Arch Dynamis lords difficultly.

[Yinnyth]

Ah my bad but you must admit they are killing awefully slow to be any relevant, there is probably a minimum% where it pops, this doesn't disprove the 30 sec I'm seeing on all my runs. All I care about anyway is the minimum% were I can bring it. What's certain is that you can fight it 30 sec with high killing speed (unlike them). Seeing as I'm only at 75% of my potential, I'm sure I can bring it at 25% with 6 if it is possible and not block by a force clone. I've never seen it pop before 30sec for me also I've seen it clone above 50% HP so HP% trigger is out of question. I'm hoping you can bring it much lower than 45% which is what I do, and I want to know what HP% it is. I'm still convinced 30 sec is in the equation.

You realize that normal enemies in Dynamis have an HP trigger for their 2hrs, right? Their HP trigger is a different % from job to job, and even then it seems like it can vary up to 10% from enemy to enemy even if they're the same job. DRG is somewhere around 95% which is why you almost never see one who doesn't summon a wyvern. NIN is somewhere around 30% which is why you frequently kill (or proc) them before they mijin. THF and PLD are somewhere around 50%.

Once they get below a certain HP, they put 2hring into their "to do" que, but their 2hr is a lower priority than WS, ranged attack, or casting a spell. That's why sometimes, even if you're killing very slow and don't proc, it's possible to prevent a NIN's 2hr by just placing stuns at the right times. Even though Mijin Gakure is on his "to do" que, it keeps getting pushed back on the que because his ranged attack is ready again, or he's ready to WS, or he's ready to cast a spell instead.

Why is any of that important to know for a discussion about ADL? Because ADL's cloning works the same way. Once you get him below a certain %, he puts splitting onto his "to do" que. If his magic or his WS are ready, he waits to split and does a spell or WS instead. His % appears to be somewhere around 60-66%. That's when he puts it on his que. Just because there isn't an exact % that he ALWAYS splits on doesn't mean it isn't % based. If you had some way to keep spamming him with stuns that last 10 seconds each (which is mostly impossible, I know), he would never have the chance to clone because everytime stun wore off, he'd attempt a WS or spell instead.

So there you go. That's why he can sometimes clone after 20 seconds and sometimes clone after 2.5 minutes. In order to disprove my theory, you would need to get a video of him cloning while he has more than 70% HP.

[MarkovChain]

You realize that normal enemies in Dynamis have an HP trigger for their 2hrs, right? Their HP trigger is a different % from job to job, and even then it seems like it can vary up to 10% from enemy to enemy even if they're the same job. DRG is somewhere around 95% which is why you almost never see one who doesn't summon a wyvern. NIN is somewhere around 30% which is why you frequently kill (or proc) them before they mijin. THF and PLD are somewhere around 50%.

Once they get below a certain HP, they put 2hring into their "to do" que, but their 2hr is a lower priority than WS, ranged attack, or casting a spell. That's why sometimes, even if you're killing very slow and don't proc, it's possible to prevent a NIN's 2hr by just placing stuns at the right times. Even though Mijin Gakure is on his "to do" que, it keeps getting pushed back on the que because his ranged attack is ready again, or he's ready to WS, or he's ready to cast a spell instead.

Why is any of that important to know for a discussion about ADL? Because ADL's cloning works the same way. Once you get him below a certain %, he puts splitting onto his "to do" que. If his magic or his WS are ready, he waits to split and does a spell or WS instead. His % appears to be somewhere around 60-66%. That's when he puts it on his que. Just because there isn't an exact % that he ALWAYS splits on doesn't mean it isn't % based. If you had some way to keep spamming him with stuns that last 10 seconds each (which is mostly impossible, I know), he would never have the chance to clone because everytime stun wore off, he'd attempt a WS or spell instead.

So there you go. That's why he can sometimes clone after 20 seconds and sometimes clone after 2.5 minutes. In order to disprove my theory, you would need to get a video of him cloning while he has more than 70% HP.

There is nothing to backup your theory. Comparing demon's2H and his cloning is farfetched. It's not as random as you think. I bet that many groups can confirm that with a given setup and strat it is going to pop at roughly the same HP%.

Ever since I do maximise my damage by replacing the whm by a monk etc it's systematically clones at lower HP. The times I had it clone above 50% that I mentionned earlier was because I was killing to slowly, and it was still at around 60% after 30s.

@ Macross, was

[Yinnyth]

There is nothing to backup your theory. Comparing demon's2H and his cloning is farfetched. It's not as random as you think. I bet that many groups can confirm that with a given setup and strat it is going to pop at roughly the same HP%.

No, it's not farfetched, it's the most logical conclusion to draw from the data available. Let us look at the evidence so far:
1. In some fights, it takes ADL as few as 20 seconds to split, generally while he is damaged extremely quickly.
2. In other fights, it takes ADL as long as 150 seconds to split, generally while he is damaged very slowly.
3. No one has yet to offer proof of an ADL kill in which he has not split at all.

So let's assume his behavior for splitting is based ONLY on how much time has passed since he was popped. Since there have been fights where he's split in 20 seconds, and fights where he's split in 150 seconds, we must then go on to assume that the time required for him to split is RANDOMIZED between those values. However, this would also mean that a group which deals damage to him quickly would, sooner or later, wind up with an ADL who takes 2 and a half minutes to split. This would mean they could kill him before he split. Since there is no evidence of this EVER happening, and not even any claims that it has happened as far as I can see, we can then conclude that his splitting is not determined solely by time.

So let's go further and assume that his splitting is based on 2 seperate triggers: HP and time- whichever elapses first. Since there is proof of a fight where he's taken longer than 2.5 minutes to split, and 2.5 minutes is more than enough time for any decent group to get him to his trigger HP, we can further conclude that IT DOESN'T MATTER if there's any time trigger at all.

Which leaves us with one possibility: his split is based partly, if not completely, on how low his HP is. So why does he sometimes get brought down to 30% HP before he manages to split, while other times he only gets brought down to 60%? We can answer that by looking at the behavior of other enemies who also have actions they take which are triggered by their HP%. He splits at varying HPs because his other actions take priority over his split, and the HP trigger itself might be randomized by as much as 10% either direction. Regardless, an HP trigger is the most logical conclusion to come to.