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

[Toxeus]

ok to add to what was just posted, idk if you have it factor'd in or not but.... it takes between 30 second to 1 min to enter and even get to floor 1 and it takes at least 2 to 2.5 mins to kill the 100 boss even with multiple 2 hours. so you could factor out 2.5 to 3.5 mins making the total time in nyzle 26.5 to 27.5 mins.

ima bring up a few things idk if were talked about..

You can set your preferred temp items at the nyzle assault orders npc, which for me is bravers, dusty wing, fools, fana, clerics, body boost. which saves time and costs 1400 tokens

i start a timer at second we enter nyzle which is when the countdown actually begins, and we all use body boost at 15 minute mark since body boost lasts 15 mins

tokens you get are based on the # of floors you get-ish, we usually get over 2k tokens for clearing 19ish floors.

[Motenten]

@Darron

You're right on that. I think the problem was that I was trying to calculate the probability as a single continuum, and adding the single floor jump calculations to the random calculations really screws it up. I'll rewrite it into separate charts and see how it goes.

Will tweak mine more tonight and see if I can get it cleaned up.

Was looking at your average time per floor, though, and it got me thinking... I wonder if I could put together a timing system in Spellcast that would allow people to see how well they're doing in their climbs? It might at least help people "get up to speed", as it were, by noting when/where they're failing and falling behind.

[Motenten]

Haven't gotten the spreadsheet probabilities to add up properly yet, when analyzing the non-stoppered scenario. Annoying.

However I do have an experimental Spellcast include file, if anyone wants to test it out to see if it's at all useful. Read the instructions at the top before using it (and only have one person per run use it).

This is solely a timer/notification tool to allow you to track your progress, and requires you to set the specific goal of the number of floors you're aiming to clear (have to edit the file manually for now). Probably most useful for those new to Nyzul, rather than those who are already good at maintaining their pace, but would still like to hear from anyone willing to try it out.

http://pastebin.com/Mzpg06aa

[Jem]

We're decent at maintaining pace but I'll give it a try if you need a test dummy.

[Motenten]

Ok, finally worked things out for climbing without a stopper, and including single jumps. Can't guarantee its accuracy, but it seems solid.

Results vary based on the amount of risk you're willing to take. That is, if you're at floors 92-98, do you take a random jump when there's a risk of going past 100? This is measured by the jump gap -- how many floors away from your target you're willing to still use a random jump. It has the same range of values as the limits on the jump -- 2 to 9 floors.

As the target of most interest is floor 100, will examine that.

         Gaps/Risk of overshooting
Jumps       9     Risk      8     Risk      7     Risk      6     Risk      5     Risk      4     Risk      3     Risk      2     Risk
13       0.01%   0.00%   0.01%   0.00%   0.01%   0.01%   0.01%   0.01%   0.01%   0.02%   0.01%   0.02%   0.01%   0.02%   0.01%   0.02%
14       0.07%   0.00%   0.11%   0.04%   0.15%   0.10%   0.17%   0.16%   0.18%   0.22%   0.20%   0.27%   0.20%   0.31%   0.21%   0.34%
15       0.40%   0.00%   0.64%   0.20%   0.83%   0.51%   0.98%   0.89%   1.09%   1.28%   1.18%   1.66%   1.24%   2.00%   1.29%   2.30%
16       1.45%   0.00%   2.25%   0.55%   2.92%   1.51%   3.47%   2.73%   3.93%   4.11%   4.30%   5.55%   4.60%   6.97%   4.79%   8.33%
17       3.85%   0.00%   5.72%   1.07%   7.36%   3.01%   8.78%   5.63%  10.02%   8.76%  11.08%  12.23%  11.93%  15.89%  12.46%  19.61%
18       8.21%   0.00%  11.65%   1.57%  14.77%   4.54%  17.60%   8.74%  20.13%  13.95%  22.32%  20.01%  24.05%  26.70%  25.07%  33.85%
19      14.97%   0.00%  20.34%   1.94%  25.34%   5.70%  29.97%  11.16%  34.15%  18.17%  37.73%  26.57%  40.45%  36.18%  41.97%  46.81%
20      24.34%   0.00%  31.85%   2.14%  38.94%   6.37%  45.52%  12.60%  51.40%  20.77%  56.29%  30.76%  57.54%  42.46%  44.27%  55.73%
21      36.40%   0.00%  46.10%   2.23%  55.24%   6.67%  63.62%  13.28%  70.90%  22.03%  67.14%  32.86%  54.27%  45.73%  39.46%  60.54%
22      50.74%   0.00%  62.50%   2.26%  73.43%   6.78%  83.18%  13.53%  77.48%  22.52%  66.29%  33.71%  52.91%  47.09%  37.38%  62.62%
23      65.92%   0.00%  79.40%   2.27%  91.62%   6.81%  86.39%  13.61%  77.33%  22.67%  66.01%  33.99%  52.44%  47.56%  36.65%  63.35%
24      79.53%   0.00%  94.22%   2.27%  93.18%   6.82%  86.37%  13.63%  77.28%  22.72%  65.93%  34.07%  52.31%  47.69%  36.43%  63.57%
25      89.53%   0.00%  97.73%   2.27%  93.18%   6.82%  86.36%  13.64%  77.27%  22.73%  65.91%  34.09%  52.28%  47.72%  36.38%  63.62%

Here we have a complex interaction between the number of jumps you make and the gap you've chosen as the limit. You need to be able to reach single-climb range earlier than your absolute max number of climbs, but the closer you try to get in that range, the higher the risk of overshooting.

Working down the list of the number of jumps you think you can clear, we can work out the highest chance for success based on the gap value.

Jumps       Gap
25            8
24            8
23            7
22            6
21            5
20          3-4
19            2
18            2
17            2
16            2
15            2
14          2-4
13          any

So until you can hit 20 jumps, you pretty much want to restrict yourself to randoms unless you hit floor 99.

But then there's the flip side -- if you -could- have cleared an additional floor, but took the risk and failed. At what point does the risk of failure outweigh the benefits?

What we want to look at here is whether the increase in risk as you drop the gap value is higher than the increase in chance of success. In all cases with a gap of 5 or less, the increase in risk is higher than the increase in reward, or at best 0.

However a better way of looking at it would be as such:
How much additional risk is incurred by dropping the gap by 1 point (eg: with 18 jumps, a 5 gap is 13.95%, a 4 gap is 20.01%)
If that additional risk was instead applied to an additional floor at the current gap level, how much of an increase in success would you see?
(eg: 18 floors at gap 5 is 20.13%; 19 floors at gap 5 is 34.15%; 20.01% - 13.95% = 6.06%; 34.15% * 6.06% = 2.07%
If that potential reward was added to the existing reward, would it beat the reward of dropping the gap value?
(eg: 20.13% for 18 floor/5 gap + 2.07% = 22.20%; 18 floor/4 gap is 22.32%, which is higher than the gain of converting the risk to an additional floor)

So for this example, the additional risk of lowering the gap is worth more than the chance at an additional floor at the current gap.

After accounting for that, we have a new set of optimal gap selections.

Jumps       Gap
25            8
24            7
23            7
22            6
21            5
20          3-4
19            3
18            3
17            3
16          2-3
15          2-3
14          2-4
13          any

For the common viable range of jumps -- 15 to 20 -- it looks like a gap of 3 is your best overall choice. That is, random jump on floors 97 or less, and only single climb from 98 or 99.


So, given that, we can give a quick summary of the chances of reaching floor 100 for a given number of jumps, using a 3 point gap (except for 21 floors, which gets a 5 point gap):

14      0.20%
15      1.24%
16      4.60%
17     11.93%
18     24.05%
19     40.45%
20     57.54%
21     70.90%

Obviously all this is moot once they introduce the stopper, but hopefully it can be useful for those using the current system.

[Motenten]

We're decent at maintaining pace but I'll give it a try if you need a test dummy.

Thanks

[takito09]

So.. if you guys have an sch embrava every run..... Wich is your method to reset their 2 hours every time?

Go to abyssea and buy a revitalizer ? Get cors ?

Because getting their 2 hours ready for 3 runs without waiting looks pretty hard.

[NynJa]

revitalizers arent purchasable, juuuuuuuuuuuus sayen

but getting in ISL in misar shouldnt take long (azure 2 eph murex, then cleave)

[transmit]

Our SCH's usually farm up a revit in abyssea in advance of our run. After that ones used up all the other members just take turns switching to COR and Wild Card.

[tenkiki]

So.. if you guys have an sch embrava every run..... Wich is your method to reset their 2 hours every time?

Go to abyssea and buy a revitalizer ? Get cors ?

Because getting their 2 hours ready for 3 runs without waiting looks pretty hard.

Did people forget about MMMv9 giving out revits to the whole pt for 3 marbles?

[NynJa]

Did people forget about MMMv9 giving out revits to the whole pt for 3 marbles?

someonessigNOONECARESABOUTMMM.jpg

[Jem]

Thanks

I couldn't get it working because I'm a spellcast idiot (I don't use i apart from for my whm and that's basic as hell). If you give me step by step instructions on how to get it working then I'll use it on our Wednesday runs. I created the .xml and put it in the right place but I didn't know where to put the extra info as I couldn't find the files it referred to in my spellcast folder.


In actual Nyzul news. We got 92 > 101 on our last run today. Thanks for that Nyzul!

[Motenten]

Save the pastebin code as NyzulTimer.xml, placed in your normal spellcast directory. Note: get the version that's up there now, I think I updated it after you mentioned you'd try it out.

In NyzulTimer.xml, edit the variables NyzulEventTrigger and UpdateTrigger. Change them to something you aren't using (eg: Footwork in NyzulEventTrigger, Mantra in UpdateTrigger). Whatever you have in UpdateTrigger, make a macro like "/ja Mantra NyzulFloor" that you can hit each time you climb a floor.


Save the following as drk.xml (or whatever job you're using):

<?xml version="1.0" ?>
<spellcast xmlns:xi="http://www.w3.org/2001/XInclude">
  
    <config Debug="false"
            ShowGearSwaps="false"
            ShowSpellInfo="false"
            RequireVersion="2.30"
            />

    <variables clear="True">
        <xi:include href="NyzulTimer.xml" xpointer="//include[@name='NyzulTimerVars']/*" />
    </variables>
    
    <sets>
        <group name="DefaultGroup" default="yes">
        </group>
    </sets>
    
    <rules>
        <xi:include href="NyzulTimer.xml" xpointer="//include[@name='NyzulTimer']/*" />
    </rules>
</spellcast>

And that should be all you need to do.


Edit: That's all you need to do to have it work. To actually run it (per instructions in the NyzulTimer.xml file), enter "/ja Footwork Nyzul:start" (or whatever ability you set the variable to) as soon as you enter the zone.

Double edit: woops, just realized that one part of it won't work properly if you edit the trigger actions. Let me edit it and reupload to pastebin.

Ok, fixed version uploaded. http://pastebin.com/Mzpg06aa

[Jem]

Ok thanks. I will use it for our runs on Weds night. Probably be 3-4 of them.

[Nemesio]

So.. if you guys have an sch embrava every run..... Wich is your method to reset their 2 hours every time?

Go to abyssea and buy a revitalizer ? Get cors ?

Because getting their 2 hours ready for 3 runs without waiting looks pretty hard.

Our DD's normally don't use 2hr unless we hit 100/80 and need em. Normally leaving 3-4 people with 2hr's available to try at COR. That with a prefarmed Revit, and mules(With COR as well) is normally enough for us to get 3-4 runs in with double Embrava.

[Kipling]

Killed the rampart beforehand for maximum TH.

Spoiler: show

http://i41.tinypic.com/9u0qrc.jpg

[Toxeus]

/names will remove that glitchy name problem, aka how i deal with it

[Kipling]

Ah, it doesn't bug me for some reason. Been using fillmode since old nyzul was new so I'm used to it.

[ikariiiii]

Killed the rampart beforehand for maximum TH.

lol'd pretty hard. my group has followed this superstition and now we're seeing more drops than ever. up next: planetary alignment drop values based on the in-game constellation/avatar map

[Jem]

Used that spellcast file and it works fine although it did give us a Body Boost warning twice on the same run at different times.

I changed all the timer messages to echo too as it was annoying people (The blah blah of blah blah seconds ones). Had to add an extra second of wait too so that the time remaining message didn't spam too fast. I think we'll just use it for the number of floors cleared and the time remaining messages. If I think of any suggestions I'll post them, just in between runs atm.


Unrelated sidenote. Gettings Spec Fam Flayers on a floor that already has Flayers on is a huge pain in the ass. I didn't think it could happen but it did. It meant that we had some Flayers that were ITG and some that were DC so had to check everything. We had it earlier with puddings too but they are easier to differentiate between. And fuck off WS restricted on Spec Fam floors following by attack speed down the next floor.