Neo Nyzul Isle (Will FFXI continue? Find out on the next FFXI XP Pack!)

[Tarage]

Did mob difficulty get tweaked along with the Lamps or is this guy in my Ig party just talking out his ass to try and justify charging 15M for NNI wins?

Pocket said to me that it's the lamps that are now fucked. Basically, because you can't datswap them it's that much harder to do those floors.

I don't know though.

[Spicyryan]

Pocket said to me that it's the lamps that are now fucked. Basically, because you can't datswap them it's that much harder to do those floors.

I don't know though.

It is the typical "oh this changed so everything must be different" and "I did not do as well as I thought I would so it must now be harder".

Should make a general paranoid parrot meme for this category.

As far as I saw last night. Same old floor jumps, same old difficulty, and same old Nyzul.

[NynJa]

Yeahhhhhhhhhh, I'm sorry, but an order lamp floor jumping from 90 seconds to a potential 5+ minutes is not "paranoid parrot meme"

[Spicyryan]

Yeahhhhhhhhhh, I'm sorry, but an order lamp floor jumping from 90 seconds to a potential 5+ minutes is not "paranoid parrot meme"

Unless I am losing what little bit of my mind I have left. I was more going on about people saying mobs and floor jumps were changed to be more difficult.

[pchan]

It is the typical "oh this changed so everything must be different" and "I did not do as well as I thought I would so it must now be harder".

Should make a general paranoid parrot meme for this category.

As far as I saw last night. Same old floor jumps, same old difficulty, and same old Nyzul.

The fact that you are insisting entirely proves the opposite. Lamps are necessarily going to fuck you unless you can still guess 1/1. Also nice job mentionning you can win without running through walls, we totally believe you (why even mention it ?). I've always wondered why the entire BG pseudo elitist cheaters try their hardest to justify the event, implying that they win 1/3 of the time ("without cheats"). Very funny.

[NynJa]

Unless I am losing what little bit of my mind I have left. I was more going on about people saying mobs and floor jumps were changed to be more difficult.

His entire post was about order lamps. Mob difficulty hasnt changed, floor jumps are still completely random, just order lamps.

[Jem]

The fact that you are insisting entirely proves the opposite. Lamps are necessarily going to fuck you unless you can still guess 1/1. Also nice job mentionning you can win without running through walls, we totally believe you (why even mention it ?). I've always wondered why the entire BG pseudo elitist cheaters try their hardest to justify the event, implying that they win 1/3 of the time ("without cheats"). Very funny.

I don't see what the issue is with people saying they go 1/3 legit. Seems completely plausible to me and I'd certainly buy that because if I went and grabbed 5 friends and had a worse win rate than 1/3-4 I'd be furious (unless we got fucked every run by Order floors) so 1/3 doesn't seem farfetched.

Order floors are going to happen but not every run (and sometimes they won't be terrible). Plus quite a few 100 wins could accommodate for 1 order floor anyway.

[pchan]

you will average like 2+ lamp floors per run, so the chance to get one is very high, can't do the math now, but since you need about 15-20 floors to win, everyone can guess. Also 3-lamps floor order are not the same as 5-lamps order.

[TSFFXI]

Assume: 20% chance to land on a lamp floor (I don't want to find Kirschy's post with the exact percentage!)
Assume: All three lamp types have an equal probability of occurring

(0.2)*16=3.2 , so on a 16 floor run we should expect, on average, 3 lamp floors. On average, one should be code.

p(order)=0.066
p(no order)=1-0.066=0.934
p(no order after 16 floors)= 0.934^16 = 0.335389930618
p(no order after 17 floors)= 0.934^17 = 0.313254195
p(no order after 18 floors)= 0.934^18 = 0.292579418
p(no order after 19 floors)= 0.934^19 = 0.273269177
p(no order after 20 floors)= 0.934^20 = 0.255233411

Pretty sure it doesn't need to be any more complicated than that. Let me know if I'm wrong!

[Jem]

you will average like 2+ lamp floors per run, so the chance to get one is very high, can't do the math now, but since you need about 15-20 floors to win, everyone can guess. Also 3-lamps floor order are not the same as 5-lamps order.

Number of Order Lamps in 25 runs

0, 1, 1, 1, 0, 1, 1, 1, 2, 0, 0, 2, 0, 1, 0, 1, 1, 2, 0, 1, 0, 1, 1, 3, 0

Average clears: 19.8 (across those 25 runs, was worse prior to that but a different set-up)
Number of times an extra 3 mins per Order floor would have stopped us clearing: 3 (1 is marginal as I have it down as exactly 3 mins remaining, counting as a fail.)

[Yugl]

Assume: 20% chance to land on a lamp floor (I don't want to find Kirschy's post with the exact percentage!)
Assume: All three lamp types have an equal probability of occurring
p(order)=0.066
p(no order)=1-0.066=0.934
p(no order after 16 floors)= 0.934^16 = 0.335389930618
p(no order after 17 floors)= 0.934^17 = 0.313254195
p(no order after 18 floors)= 0.934^18 = 0.292579418
p(no order after 19 floors)= 0.934^19 = 0.273269177
p(no order after 20 floors)= 0.934^20 = 0.255233411
p(two order floors)= 0.066^2 = 0.004356
p(you don't get two order floors after 2 floors) = 1-0.004356=0.995644
p(you don't get two order floors after 16 floors) = 0.932535331
p(you don't get two order floors after 17 floors) = 0.928473207
p(you don't get two order floors after 18 floors) = 0.924428777
p(you don't get two order floors after 19 floors) = 0.920401966
p(you don't get two order floors after 20 floors) = 0.916392695

Pretty sure it doesn't need to be any more complicated than that. Let me know if I'm wrong!

p(Two orders) isn't representative because that shows the probability of two order floors back-to-back.

[Nekradarken]

The fact that you are insisting entirely proves the opposite. Lamps are necessarily going to fuck you unless you can still guess 1/1. Also nice job mentionning you can win without running through walls, we totally believe you (why even mention it ?). I've always wondered why the entire BG pseudo elitist cheaters try their hardest to justify the event, implying that they win 1/3 of the time ("without cheats"). Very funny.

I know that you have excellent math skill and you mentioned that in general thaumas gears are inferior to MNK AF3+2 (my apology if it wasn't you). So I am just curious why do you keep arguing in this thread if this event is not worth doing?

[TSFFXI]

Yep, you're right. I realized it and refreshed hoping to make a quick edit, but you caught me. I'm going to get rid of it anyway, so that no one goes away from the thread thinking it's correct.

I'm hardly the person to answer this question, seeing as I haven't used probability for very much at all in the last four years, but since I'm awake I thought I'd give it another shot.

I count 120 (16!/[14!*2!]) ways for two order floors to be arranged among 14 non-lamp floors. Each of these occurs with probability p(two order floors)=[(0.0666666666666)^2 * (1-0.06666666666666666)^14]. Hence the probability is 120*[(0.0666666666666)^2 * (1-0.06666666666666666)^14]=0.203008209425225.

I continued this for 0->16 order floors for a 16 floor run, and I will now paste the results from my spreadsheet here in a way that will likely be indiscernible by anyone.

# of order floors _ Arrangements _ P(arrangement) _ P(# of order floors)
0 _ 1 _ 0.331580075394535 _ 0.331580075394535
1 _ 16 _ 0.0236842910996096 _ 0.378948657593754
2 _ 120 _ 0.00169173507854354 _ 0.203008209425225
3 _ 560 _ 0.000120838219895967 _ 0.0676694031417418
4 _ 1820 _ 0.000008631301421140 _ 0.0157089685864758
5 _ 4368 _ 0.000000616521530081 _ 0.00269296604339585
6 _ 8008 _ 0.000000044037252148 _ 0.000352650315206599
7 _ 11440 _ 0.000000003145518010 _ 0.0000359847260414897
8 _ 12870 _ 0.000000000224679857 _ 0.00000289162977119113
9 _ 11440 _ 1.60486E-11 _ 0.000000183595541028009
10 _ 8008 _ 1.14633E-12 _ 0.00000000917977705140043
11 _ 4368 _ 8.18804E-14 _ 0.000000000357653651353263
12 _ 1820 _ 5.8486E-15 _ 1.06445E-11
13 _ 560 _ 5.8486E-15 _ 3.27522E-12
14 _ 120 _ 4.17757E-16 _ 5.01309E-14
15 _ 16 _ 2.13141E-18 _ 3.41026E-17
16 _ 1 _ 1.52244E-19 _ 1.52244E-19

Hopefully that's not completely wrong. I'd post numbers for three, etc. order floors on 15-20 jump runs, but it's almost totally unreadable pasted here.
The short answer is that on a 16 jump run you should expect two order floors only 20% of the time.

[Jem]

For what it's worth, these are the numbers BG wiki links to

77.53% Chance of Fighting Critters:
26.87% Chance of Defeat All
17.18% Chance of Defeat 1 Enemy
16.74% Chance of Defeat an Enemy Family
16.74% Chance of Defeat an Enemy Leader (I took boss encounters off of this measurement)

21.59% Chance of Lamps:
8.37% Chance of Single Lamp
4.85% Chance of Ordered Lamps
8.37% Chance of Same time Lamp

.88% Chance of Free floors

Not sure on sample size. And like he says he has removed 20/40/60/80/100 floor objectives which are always Leader.

[Byrthnoth]

Yeah, so it's about a 5% lamp rate. If you do get lamps, it can be 3, 4, or 5 lamps in the order (2, 3, and 4 average guesses required if I'm not mistaken). Assuming an even distribution of those, you're looking to lose about 3 guesses worth of time on average every time you have a lamp floor. I know that a lot of my wins would have been losses if it had taken 3 attempts instead of 1 to clear lamps.

At the same time though, the odds of getting no lamps are about 40%. So even if every set of lamps is an insta-loss, they've only reduced your win rate by 60%. Even if you do get lamps (and assuming an even distribution again), your odds of going 1/1 are about 7% on average. 1/2 is higher and may not kill the run depending how much extra time you would have had, etc. I'd say that they reduced your odds of winning by something like 30-50% with the lamp patch.

[Spicyryan]

I know that you have excellent math skill and you mentioned that in general thaumas gears are inferior to MNK AF3+2 (my apology if it wasn't you). So I am just curious why do you keep arguing in this thread if this event is not worth doing?

Because I did not select a better post to throw a quick vent out about the random people who always seemingly think things were ninjaed with a sort of difficulty increase

[Eurell]

Pretty sure he wasn't talking to you lol

[TSFFXI]

Nevermind

[Byrthnoth]

3 Lamps: 6 combinations
4 Lamps: 24 combinations
5 lamps: 120 combinations

If there's an even distribution of 3, 4, and 5 lamps, then you would see the average of your odds of guessing it on the first guess.
1/6 + 1/24 + 1/120 = 20/120 + 5/120 + 1/120 = 26/120
26/120/3 = 7.2%

Spoiler: show

As for the average number of guesses, here it is worked out for 3 lamps:

Lets say 1, 2, 3 is the right order
1 2 3 = Correct
2 3 1 = None right
3 1 2 = None right
1 3 2 = 1 is correct
3 2 1 = 2 is correct
2 1 3 = 3 is correct

1/6 = Get it right first try
3/6 = Get one of the numbers right
2/6 = Get no number right

If you get none right, there are only two other valid solutions so you have a 50% chance of getting it on the second guess and a 100% chance of getting it by the third guess (2.5 guesses average).

If you get one right, you have a 100% chance of getting it on the next guess (2 average)

If you get all of them right, obviously you get it on the first shot.

2.5*2/6 + 2*3/6 + 1*1/6 = 12/6 = 2 average

If you work it out for 4 lamps, you get 3 average. I'm pretty sure 5 lamps would be 4 guesses average. So assuming an even distribution you're looking to waste an average of 3 attempts per floor. However, you don't always see the average. Some lamp floors are going to totally kill your run (6 guesses on a single floor means you're basically boned) and others are going to not matter (1/1, or 1/2 on a run where you would have had an extra minute, 1/3 on a run where you would have had an extra 2, etc.) If you get 2 Lamp floors on the same run, I'd guess you're going to be fucked on average.

[TSFFXI]

Agggh, you guys reply so fast. I was cOming in to delete that.