thos
(9) 2016-11-15 Humongris
(10) 2016-11-22 The Soultakers
(11) 2016-11-29 Nithogg
(12) 2016-12-06 Levantus
(13) 2016-12-06 Na'zak the Fiend
(14) 2016-12-13 Withered J'im
(15) 2016-12-20 Calamir
(16) 2016-12-27 Drugon the Frostblood
(17) 2017-01-03 Flotsam
(18) 2017-01-10 Shar'thos
(19) 2017-01-17 Ana-Mouz
(20) 2017-01-17 Humongris
(21) 2017-01-24 The Soultakers
(22) 2017-03-28 Drugon the Frostblood
(23) 2017-04-04 The Soultakers
(24) 2017-04-11 Shar'thos
(25) 2017-04-18 Withered J'im
(26) 2017-04-25 Shar'thos
(27) 2017-05-02 Drugon the Frostblood
(28) 2017-05-09 Ana-Mouz
(29) 2017-05-16 Flotsam
(30) 2017-05-23 Levantus
(31) 2017-05-30 Withered J'im
(32) 2017-06-06 Calamir
(33) 2017-06-13 Ana-Mouz
(34) 2017-06-20 Levantus
(35) 2017-06-27 Withered J'im
(36) 2017-07-04 Withered J'im

Resulting in the following graph:


Looks like Withered J'im and Na'zak are outliers. Rest you can expect has a normal distribution for a RNG based system.

Feel free to correct me. I just found the lack of real numbers a issue as people posted based on feelings. Hopefully this helps steer the thread in the right direction.

[Seranthor]

Thanks to JolietJerry for the numbers. You can find his comment here on Wowhead.

His log basically results in the following table:

(1) 2016-09-20 Nithogg
(2) 2016-09-27 Levantus
(3) 2016-10-04 Withered J'im
(4) 2016-10-11 Withered J'im
(5) 2016-10-18 Calamir
(6) 2016-10-25 Drugon the Frostblood
(7) 2016-11-01 Flotsam
(8) 2016-11-08 Shar'thos
(9) 2016-11-15 Humongris
(10) 2016-11-22 The Soultakers
(11) 2016-11-29 Nithogg
(12) 2016-12-06 Levantus
(13) 2016-12-06 Na'zak the Fiend
(14) 2016-12-13 Withered J'im
(15) 2016-12-20 Calamir
(16) 2016-12-27 Drugon the Frostblood
(17) 2017-01-03 Flotsam
(18) 2017-01-10 Shar'thos
(19) 2017-01-17 Ana-Mouz
(20) 2017-01-17 Humongris
(21) 2017-01-24 The Soultakers
(22) 2017-03-28 Drugon the Frostblood
(23) 2017-04-04 The Soultakers
(24) 2017-04-11 Shar'thos
(25) 2017-04-18 Withered J'im
(26) 2017-04-25 Shar'thos
(27) 2017-05-02 Drugon the Frostblood
(28) 2017-05-09 Ana-Mouz
(29) 2017-05-16 Flotsam
(30) 2017-05-23 Levantus
(31) 2017-05-30 Withered J'im
(32) 2017-06-06 Calamir
(33) 2017-06-13 Ana-Mouz
(34) 2017-06-20 Levantus
(35) 2017-06-27 Withered J'im
(36) 2017-07-04 Withered J'im

Resulting in the following graph:


Looks like Withered J'im and Na'zak are outliers. Rest you can expect has a normal distribution for a RNG based system.

Feel free to correct me. I just found the lack of real numbers a issue as people posted based on feelings. Hopefully this helps steer the thread in the right direction.

31-Jan Nithogg (3)
07-Feb Levantus (3)
14-Feb J'im (4)
21-Feb Calamir (3)
28-Feb Drugon (3)
07-Mar Flotsam (3)
14-Mar Shar'thos (3)
21-Mar Humongris (3)

(3) indicates it was the 3rd spawn of that NPC

So the comment on 4 July would be that it has been 30 wks since Na'Zak, 22 wks since Nithogg, J'im has spawned 6x since the last Na'Zak spawn, and J'im has spawned 5x since the last Nithogg spawn.

[Elim Garak]

Looks like Withered J'im and Na'zak are outliers. Rest you can expect has a normal distribution for a RNG based system.

Feel free to correct me. I just found the lack of real numbers a issue as people posted based on feelings. Hopefully this helps steer the thread in the right direction.

There's not enough data to draw any conclusions.

[Vladinator]

There's not enough data to draw any conclusions.

Yeah. I'm looking forward to in another half year to see how it develops.

[Shakou]

There are a few classes, who doesnt really need UA, and you still need it to proc pretty high TF, and the other thing. The game is for everyone, so lets just have WJ spawn every 8-11 weeks ;-)

This is simply not true, a 905 Unstable Arcanocrystal is legitimately only marginally better than the 860 base version. It's just that good of a trinket, far overbudgetted for it's I lvl and besides, secondary stat increases on gear are incremental every 5 I lvls.

However, there are better options available for most classes at this point.

[Svad1287]

This is simply not true, a 905 Unstable Arcanocrystal is legitimately only marginally better than the 860 base version. It's just that good of a trinket, far overbudgetted for it's I lvl and besides, secondary stat increases on gear are incremental every 5 I lvls.

However, there are better options available for most classes at this point.

I wrote the UA into the post, because people mentioned they wanted WJ, to spawn alot, because of TF version etc ;-)