After much puzzling and experimenting, I believe I figured out the turn order.
To track the sequence of turns, each dragon has a third hidden "bar", like health and breath. I will call this "initiative". A dragon or monster's starting "initiative" is equal to their quickness (QCK). This "initiative" is used to pay for a turn. The "initiative cost" of a turn is equal to the QCK of the bottom right monster (the one in the last position).
The turn order:
- Ambush rounds
- At the start of the fight, after the ambush rounds, each dragon and monster gets an amount of "initiative" equal to their quickness (QCK).
- Whoever has the most "initiative" gets a turn. In the case of a tie, the order is dragons before monsters, top to bottom. If there are four monsters, their tiebreak order is instead clockwise from the top right. Each turn costs an amount of "initiative" equal to the QCK of the monster in the bottom right.
- Repeat step (3) until no one has enough "initiative" to pay for another turn. Conveniently, since the cost of a turn is determined by the bottom right monster, this will happen after the bottom right monster's turn.
- Now that everyone is low on "initiative", each dragon and monster generates an amount of additional "initiative" equal to their QCK. Add this to their previous total.
- Head back to step (3) and keep going until the battle is over.
Haste:
- Haste takes effect right away. This matches the visual buff icon.
- Haste immediately increases your QCK by an amount equal to five plus the casters level.
- Haste does not change your "initiative" at all. Since the turn order is only indirectly determined by QCK, this means you will not always get an immediate effect from using Haste.
- The duration of Haste is listed as five rounds but in fact works more as six. This matches the visual buff icon.
The counting is a bit strange. What this means is that the buff wears off the sixth time you start a turn with it. Since QCK is only used after the monster in the final position acts and the "initiatives" tick up, and you end six turns with the buff, you in fact benefit from the buff for six turns instead of five. The number of rounds is effectively the same if a different dragon casts Haste, but the phrasing becomes a more nuanced.
Aside: The duration of Rally is counted the same way. However, since Rally is used for your attacks, this gives you five actions with the buff, so the duration of five makes sense here.
A Simulator:
PigeonPhD put together a wonderful site for predicting the turn order that can be found
here.
To be improved:
Unfortunately, this does not explain what happens when the bottom right monster dies. This was the monster that determined the "initiative cost" of each turn. When this monster goes away, the order jumbles, but I have not come up with a good explanation for the new order.
Acknowledgements:
Sylvandyr for feedback, logs, and another
guide focused on numbers.
Maki for logs that helped test in
this thread.
liramyne for details on Haste.
PigeonPhD for coding the Turn Order Calculator.
After much puzzling and experimenting, I believe I figured out the turn order.
To track the sequence of turns, each dragon has a third hidden "bar", like health and breath. I will call this "initiative". A dragon or monster's starting "initiative" is equal to their quickness (QCK). This "initiative" is used to pay for a turn. The "initiative cost" of a turn is equal to the QCK of the bottom right monster (the one in the last position).
The turn order:
- Ambush rounds
- At the start of the fight, after the ambush rounds, each dragon and monster gets an amount of "initiative" equal to their quickness (QCK).
- Whoever has the most "initiative" gets a turn. In the case of a tie, the order is dragons before monsters, top to bottom. If there are four monsters, their tiebreak order is instead clockwise from the top right. Each turn costs an amount of "initiative" equal to the QCK of the monster in the bottom right.
- Repeat step (3) until no one has enough "initiative" to pay for another turn. Conveniently, since the cost of a turn is determined by the bottom right monster, this will happen after the bottom right monster's turn.
- Now that everyone is low on "initiative", each dragon and monster generates an amount of additional "initiative" equal to their QCK. Add this to their previous total.
- Head back to step (3) and keep going until the battle is over.
Haste:
- Haste takes effect right away. This matches the visual buff icon.
- Haste immediately increases your QCK by an amount equal to five plus the casters level.
- Haste does not change your "initiative" at all. Since the turn order is only indirectly determined by QCK, this means you will not always get an immediate effect from using Haste.
- The duration of Haste is listed as five rounds but in fact works more as six. This matches the visual buff icon.
The counting is a bit strange. What this means is that the buff wears off the sixth time you start a turn with it. Since QCK is only used after the monster in the final position acts and the "initiatives" tick up, and you end six turns with the buff, you in fact benefit from the buff for six turns instead of five. The number of rounds is effectively the same if a different dragon casts Haste, but the phrasing becomes a more nuanced.
Aside: The duration of Rally is counted the same way. However, since Rally is used for your attacks, this gives you five actions with the buff, so the duration of five makes sense here.
A Simulator:
PigeonPhD put together a wonderful site for predicting the turn order that can be found
here.
To be improved:
Unfortunately, this does not explain what happens when the bottom right monster dies. This was the monster that determined the "initiative cost" of each turn. When this monster goes away, the order jumbles, but I have not come up with a good explanation for the new order.
Acknowledgements:
Sylvandyr for feedback, logs, and another
guide focused on numbers.
Maki for logs that helped test in
this thread.
liramyne for details on Haste.
PigeonPhD for coding the Turn Order Calculator.
If you find a battle that does not follow this sequence, I would love the details to revise this theory.
Ideas about what happens when the enemy in the bottom right goes away would also be greatly appreciated.
If you find a battle that does not follow this sequence, I would love the details to revise this theory.
Ideas about what happens when the enemy in the bottom right goes away would also be greatly appreciated.
@
RageAsylum One note about "top to bottom": In a party of 4, the default turn order is Top Right -> Bottom Right -> Bottom Left -> Top Left. This can be confirmed by the Crowned Bonepriest pack in Sandswept Delta if it still exists. This default order also also used to be observed in 1.0 in the all caster Mire pack for the Croakers + Shellion but not the Salve Kamaitachi (which had different QCK levels, but some sort of threshold was hit) - it may not happen in 2.0 for the Mire pack now that the first turn bug was fixed and the first turn is consumed properly.
Which packs made you believe that the QCK cost is the enemy in the last position? It's possible that it wasn't the "last position", but rather a specific QCK level or some other criteria, or perhaps a combination involving tiebreaker default turn order. (I suspect some sort of average because if certain QCK levels get removed from the battle, that causes the turn order to change in specific ways.) Maybe it'll be easier to figure out why they seem to center on the "last position" if we know which packs you noticed it for.
@
RageAsylum One note about "top to bottom": In a party of 4, the default turn order is Top Right -> Bottom Right -> Bottom Left -> Top Left. This can be confirmed by the Crowned Bonepriest pack in Sandswept Delta if it still exists. This default order also also used to be observed in 1.0 in the all caster Mire pack for the Croakers + Shellion but not the Salve Kamaitachi (which had different QCK levels, but some sort of threshold was hit) - it may not happen in 2.0 for the Mire pack now that the first turn bug was fixed and the first turn is consumed properly.
Which packs made you believe that the QCK cost is the enemy in the last position? It's possible that it wasn't the "last position", but rather a specific QCK level or some other criteria, or perhaps a combination involving tiebreaker default turn order. (I suspect some sort of average because if certain QCK levels get removed from the battle, that causes the turn order to change in specific ways.) Maybe it'll be easier to figure out why they seem to center on the "last position" if we know which packs you noticed it for.
@
Sylvandyr
It was for packs in the Ghostlight Ruins. When the fastest enemy was in the bottom right, it took the others an unusually long time to act. As I recall, it was a Tengu Caller, Longneck Scholar, Greybeak Reaper pack.
Thank you for the information about the packs of four. Do you know of a zone where there are four enemies of the same speed so I can see that myself?
@
Sylvandyr
It was for packs in the Ghostlight Ruins. When the fastest enemy was in the bottom right, it took the others an unusually long time to act. As I recall, it was a Tengu Caller, Longneck Scholar, Greybeak Reaper pack.
Thank you for the information about the packs of four. Do you know of a zone where there are four enemies of the same speed so I can see that myself?
@
RageAsylum Crowned Bonepriest x4 in Sandswept Delta, if that pack is still there. Do you know if it still exists?
[EDIT]: There used to be Gale Wolf x4 in Bamboo Falls too.
Sadly, these packs were pretty rare and annoying to run into, so if you're keeping track of an updated guide with existing packs, it'd be better to check first. (Sorry, I haven't been playing a lot lately.)
@
RageAsylum Crowned Bonepriest x4 in Sandswept Delta, if that pack is still there. Do you know if it still exists?
[EDIT]: There used to be Gale Wolf x4 in Bamboo Falls too.
Sadly, these packs were pretty rare and annoying to run into, so if you're keeping track of an updated guide with existing packs, it'd be better to check first. (Sorry, I haven't been playing a lot lately.)
@
Sylvandyr
Thank you for the turn information for four packs. I managed to chase down 4x Crowned Bonepriest I will update that.
I settled on the "last position" idea in part because it worked for all the test data I threw at it, and because it would be easy to code. Presumably, the enemies are stored in a list, and accessing index -1 (or length-1, depending on the language) would be simpler than some sort of average calculation.
@
Sylvandyr
Thank you for the turn information for four packs. I managed to chase down 4x Crowned Bonepriest I will update that.
I settled on the "last position" idea in part because it worked for all the test data I threw at it, and because it would be easy to code. Presumably, the enemies are stored in a list, and accessing index -1 (or length-1, depending on the language) would be simpler than some sort of average calculation.
@
RageAsylum I see. In that case, I suggest taking turn order data in a place of larger QCK variability like Harpy's Roost for most of the packs to see if your pattern holds true.
The fact that you mentioned not having an explanation for what happens when an enemy gets KOed makes me think something else is at work. I've also noticed KOing enemies in different slots with the same QCK has the same effect under certain conditions.
(I still have the data saved that you sent me, sorry that I haven't had the time to sit down and actually work on this.)
@
RageAsylum I see. In that case, I suggest taking turn order data in a place of larger QCK variability like Harpy's Roost for most of the packs to see if your pattern holds true.
The fact that you mentioned not having an explanation for what happens when an enemy gets KOed makes me think something else is at work. I've also noticed KOing enemies in different slots with the same QCK has the same effect under certain conditions.
(I still have the data saved that you sent me, sorry that I haven't had the time to sit down and actually work on this.)
@Sylvandyr
Oh absolutely. I intend to continue testing this out. So far, it holds. For an example, here is the Psywurm - Salve Kamaitachi - Psywurm pack you found odd. The numbers are what their "initiative" values would be tracing through the turns. The names on the right side are the order. It seems to hold up, but of course I have not tested, nor will I ever test, every single QCK combination.
[code]
Psywurm 1 Psywurm 2
Sylvanshal Salve
ambush Sylvanshal haste QCK 64 -> 94
ambush Sylvanshal
94 62 56 62 Sylvanshal
32 62 56 62 Psywurm 1
32 0 56 62 Psywurm 2
126 62 112 62 Sylvanshal
64 62 112 62 Salve
64 62 50 62 Sylvanshal KO Salve
2 62 62 Psywurm 1
2 0 62 Psywurm 2
96 62 62 Sylvanshal
34 62 62 Psywurm 1
34 0 62 Psywurm 2
128 62 62 Sylvanshal KO Pswurm 1, lose haste QCK 94 -> 64
66 62 Sylvanshal
4 62 Psywurm 2
98 62 Sylvanshal KO Pswurm 2
[/code]
@
Sylvandyr
Oh absolutely. I intend to continue testing this out. So far, it holds. For an example, here is the Psywurm - Salve Kamaitachi - Psywurm pack you found odd. The numbers are what their "initiative" values would be tracing through the turns. The names on the right side are the order. It seems to hold up, but of course I have not tested, nor will I ever test, every single QCK combination.
Code:
Psywurm 1 Psywurm 2
Sylvanshal Salve
ambush Sylvanshal haste QCK 64 -> 94
ambush Sylvanshal
94 62 56 62 Sylvanshal
32 62 56 62 Psywurm 1
32 0 56 62 Psywurm 2
126 62 112 62 Sylvanshal
64 62 112 62 Salve
64 62 50 62 Sylvanshal KO Salve
2 62 62 Psywurm 1
2 0 62 Psywurm 2
96 62 62 Sylvanshal
34 62 62 Psywurm 1
34 0 62 Psywurm 2
128 62 62 Sylvanshal KO Pswurm 1, lose haste QCK 94 -> 64
66 62 Sylvanshal
4 62 Psywurm 2
98 62 Sylvanshal KO Pswurm 2
@
RageAsylum
I've been testing out a new mage build for the Golem Workshop that will eventually have ~110 QCK (97 at level 21, currently), and I've learned that, in terms of this formula, the initiative added from using Haste does not apply until the set turn order of a round has finished. That is to say, if I use Haste but the last monster hasn't attacked yet (it's the middle of the round), I don't get an immediate extra turn. However, if I use Haste before the round has started - in the Ambush turns - I do get that second turn before the monsters attack. Haste's added QCK (its bonus to initiative) needs to be there before turn order is calculated, or it won't be applied until the next round, when the Hasted QCK is added to any remaining initiative.
If you need me to delete this, let me know! I just thought it might be useful.
@
RageAsylum
I've been testing out a new mage build for the Golem Workshop that will eventually have ~110 QCK (97 at level 21, currently), and I've learned that, in terms of this formula, the initiative added from using Haste does not apply until the set turn order of a round has finished. That is to say, if I use Haste but the last monster hasn't attacked yet (it's the middle of the round), I don't get an immediate extra turn. However, if I use Haste before the round has started - in the Ambush turns - I do get that second turn before the monsters attack. Haste's added QCK (its bonus to initiative) needs to be there before turn order is calculated, or it won't be applied until the next round, when the Hasted QCK is added to any remaining initiative.
If you need me to delete this, let me know! I just thought it might be useful.
|
|
|
Lira
she/her
|
@
liramyne
Oh no, that is very useful. I will try to work that in there more clearly. Perhaps the cleanest way would be to add a section about haste. It seems that since haste only adds to your QCK, and QCK is only used when adding to what I called your "initiative", as you also found, it does not always take effect right away.
@
liramyne
Oh no, that is very useful. I will try to work that in there more clearly. Perhaps the cleanest way would be to add a section about haste. It seems that since haste only adds to your QCK, and QCK is only used when adding to what I called your "initiative", as you also found, it does not always take effect right away.
