+ML where and where not?

This is where you spade, shovel and sickle. Or is it reap?
Post Reply
Brulak
AFH
Posts: 945
Joined: Sat Apr 14, 2007 2:51 pm
Location: Hamburg, Germany

+ML where and where not?

Post by Brulak » Wed Jan 09, 2008 4:10 am

Given that the KoL forums discuss the stats of +ML vs. the need to obtain certain adventures rather quickly (see recent post by emily yaffle), here's some data for you to base your pull decisions on:

+ML is good for you here:

Council quests:
- Mosquito (IMHO more stats early on wins over getting the temple before the superlikely mosquito)
- Tavern
- Bathole
- All of the Knob
- Friars part 2
- Cyrpt: Areas that don't have nocombats boosting your main stat (two out of skeletons, zmobies, lihces)
- Mines if you have good +items
- Goatlet
- entire Pirate quest
- Valley
- Hole in the Sky
- entire level 11 quest sans pyramid, excepting ballroom and palindome as moxie class
- entire level 12 quest except for starting the war

Spookyraven:
- Haunted Pantry
- Billard room until you have the cue
- Ballroom if not a moxie class
- Cellar

Other:
- Degrassi knoll
- 8 bit realm

no +ML (except MCD which doesn't ruin noncombats) is better here:

Spookyraven:
- Billard room after cue
- Library
- Bedroom
- Gallery / Bathroom / Ballroom for mainstats

Council quests:
- Friars Part 1
- Cyrpt Ghuols and your mainstat nocombat area
- Airship
- Castle
- Palindome if moxie class
- MacGuffin, Pyramid part
- Starting the Island war.


Unknown:
- Dungeons of Doom

mod edit: fix html link

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Wed Jan 09, 2008 7:13 am

I thought using +noncombats in the Cyrpt let you find the bosses earlier... but then I'm not sure exactly how the boss encounters work.

Brulak
AFH
Posts: 945
Joined: Sat Apr 14, 2007 2:51 pm
Location: Hamburg, Germany

Post by Brulak » Wed Jan 09, 2008 7:24 am

I did not record turncounts for this, but I "feel" there is a set minimum of turns like eight per area or so. Mind you, do not increase combat rate there, this does turn away bosses. But just pump ML and walk away from those off-stat non-combats. Has worked ok for me over the last runs to split the cyrpt that way.

greycat
AFH
Posts: 359
Joined: Sat Sep 22, 2007 9:42 pm

Post by greycat » Wed Jan 09, 2008 9:57 am

For hardcore, the list will look a bit different. You won't have massive, huge, gargantuan +item and +ML modifiers in hardcore, so there are some places where you want noncombats, or at least want to consider them:
  • Harem. Getting the outfit from combat drops is a crapshoot at best. You probably want the outfit noncombat choice adventures.
  • Itznotyerzitz Mine. As above.
  • Obligatory Pirate's Cove. As above (although the drop rates are quite a bit better here).
  • Barrrney's Barrr. The noncombats here give some good stats, or shots of rotgut which aren't bad hardcore booze.

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Wed Jan 09, 2008 6:39 pm

Brulak wrote:I did not record turncounts for this, but I "feel" there is a set minimum of turns like eight per area or so. Mind you, do not increase combat rate there, this does turn away bosses. But just pump ML and walk away from those off-stat non-combats. Has worked ok for me over the last runs to split the cyrpt that way.

Okay, since we're spending about 35 turns here and don't really know how to optimise it, is anybody interested in spading this once and for all? I'd suggest the following:

Test 1: Determine the minimum turncount to activate the boss encounter
1) Freshly ascended character with +24% noncombats (+29% if you want to do the ballroom first).
2) Adventure in the first subzone - Green smoke bomb from combats and skip noncombats. Do this a few times (10?) then kill a monster to use up a turn.
3) Repeat (2) until you find the boss.
4) Do this for all four subzones.

Test 2: Determine the first instance of superlikeliness
-- Same as the first test but using a character with +20 Combats.

It will cost hundreds of green smoke bombs/tattered scraps and will be pretty tedious... calling all volunteers.

EDIT: This bugs me, so I'll do it.
Last edited by Seal Lubber on Thu Jan 10, 2008 6:38 pm, edited 1 time in total.

User avatar
QuantumNightmare
AFH
Posts: 1827
Joined: Mon Jan 29, 2007 10:28 pm
Location: Montreal

Post by QuantumNightmare » Thu Jan 10, 2008 5:29 pm

Seal Lubber wrote:Test 1: Determine the minimum turncount to activate the boss encounter
1) Freshly ascended character with +24% noncombats (+29% if you want to do the ballroom first).
2) Adventure in the first subzone - Green smoke bomb from combats and skip noncombats. Do this a few times (10?) then kill a monster to use up a turn.
3) Repeat (2) until you find the boss.
4) Do this for all four subzones.
This part of the test seems interesting to me, although you seem to be the one with the most crypt knowledge at the moment. I'm willing to donate a few 100k if you're willing to work on it.

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Thu Jan 10, 2008 10:34 pm

Noncombats=+24% (sonata, fresh scent, conflict, ballroom, yendorian)

What I Did: Green smoke bomb all combats, skip all noncombats, kill the boss.

Legend: c-combat, n-noncombat, B-boss, X-killed monster
------------------------------------------------------------------------

Alcove:
cccncncccnXcccncnncncB

n=7
c=14

Nook:
ccncncncncncB

n=5
c=7

Niche:
ccccccccnnnnnnnccnccnccB

n=9
c=14

Cranny:
nccccncnnnccnccncncnB

n=9
c=11

------------------------------------------------------------------------

What Happen?: The original plan was to skip at least 10 encounters then kill a monster to advance the turncount. Turns out that wasn't necessary, since the bosses could be found without spending any turns.

So?: I've previously done a test at +15 combats, which showed that the boss encounters consistently (well, 4 times) show up after 12 turns. This most recent test shows bosses appearing after 0 turns, but there is a minimum of 12 encounters before a boss is found. I have no idea what the actual mechanism is.

-You could run max +ML: You will find the boss after 12 turns. (This was consistent for 4 sub-zones, haven't tested it further than that)
-You could run max +noncombats: By skipping the noncombats, you may be advancing the encounter counter, thereby possibly allowing you to find the boss in less than 12 turns. (I've found a boss in 7 turns myself).


Theory
There is a range of 12-23 encounters before a boss is found in this most recent test. Perhaps the mechanism is something like this:

-After 12 encounters, possibly give the boss (say, 20%)
-After 12 turns, give the boss

So by pumping noncombats, you give yourself a chance at speeding up the zone, but it involves a bit of RNG. It won't take longer than 12 turns though. The reason I'm liking this idea is that I vaguely remember talk about this sort of mechanism by Jick in one of the NS-13 whine threads.

salien
Pie in the Sky
Posts: 115
Joined: Fri Jan 11, 2008 1:50 pm

Post by salien » Fri Jan 11, 2008 2:29 pm

I'd be interested to see how frequently you get the boss encounter after encountering it once; it may shed some light on how it occurs once whatever 'threshold' is achieved. I think the newest build of Mafia should have a choice preference for skipping the boss encounters, so this should be easy to spade; just set it to skip the boss and then run a whole bunch of turns, with and without combat modifiers. Mostly I'd be curious to see if it becomes periodic (i.e. every 12 turns), is affected by +/- noncombat, or is superlikely at some fixed percent.

User avatar
KujjieKujjieKoo
Spade Ninja
Posts: 2240
Joined: Mon Apr 30, 2007 8:50 pm
Location: !Chicago
Contact:

Post by KujjieKujjieKoo » Fri Jan 11, 2008 2:39 pm

The old spading thread "The Crypt: Strange Happenings" has been moved to the public forums. Feel free to grab any other useful information out of that one.
Image

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Wed Feb 13, 2008 7:53 am

Seal Lubber wrote:Theory
There is a range of 12-23 encounters before a boss is found in this most recent test. Perhaps the mechanism is something like this:

-After 12 encounters, possibly give the boss (say, 20%)
-After 12 turns, give the boss
Had another crack at this and came up with more questions than answers. Data follows:

Niche (top right):
+15% 4 combats, killed 4 liches
10 turns at hidden temple
+15% 8 combats
-19% 2 combats
-19% 1 non-combat, skipped
-19% Boss, killed

**14 combats, a non-combat, then a boss immediately following. So far so good, behaving as expected by the above theory.


Cranny (bottom right):
+15% 8 combats
+15% Boss, killed

**8 combats, then a boss. Well, maybe the minimum is less than 8 turns.


Nook (bottom left):
+15% 11 combats
+15% Boss, killed

**11 combats, then a boss. Back to normal maybe?


Alcove (top left):
+15% 25 combats
-19% 10 combats
-19% 1 non-combat, skipped
-19% Boss, killed

**35 combats, a non-combat, then a boss. No, not back to normal at all. This kind of blows away any theory I've come up with so far.

Another multi will pass through tomorrow.

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Thu Feb 14, 2008 4:49 am

Recent radio show transcript:
44:27
Q: Why do you lose a charge in the El Vibrato portal even if you use an item that allows you to run away without using an adventure?
A: The charge is used up when you go there, not because you killed something. If you use a tattered scrap of paper, it still counts as you being in that zone. If there is something that requires you to be in a zone for x turns, you can increment that without using up the turn.
So if there is any turncount mechanic going on, the skippable nature of the non-combats will very likely let you find a boss more quickly.

alacrity
AFH
Posts: 501
Joined: Sun May 27, 2007 6:55 am
Location: Santa Barbara, CA
Contact:

Post by alacrity » Thu Feb 14, 2008 3:31 pm

Could this be a mechanic built on the same core mathematics as the Wormwood choice adventures? I haven't done formal spading there but it seems that non-combat% boosts don't speed up the occurance of those choice adventures at all, and that combat% boosts don't eliminate them.

I have a lot of data (some of which is at +5% combats, some of which isn not) and the distribution of the first choice is evenly distributed among the 3 possibilities (+/- 5% statistical error) over 6000 wormwood loops. Similarly the second choice is evenly distributed amont the 2 possible options, and obviously the last choice only has one turn it can occur on.

This could be modeled as a simple varying encounter rate, like so:
Encounter% = 1 / (MaxTurns - TurnsSpentInZone)

With 3 max turns (e.g. 1st Wormwood choice adv) chance of getting it on the 1st turn is 1/3, 2nd turn is 1/2, 3rd turn is 1/1, caveated with the "you only get it once" limitation in the wormwood.

Obviously that limitation doesn't exist in the sub-zones of the cyrpt. But it may be useful to test this model against cyrpt data with maybe some base number of turns with a 0% frequency, and then an increasing frequency following a probability curve of 1/(max - spent).

One quick test that seems interesting here is to see whether you can tattered scrap Wormwood combats and increase the turn counter for the purposes of getting the choice adventure. If not that would indicate that the mechanics involved in the Wormwood are not related to the mechanics alluded to in the J&S radio transcript.

Is this too out there? Are the devs more likely to reuse basic mechanics like this, or build totally new ones?

Antipasta
AFH
Posts: 251
Joined: Fri Sep 21, 2007 5:24 pm

Post by Antipasta » Mon Feb 18, 2008 6:21 pm

I have to admit that I am completely confused about what to do in the cyrpt. I'd been thinking that the best approach was to try to get the choice adventure, and decline it, in hopes that it would bring the boss on. But I haven't had much luck with that - on my last run, declining the choice seemed to delay the boss arrival, compared with the zones where I accepted it (see below). I also had a recent run where my very first cyrpt adventure was a boss.

And I can't make anything out of the posted data in this and the other thread! Help! Does anybody have a current guess at best practice?

here's my last run, with noncombat boosted 15%, with the zones in order (c=combat, nc=noncombat stat/item choice adv):
Nook
c, c, c, c, c, c, c, c,
nc (accept)
c, c,
boss

Cranny
c, c, c
nc (accept)
nc (accept)
boss

Niche
c, c, c, c, c, c
nc (decline)
c, c
nc (decline)
c
nc (decline)
nc (decline)
c
nc (decline)
c
nc (decline)
nc (decline)
c
boss

Alcove
nc (decline)
c, c, c
nc (decline)
c, c, c, c, c
nc (decline)
c, c, c, c
nc (decline)
c
boss

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Mon Feb 18, 2008 9:04 pm

Antipasta wrote:And I can't make anything out of the posted data in this and the other thread! Help! Does anybody have a current guess at best practice?
I've done a few spading runs through and still have no idea what's going on... I've tried to summarise what has been discovered so far and has reasonable evidence to show for it.

- If you run +15%combats, from the start, in the Alcove, you won't find a boss in any amount of turns. This suggests that a non-combat is required to unlock the Boss.

- After a certain amount of turns, the Boss is guaranteed. Evidence of this is several occasions of running +15%combats for many (15-50) turns, finding a non-combat and then immediately finding the Boss. This was tested in the Alcove at least 3 times, the Niche once.

- There is no minimum turn count; you can do find a boss in zero turns if you green smoke bomb away from all combats. However, it may take up to 23+ encounters to find a Boss with this method, while I've never seen it take more than about 12 turns using conventional adventuring.

- All zones may not be equal. In my most recent test I started in the Niche instead of the Alcove. Finding a non-combat in the Niche should have unlocked the Boss in the Alcove if things were being consistent. However, when I finally got to the Alcove the Boss was nowhere to be seen and I needed another non-combat to find it. This is just confusing, because I've seen that both the Alcove and the Niche need a non-combat to unlock the Boss, but doing the Alcove first unlocked the Niche while doing the Niche first failed to unlock the Alcove.

Some other thoughts which may or may not be accurate:
- The regular monsters you fight are probably not part of the mechanic. I'd tend to think that if they were involved you'd have to kill them rather than just find them. Since you don't need to kill anything to find a Boss, I don't think they're involved.
- I suspect some sort of global turn count used by Jick to prevent RNG screw. My first test involved about 70 turns before finding a non-combat in the Alcove before finding a Boss since I was at +15%combats. After that, all of the Bosses were found either immediately or before <6 turns, I can't remember which exactly. Also, in my most recent test I spent a few extra turns in the Niche (14), while the next zone (Cranny) was done in only 8 turns. The zone after that (Nook) was a more typical 11 turns. This is very much a guess.
- Best practice would I think be to first try and prove that non-combats are the unlock mechanism. This can be done by running +15%combats in all of the zones for 30ish turns. I suspect you'll find you don't encounter any Bosses. Then we need to know if a non-combat from one zone unlocks the Boss encounter in all other zones; there are conflicting results for this one. Once the unlocking mechanism is nailed down, you can try to figure out when exactly the Bosses will show up in each zone. Grab a single non-combat, pump +combats and count turns. Use green smoke bombs and count encounters. Pump combats in the first zone for 20 turns then kill the boss, see if that effects other zones. Maybe it's nothing to do with counters at all. Maybe there are other ways to speed it up to do with the regular non-combats. Those are just some ideas at the moment...

User avatar
KujjieKujjieKoo
Spade Ninja
Posts: 2240
Joined: Mon Apr 30, 2007 8:50 pm
Location: !Chicago
Contact:

Post by KujjieKujjieKoo » Mon Feb 18, 2008 9:20 pm

Seal Lubber wrote:Best practice would I think be to first try and prove that non-combats are the unlock mechanism.

Sorry Seal... current run, Disco Bandit, Blender Sign. Fought all combats, no run aways, no special items. (edit: and no buffs or combat modifiers either!)

Code: Select all

[551-559] Defiled Cranny
gluttonous ghuol
gaunt ghuol
gaunt ghuol
gaunt ghuol
gluttonous ghuol
gluttonous ghuol
gaunt ghuol
gluttonous ghuol
Lunchtime: huge ghuol

[560-574] Defiled Nook
toothy sklelton
spiny skelelton
toothy sklelton
toothy sklelton
Skull, Skull, Skull: Moxie
toothy sklelton
spiny skelelton
toothy sklelton
toothy sklelton
toothy sklelton
spiny skelelton
Skull, Skull, Skull: Moxie
toothy sklelton
toothy sklelton
Pileup: giant skeelton

[575-586] Defiled Alcove
corpulent zobmie
grave rober zmobie
grave rober zmobie
corpulent zobmie
Turn Your Head and Coffin: Combat item - Brain
corpulent zobmie
grave rober zmobie
corpulent zobmie
corpulent zobmie
corpulent zobmie
grave rober zmobie
Doublewide: conjoined zmombie

[587-595] Defiled Niche
senile lihc
Urning Your Keep: Combat Item - Plus size
slick lihc
Urning Your Keep: Meat
slick lihc
senile lihc
senile lihc
senile lihc
Lich in the Niche: gargantulihc

[596] Haert of the Cyrpt
Bonerdagon
Last edited by KujjieKujjieKoo on Mon Feb 18, 2008 9:47 pm, edited 1 time in total.
Image

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Mon Feb 18, 2008 9:42 pm

Oh well, making progress.

Alcove: Almost definitely (I like that phrase) needs a non-combat, not sure about turn-counts etc. We've had three separate tests of running +combats for large numbers of turns without seeing a Boss.

Cranny: No non-combat required, seems to not take too many turns. Antipasta-3 combats, 2 noncombats; Seal- 8 combats; Kujjie- 8 combats.

Niche: I went 14 turns before a non-combat, Boss next turn. Need to spend more turns here.

Nook: Haven't started a test here yet, not sure. I've got Soda Seven waiting in the wings to ascend and do this when I have time.

greycat
AFH
Posts: 359
Joined: Sat Sep 22, 2007 9:42 pm

Post by greycat » Wed Feb 20, 2008 8:13 am

hardcore pastamancer opossum
smooth + sonata + MCD 11 the whole time

cranny (SE): GLUT drawers(2) GAUNT drawers(2) GLUT GLUT GLUT GAUNT GLUT GLUT drawers(2) GAUNT GLUT GAUNT drawers(2) boss

alcove(NW): CORP ROBER CORP (left zone) coffin(skip) coffin(skip) ROBER CORP (rollover) CORP CORP coffin(skip) coffin(skip) CORP ROBER CORP boss

nook (SW): TOOTH skulls(skip) skulls(skip) skulls(skip) skulls(skip) SPINY SPINY skulls(skip) TOOTH skulls(skip) skulls(skip) skulls(skip) skulls(skip) TOOTH TOOTH SPINY TOOTH skulls(skip) SPINY skulls(skip) SPINY SPINY skulls(skip) TOOTH TOOTH SPINY SPINY boss

niche (NE): urn(1) SENILE SLICK SLICK SENILE boss

Summary:
cranny: 16 turns, 11 regular combats, 4 choices (taken), 1 boss
alcove: 11 turns, 10 regular combats, 4 choices (skipped), 1 boss
nook: 16 turns, 15 regular combats, 12 choices (skipped), 1 boss
niche: 6 turns, 4 regular combats, 1 choice (taken), 1 boss

User avatar
QuantumNightmare
AFH
Posts: 1827
Joined: Mon Jan 29, 2007 10:28 pm
Location: Montreal

Post by QuantumNightmare » Fri Feb 22, 2008 10:42 pm

[hardcore] Kyrivian: Wow. Huge ghoul bos battle on the first adventure in that part of the Cyrpt. Thanks, RNG.

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Fri Mar 07, 2008 9:37 pm

More Test (Casual Sauceror, Blender):

Cranny +15% combats
[177] gaunt ghuol (Harold's Bell)
[178] gluttonous ghuol (Green Smoke Bomb)x4
[178] huge ghuol (Kill)

Nook +15% combats
[179] spiny skelelton (divine champagne popper)
[179] toothy sklelton (Green Smoke Bomb) x10
[179] giant skeelton (Kill)

Niche +15% combats
[180] slick lihc (divine champagne popper)
[180] senile lihc (Green Smoke Bomb) x9
[180] gargantulihc (Kill)

Alcove +15% combats
[181] grave rober zmobie (divine champagne popper)
[181] corpulent zobmie x25
[181-187] Hidden Temple (wear off popper effect, which I now realise wasn't necessary since I could just popper a corpulent zmobie for the same effect)
[188] grave rober zmobie (Green Smoke Bomb)
[188] corpulent zobmie (divine champagne popper)
[188] grave rober zmobie (Green Smoke Bomb) x44
Drop to +10% Combats
[188] grave rober zmobie (Green Smoke Bomb) x38
[188] Turn Your Head and Coffin (Skipped)
[188] conjoined zmombie (Kill)

-Cranny: again, low number of turns
-Nook: no non-combat required
-Niche: no non-combat required
-Alcove: still behaving as expected.
Last edited by Seal Lubber on Sat Mar 08, 2008 12:16 am, edited 2 times in total.

User avatar
KujjieKujjieKoo
Spade Ninja
Posts: 2240
Joined: Mon Apr 30, 2007 8:50 pm
Location: !Chicago
Contact:

Post by KujjieKujjieKoo » Fri Mar 07, 2008 9:54 pm

3 of the crypt zones very clearly correspond to a specific mainstat. I wonder if the turn requirements vary depending on your current class? ie: Need more turns in your mainstat zone to complete than your offstat area?

Request for future data: List your class and moon sign.

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Sat Mar 08, 2008 12:10 am

More Test (Casual Disco Bandit, Blender):

Cranny +15% combats
[158] ghuols (Green Smoke Bomb)x10
[158] huge ghuol (Kill)

Nook +15% combats
[159] skleltons (Green Smoke Bomb) x4
[159] giant skeelton (Kill)

Niche +15% combats
[160] lihcs (Green Smoke Bomb) x13
[160] gargantulihc (Kill)

Alcove +15% combats
[161] zobmies x25
Drop to -20% Combats
[161] zmobies (Green Smoke Bomb) x1
[161] Turn Your Head and Coffin (Skipped)
[161] conjoined zmombie (Kill)

-This time the Nook was quick
-Alcove shown to need a non-combat as Sauceror and Disco Bandit

User avatar
Flolle
Fie the Pie
Posts: 848
Joined: Mon Apr 02, 2007 10:13 am

Post by Flolle » Sat Mar 08, 2008 7:51 am

Hmm, interesting. Maybe try a different order to see whether this is really an Alcove only mechanic.

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Sat Mar 08, 2008 5:29 pm

Flolle wrote:Hmm, interesting. Maybe try a different order to see whether this is really an Alcove only mechanic.

The data's fairly scattered, but I count three tests run at +15% combats and with different orders in which the alcove is the only place which needs a noncombat.

lordhades15
Fie the Pie
Posts: 1175
Joined: Sat Feb 03, 2007 12:02 am
Location: Kent, OH

Post by lordhades15 » Tue Mar 11, 2008 9:16 am

More Testing on the Crypt (Casual, Sauceror, Blender), in order visited.

Used smoke bomb to skip combats (only killing bosses), skipped noncombats.

All tests were run with +15% combats (had noncombat ballroom song playing, but used +26 ML to counteract it).


Alcove:

11 Zombies
Noncombat
Zombie Boss

Niche:

8 lihcs
Noncombat
2 lihcs
Noncombat
2 lihcs
Boss

Nook:
9 skeletons
Boss

Cranny:
8 ghuol
Noncombat
4 ghuol
Boss

Interpret as you will.
j12601 wrote: Turns are now sub-optimal. It is a brand new fucking day.

User avatar
KujjieKujjieKoo
Spade Ninja
Posts: 2240
Joined: Mon Apr 30, 2007 8:50 pm
Location: !Chicago
Contact:

Post by KujjieKujjieKoo » Tue Mar 11, 2008 9:25 am

lordhades15 wrote:All tests were run with +15% combats (had noncombat ballroom song playing, but used +26 ML to counteract it).

Plus combats and Minus combats are summed first, then +ML is checked. So if you had +15% combats from equip/gear and -5% combats from the song, your effective combat modification is +10%. The game then checks to see if ML needs to be applied, which is does not since the combat modifier is positive.

lordhades15
Fie the Pie
Posts: 1175
Joined: Sat Feb 03, 2007 12:02 am
Location: Kent, OH

Post by lordhades15 » Tue Mar 11, 2008 9:28 am

Thank you for the clarification, Kujjie.

Still, at +10% combats, got boss in Alcove. That is contrary to findings thus far, no?
j12601 wrote: Turns are now sub-optimal. It is a brand new fucking day.

User avatar
KujjieKujjieKoo
Spade Ninja
Posts: 2240
Joined: Mon Apr 30, 2007 8:50 pm
Location: !Chicago
Contact:

Post by KujjieKujjieKoo » Tue Mar 11, 2008 9:43 am

lordhades15 wrote:Still, at +10% combats, got boss in Alcove. That is contrary to findings thus far, no?

As I understand what SL has found, a non-combat must be encountered before the Alcove boss will appear. That is consistent with your data point as well.

More data: Sauceror, Oppossum Sign, Softcore, 0% combat modifiers, MCD @ 11.

[255-261] Defiled Cranny: (0 non-combats before boss)
gaunt ghuol
gluttonous ghuol
gluttonous ghuol
gluttonous ghuol
gaunt ghuol
gluttonous ghuol
Lunchtime (huge ghuol)

[262-276] Defiled Niche (3 non-combats before Boss)
slick lihc
senile lihc
senile lihc
Urning Your Keep (Myst)
Urning Your Keep (Myst)
slick lihc
slick lihc
Urning Your Keep (Myst)
slick lihc
slick lihc
senile lihc
slick lihc
senile lihc
Lich in the Niche (gargantulihc)

[277-287] Defiled Alcove (1 non-combat before boss)
corpulent zobmie
Turn Your Head and Coffin (Meat)
grave rober zmobie
corpulent zobmie
grave rober zmobie
grave rober zmobie
grave rober zmobie
corpulent zobmie
grave rober zmobie
corpulent zobmie
Doublewide (conjoined zmombie)

[288-293] Defiled Nook (0 non-combats before boss)
toothy sklelton
spiny skelelton
toothy sklelton
spiny skelelton
toothy sklelton
Pileup (giant skeelton)

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Tue Mar 11, 2008 9:50 am

lordhades15 wrote:Still, at +10% combats, got boss in Alcove. That is contrary to findings thus far, no?
Other than being reasonably lucky to hit a non-combat in 12 turns, this is expected behaviour. The non-combat rate is 15% (in the Alcove, not completely sure about the other zones), so while running +10% combats finding a non-combat and a Boss immediately after would be the most likely result.

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Thu Mar 20, 2008 8:04 am

More Test (Casual Disco Bandit, Blender):

Alcove -19% combats
[93] zobmie (Kill) x1
[94] Turn Your Head and Coffin (Skipped) x1
Change to +20% Combats
[94-103] zmobies (Kill) x10
[104] conjoined zmombie (Kill)

Niche +20% combats
[105-117] lihcs (Kill) x13
[118] gargantulihc (Kill)

Nook +20% combats
[119-132] skleltons (Kill) x14
[133] giant skeelton (Kill)

Cranny +20% combats
[134-145] ghuols (Kill) x12
[146] huge ghuol (Kill)

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Thu Mar 20, 2008 9:41 am

Going to throw out some theories (please try to pick holes in them):

Assumption 0 -- The Boss is not a conventional non-combat.

Evidence 0 -- There's plenty of evidence that the Boss is a Superlikely, and that is isn't a regular non-combat initially. It is harder to tell whether it turns into a regular non-combat at some point, but I assume it doesn't.

Assumption 1 -- There are three mechanics possible:
Mechanic (A)- The first is what I will call Encounter Count. In this mechanic, each time you adventure, regardless of what happens, a counter is decreased until finally a Boss is found.
Mechanic (B)- Increasing Chance Superlikely. In this mechanic, each time you adventure, regardless of what happens, the chance of finding the Boss will increase.
Mechanic (C)- Minimum number of encounters, with increasing superlikely once this minimum is reached. Just a combination of the first two.

Evidence 1 -- I'll try to support/put holes in Assumption 1 above:
Mechanic (A)- Evidence is that tests have shown that adventuring at +15 combats and Green Smoke Bombing away will still allow the Boss to be found. Therefore any counting mechanic must be based on encounters and not turns.

However, there has been one report of a 1-turn Boss. It would be dangerous to dismiss this, so I'll discuss it here. My thought is that if this report is accurate, then it would disprove this mechanic. The reason for this is that if there was a minimum Encounter Count, surely it would be fixed? If it was fixed, it would be pointless to make it fixed at only 1 turn. There is the possibility of a randomly generated minimum encounter count. This has precedent in semi-rare counters, but I can't help but feel that it would be out of place in the Cyrpt. The semi-rare mechanic is optional with a small range and which doesn't cost turns, while the Cyrpt would be required with an apparent range of about 20-60 turns (4-15 each subzone). That much RNG is ridiculous when you have no control over it (or is it).

Mechanic (B)- A 1-turn Boss has been reported. I've counted 37 legit tests in this thread, but the quickest we've found a Boss is 4 turns. If at turn 0 the chance was 0%, and turn 20 the chance was 100%, this might explain the data we've got (including the rare 1-turn Boss). You'd be lucky to do it quicker than 4 turns, and unlucky to hit 16 turns.


Assumption 2 -- Only combat encounters are counted in whatever the mechanic is.

Evidence 2 -- In the mechanics suggested, finding and skipping noncombats would be expected to decrease the time taken to finish the quest. However, this might not be the case:
Mechanic (A)- nook: 16 turns, 15 regular combats, 12 choices (skipped), 1 boss. 37 encounters is huge, perhaps only combats increase the chance of finding the Boss? The +15% combat tests have a range of 4-15 combats; when you look at the +non-combat tests, it looks as though the same 4-15 turn range is in place, but only if you ignore the many skipped noncombats.
Mechanic (B)- With a % chance of finding a Boss, lots of skipped non-combats should let you find the Boss earlier. I don't think there's enough data currently to say either way.


Testing -- So how to test these assumptions and hopefully pin down one of the mechanics:
More tests are needed with as much +noncombats as people can muster, skipping all non-combat encounters.
- If these tests require less combat encounters than +combat tests, then we could say that the Bosses are Increasing Chance Superlikely. This is because the skipped non-combats will give more chances of hitting a Boss.
- If, despite boosting non-combats, the tests find the same number of combat encounters as in the +combat tests, this would be evidence of a fixed combat number mechanism. This is because with a counter mechanism which isn't decremented by the skipped non-combats, the number of combats required will obviously be independent of combat rate.

Fatal Flaws:
- Assumption 2 is not necessarily true, although it is consistent for 12 tests in this thread.
- Something strange where a Boss is a % chance when a combat is encountered would make the test above pointless.
- Another mechanic entirely involving item drops, ML, speed of European Swallows, etc.
- I don't even want to think about Mechanic C suggested initially.
- Maybe at some point, the Boss becomes a regular non-combat. It would be hard (or impossible) to determine this from the test I've suggested.

--Anyhoo, just a suggestion for further testing. Anybody have more theories or a better plan?

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Fri Apr 04, 2008 3:10 am

More Test (Casual Disco Bandit, Blender):

Alcove -24% combats
[125] Turn Your Head and Coffin (Skipped)
[125] zobmie (Kill)
[126] Turn Your Head and Coffin (Skipped)
[126] Turn Your Head and Coffin (Skipped)
[126] Turn Your Head and Coffin (Skipped)
[126] zmobie (Kill)
[127] zmobie (Kill)
[128] zmobie (Kill)
[129] Turn Your Head and Coffin (Skipped)
[129] Turn Your Head and Coffin (Skipped)
[129] Turn Your Head and Coffin (Skipped)
[129] zmobie (Kill)
[130] Turn Your Head and Coffin (Skipped)
[130] Turn Your Head and Coffin (Skipped)
[130] zmobie (Kill)
[131] Turn Your Head and Coffin (Skipped)
[131] zmobie (Kill)
[132] zmobie (Kill)
[133] Turn Your Head and Coffin (Skipped)
[133] zmobie (Kill)
[134] Turn Your Head and Coffin (Skipped)
[134] conjoined zmombie (Kill)

- 9 combats, 12 noncombats. Interesting that a boss directly followed a noncombat, so maybe it is a percent chance.

Niche -24% combats
[135] lihc (Kill)
[136] Urning Your Keep (Skipped)
[136] lihc (Kill)
[137] lihc (Kill)
[138] lihc (Kill)
[139] lihc (Kill)
[140] lihc (Kill)
[141] gargantulihc (Kill)

- 6 combats, 1 noncombat

Nook -24% combats
[142] sklelton (Kill)
[143] sklelton (Kill)
[144] Skull, Skull, Skull (Skipped)
[144] sklelton (Kill)
[145] Skull, Skull, Skull (Skipped)
[145] sklelton (Kill)
[146] sklelton (Kill)
[147] Skull, Skull, Skull (Skipped)
[147] sklelton (Kill)
[148] Skull, Skull, Skull (Skipped)
[148] sklelton (Kill)
[149] sklelton (Kill)
[150] Skull, Skull, Skull (Skipped)
[150] sklelton (Kill)
[151] Skull, Skull, Skull (Skipped)
[151] sklelton (Kill)
[152] giant skeelton (Kill)

- 10 combats, 6 noncombats

Cranny -24% combats
[153] ghuol (Kill)
[154] Go Slow Past the Drawers (Skipped)
[154] Go Slow Past the Drawers (Skipped)
[154] ghuol (Kill)
[155] ghuol (Kill)
[156] Go Slow Past the Drawers (Skipped)
[156] ghuol (Kill)
[157] ghuol (Kill)
[158] Go Slow Past the Drawers (Skipped)
[158] ghuol (Kill)
[159] ghuol (Kill)
[160] ghuol (Kill)
[161] Go Slow Past the Drawers (Skipped)
[161] ghuol (Kill)
[162] Go Slow Past the Drawers (Skipped)
[162] ghuol (Kill)
[163] Go Slow Past the Drawers (Skipped)
[163] ghuol (Kill)
[164] ghuol (Kill)
[165] huge ghuol (Kill)

- 12 combats, 7 noncombats

- TOTAL: 37 combats, 26 noncombats

User avatar
QuantumNightmare
AFH
Posts: 1827
Joined: Mon Jan 29, 2007 10:28 pm
Location: Montreal

Post by QuantumNightmare » Tue May 13, 2008 12:32 pm

JiK4eva (private): That's what I had assumed. A random and probably irrelevant question, do you know anything about ghoststop?
JiK4eva (private): He's the player whose turncount is stuck at MAXINT. Since then, his ascensions have all been 0 turns. He ran away from a cyrpt mini-boss noncombat and never saw it again. I thought it was an interesting insight into the game mechanics, if a bit irrelevant

We need to recruit ghoststop to test out mechanics!

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Tue May 13, 2008 7:29 pm

QuantumNightmare wrote:JiK4eva (private): That's what I had assumed. A random and probably irrelevant question, do you know anything about ghoststop?
JiK4eva (private): He's the player whose turncount is stuck at MAXINT. Since then, his ascensions have all been 0 turns. He ran away from a cyrpt mini-boss noncombat and never saw it again. I thought it was an interesting insight into the game mechanics, if a bit irrelevant

We need to recruit ghoststop to test out mechanics!

Hehe, poor guy. When you run away from the Boss encounter I'm pretty sure it will only reappear after 4-6 turns, spent anywhere. If his turn count wasn't increasing, he wouldn't see it again.

This could actually be useful now that I think about it:
- Run just enough buffs to kill the regular monsters.
- Adventure until you find and skip the Boss encounter in each area.
- Spend 3 turns somewhere else.
- Start killing Bosses in the same order you found them, maybe drop Ur-kel's or use the Telescope buff or something.

That wouldn't cost you anything if the mechanic works how I think it does, but would allow you to deal with the whole zone more efficiently.

User avatar
QuantumNightmare
AFH
Posts: 1827
Joined: Mon Jan 29, 2007 10:28 pm
Location: Montreal

Post by QuantumNightmare » Tue May 13, 2008 10:43 pm

Seal Lubber wrote:Hehe, poor guy. When you run away from the Boss encounter I'm pretty sure it will only reappear after 4-6 turns, spent anywhere. If his turn count wasn't increasing, he wouldn't see it again.
Do you happen to have a log of that? Either way, I'd be curious to see a few hundred turns in a single subzone, skipping the boss every time. 200 turns with +combats and 200 turns with +noncombats would be great. I'll see if I have a multi available for this after rollover...

Seal Lubber
AFH
Posts: 318
Joined: Fri Sep 21, 2007 12:27 am

Post by Seal Lubber » Tue May 13, 2008 10:53 pm

QuantumNightmare wrote:Do you happen to have a log of that? Either way, I'd be curious to see a few hundred turns in a single subzone, skipping the boss every time. 200 turns with +combats and 200 turns with +noncombats would be great. I'll see if I have a multi available for this after rollover...
No logs, but I reckon I would of spent about 100 turns messing around with it. There's some not very useful information in the first and third posts here, but it would be worthwhile doing a fresh test.

Post Reply