Register

Simulationcraft APL Discussion

Face-rippin fun.

Moderator: Forum Administrators

Exalted
User avatar
Posts: 2278
Joined: Fri Nov 25, 2011 7:49 pm

Simulationcraft APL Discussion

Postby aggixx » Sat Oct 11, 2014 7:38 pm

This thread is for discussion and improvement of the default action priority list in the latest revision of WoD SimulationCraft. The most recent list can generally be found here, although keep in mind that some elements of the action list are dynamic based on the profile, eg. using Incarnation differently because the profile has Assurance of Consequence.
ImageImage

Exalted
Posts: 564
Joined: Tue Feb 08, 2011 5:51 pm

Re: WoD - Simulationcraft APL Discussion

Postby ShmooDude » Sat Oct 11, 2014 7:40 pm

Gonna try some Simc optimizing tonight since things seem fairly stable balance wise. Some of the current logic I think is out of date due to the balance changes we've received. What I've found so far:

1) Thrash is far too low prioirty when dealing with multiple targets. Its actually a DPS up to always thrash where the clarity proc is instead of prioritizing finishers/rake.
2) Not always pooling energy for FB. This should always be done unless Rip is actually going to fall off.

The other thing I'm going to mess with is trying to build an action list that uses buff checks instead of (exclusively) a comparison to the old DoT. This will allow us to do things such as waiting to apply rake towards the end of a Tiger's Fury if it won't waste ticks then, but will now. That kind of thing, may not pan out but I think its worth a shot.

The one question I have is with the cycle targets. Right now, the action list is trying to game Rake/Rip DoTs on multiple targets, and quite frankly, I think that's something only a computer can do (maaaybe a player on 2 targets, but certainly not more than that). Should it get reduced to only trying to maintain bleeds on non-current targets?

Exalted
User avatar
Posts: 2278
Joined: Fri Nov 25, 2011 7:49 pm

Re: WoD - Simulationcraft APL Discussion

Postby aggixx » Sat Oct 11, 2014 7:41 pm

For execute a skilled player should be making sure their Rip multipliers are as high as possible on all targets. Rip in general isn't that unreasonable if you use an addon to track multipliers for you.

On Rake though I can agree. For the most part realistically good play means reapplying as many Rakes as possible during Tiger's Fury, but that's more buff tracking than multiplier tracking (like you said), and tracking Bloodtalons is a whole different ballgame. Although I could see tracking Bloodtalons'd Rakes on 2 or 3 targets like you said, but beyond that its just unreasonable for a player to waste time thinking about that sort of thing because of the scarcity of Bloodtalons procs they actually have to utilize. Its not impossible to do, just stupid because of the amount of brain power it would take to manage something completely negligible.
ImageImage

Exalted
Posts: 564
Joined: Tue Feb 08, 2011 5:51 pm

Re: WoD - Simulationcraft APL Discussion

Postby ShmooDude » Sat Oct 11, 2014 7:42 pm

I'm gonna try some stuff with both multipliers and buff tracking.

These looked correct for the few things I've tried so far (all assume Savage Roar is up):

1.45 Base
1.6675 TF
1.885 BT
2.16775 TF + BT
2.9 Inc (and Prowl/SM)
3.335 Inc + TF
3.77 Inc + BT
4.3355 Inc + TF + BT

Multipliers have the advantage of being able to catch more than one thing with the same line. For example I tried a line that said something like if persistent_multiplier>=dot.rake.pmultiplier&persistent_multiplier>3 catches anytime a rake has incarnation + at least one other buff. Though I suppose the same thing could be done like persistent_multiplier>=dot.rake.pmultiplier&buff.king_of_the_jungle.up&(buff.tigers_fury.up|buff.bloodtalons.up). Might just go ahead and switch to that as its probably slightly easier to read.

Exalted
User avatar
Posts: 2278
Joined: Fri Nov 25, 2011 7:49 pm

Re: WoD - Simulationcraft APL Discussion

Postby aggixx » Sat Oct 11, 2014 7:42 pm

Yeah I would use buffs wherever possible. Generally you should avoid looking at the absolute persistent multiplier value in the APL because something else could be added that invalidates the whole idea of the list.
ImageImage

Exalted
User avatar
Posts: 2278
Joined: Fri Nov 25, 2011 7:49 pm

Re: Simulationcraft APL Discussion

Postby aggixx » Sat Oct 11, 2014 7:50 pm

I made a change to how TF and Berserk are used earlier today. It turns out that even after the removal of the restriction on TF being used during Berserk it is still beneficial to synchronize the usage of the two cooldowns as we've been doing live. This was fairly significantly (1%) DPS gain at level 90 due to Assurance of Consequence majorly messing with the way it was working before, and a small gain at level 100 (0.5% or something).
ImageImage

Exalted
Posts: 564
Joined: Tue Feb 08, 2011 5:51 pm

Re: Simulationcraft APL Discussion

Postby ShmooDude » Sat Oct 11, 2014 7:57 pm

One thing I've found so far is that gaming rake is a fairly small gain with the current APL. I think even at its highest (BT+Inc) it was only ~2% over simply refreshing asap for maximum uptime. Even if we find a better way to do it, I don't see it being a particularly large gain (certainly not more than 5%). Something that people that are good at the class should do, but isn't going to kill someone who is unable to/do not want to as long as they maintain their uptimes.

Haven't messed with Rip, but its probably a bigger gain since we can keep the snapshot with an execute FB refresh (though didn't I read somewhere that was bugged last I saw, idk if its been fixed yet) and its duration is significantly longer.

Exalted
Posts: 564
Joined: Tue Feb 08, 2011 5:51 pm

Re: Simulationcraft APL Discussion

Postby ShmooDude » Sat Oct 11, 2014 11:20 pm

How did you change berserk specifically? Did you simply change the if to a sync=tigers_fury? If so, I'm only seeing a DPS up in certain specs doing that. Some (like incarnation ones) are seeing a loss.

Exalted
Posts: 564
Joined: Tue Feb 08, 2011 5:51 pm

Re: Simulationcraft APL Discussion

Postby ShmooDude » Sat Oct 11, 2014 11:36 pm

Changes I've found beneficial so far:

1) Fixing sync issues. Spells/Items you want to sync must be placed above the spell they're syncing to or the ability never fires. The potion synced to berserk was too low on the action list to ever fire, moved it up and got a small gain.
2) Ensuring Ferocious Bite always uses max energy. Not sure I even have it the best way atm, gonna look at the old FB code and see if that works better than what I got now.
3) Increase Thrash priority significantly. Currently even at 2 targets its a DPS up to always thrash where the current clearcast Thrash is. Haven't tested it sans 2-piece though, so it might require 3 targets without the 2 piece.1

What I have not particularly found yet, interestingly, is any notably better bleed overwrite logic. I have a feeling that the gains at this point are so low, that optimizing it doesn't really do much. Either that, or I simply haven't found the proper way to do it yet.

Exactly what I have ATM:

Spoiler: show
Code: Select all
actions=rake,if=buff.prowl.up|buff.shadowmeld.up
actions+=/auto_attack
actions+=/skull_bash
actions+=/force_of_nature,if=charges=3|trinket.proc.all.react|target.time_to_die<20
actions+=/incarnation,sync=berserk
actions+=/potion,name=draenic_agility,sync=berserk,if=target.health.pct<25
actions+=/berserk,if=cooldown.tigers_fury.remains<8
actions+=/potion,name=draenic_agility,if=target.time_to_die<=40
actions+=/use_item,slot=trinket1,sync=tigers_fury
actions+=/blood_fury,sync=tigers_fury
actions+=/berserking,sync=tigers_fury
actions+=/arcane_torrent,sync=tigers_fury
actions+=/tigers_fury,if=(!buff.omen_of_clarity.react&energy.max-energy>=60)|energy.max-energy>=80
actions+=/shadowmeld,if=(buff.bloodtalons.up|!talent.bloodtalons.enabled)&dot.rake.remains<0.3*dot.rake.duration
# Keep Rip from falling off during execute range.
actions+=/ferocious_bite,cycle_targets=1,if=dot.rip.ticking&dot.rip.remains<=3&target.health.pct<25
actions+=/healing_touch,if=talent.bloodtalons.enabled&buff.predatory_swiftness.up&(combo_points>=4|buff.predatory_swiftness.remains<1.5)
actions+=/savage_roar,if=buff.savage_roar.remains<3
actions+=/pool_resource,for_next=1
actions+=/thrash_cat,if=remains<=duration*0.3&active_enemies>1
actions+=/rip,cycle_targets=1,if=combo_points=5&remains<=duration*0.3&persistent_multiplier>dot.rip.pmultiplier
actions+=/ferocious_bite,cycle_targets=1,if=combo_points=5&target.health.pct<25&dot.rip.ticking&(energy>=50|buff.berserk.up&energy>=25)
actions+=/rip,cycle_targets=1,if=combo_points=5&remains<=3
actions+=/savage_roar,if=combo_points=5&(energy.time_to_max<=1|buff.berserk.up|cooldown.tigers_fury.remains<3)&buff.savage_roar.remains<42*0.3
actions+=/ferocious_bite,if=combo_points=5&(energy.time_to_max<=1|buff.berserk.up&energy>=25|cooldown.tigers_fury.remains<3&energy>=50)
actions+=/rake,cycle_targets=1,if=remains<=3&combo_points<5
actions+=/rake,cycle_targets=1,if=remains<=duration*0.3&combo_points<5&persistent_multiplier>dot.rake.pmultiplier
actions+=/moonfire,cycle_targets=1,if=combo_points<5&remains<=duration*0.3&active_enemies<=10
actions+=/rake,cycle_targets=1,if=persistent_multiplier>dot.rake.pmultiplier&combo_points<5
actions+=/swipe,if=combo_points<5&active_enemies>=3
actions+=/shred,if=combo_points<5&active_enemies<3


Note, I need to double check that Thrash change on the current code, haven't messed with it for a while.

Exalted
User avatar
Posts: 2091
Joined: Mon May 24, 2010 10:21 pm

Re: Simulationcraft APL Discussion

Postby Tinderhoof » Sun Oct 12, 2014 2:21 am

Just as a note. As we aren't going to even see 2p until we are in the middle of January making any action list favor it right now is way to early. I would focus on what we are going to actually have available for the first 3 months.

Exalted
User avatar
Posts: 2278
Joined: Fri Nov 25, 2011 7:49 pm

Re: Simulationcraft APL Discussion

Postby aggixx » Sun Oct 12, 2014 4:11 am

ShmooDude wrote:How did you change berserk specifically? Did you simply change the if to a sync=tigers_fury? If so, I'm only seeing a DPS up in certain specs doing that. Some (like incarnation ones) are seeing a loss.

Berserk is now used any time Tiger's Fury is active (so, immediately after activating TF in all but edge cases) and Incarnation is synced to Berserk. If you have Assurance it instead gets synced to TF.

ShmooDude wrote:1) Fixing sync issues. Spells/Items you want to sync must be placed above the spell they're syncing to or the ability never fires. The potion synced to berserk was too low on the action list to ever fire, moved it up and got a small gain.
2) Ensuring Ferocious Bite always uses max energy. Not sure I even have it the best way atm, gonna look at the old FB code and see if that works better than what I got now.
3) Increase Thrash priority significantly. Currently even at 2 targets its a DPS up to always thrash where the current clearcast Thrash is. Haven't tested it sans 2-piece though, so it might require 3 targets without the 2 piece.

1) Nice catch. That was probably my fault that it got like that but its an easy thing to miss.

2) Yeah I think originally I had messed with 50 energy FBs but the damage was just so bad at the time that the gain was negligible (vs using the energy on more shreds) so it wasn't worth the effort. I can certainly see it being worth it now of course after the series of buffs (and nerfs) to it. I'm skeptical the way your doing it is ideal for sure, especially the BitW line:
Code: Select all
actions+=/ferocious_bite,cycle_targets=1,if=combo_points=5&target.health.pct<25&dot.rip.ticking&(energy>=50|buff.berserk.up&energy>=25)

There's no real reason to FB at 50 when you could wait til higher instead, which reduces SR collision and increases Bloodtalons success rate. Also I think adding an energy conditional on the Berserk half is probably a bit excessive since you're not likely to even get down to 25 energy in the Berserk anyway, at least at most gear levels. I'll give it a shot and see what I can do.

3) Pretty sure its always a DPS up at 2. Its basically break-even at 1 target regardless of how little mastery you have.

ShmooDude wrote:What I have not particularly found yet, interestingly, is any notably better bleed overwrite logic. I have a feeling that the gains at this point are so low, that optimizing it doesn't really do much. Either that, or I simply haven't found the proper way to do it yet.

Yeah in general I've found bleed optimizations that involve something other than uptime are very difficult to get any significant gain out of. Gaming BitW is the biggest by far and anything else is pretty much in the neighborhood of a couple % at most.

Tinderhoof wrote:Just as a note. As we aren't going to even see 2p until we are in the middle of January making any action list favor it right now is way to early. I would focus on what we are going to actually have available for the first 3 months.

While its total impact is quite significant its impact on a single action selection so insignificant that I'm very skeptical it makes any practical difference in the APL. At most it could cause energy flooding issues at a high number of targets but that's not really anything you'd typically sim anyway.
ImageImage

Exalted
User avatar
Posts: 2278
Joined: Fri Nov 25, 2011 7:49 pm

Re: Simulationcraft APL Discussion

Postby aggixx » Sun Oct 12, 2014 5:12 am

Found that not having an energy requirement on the Shadowmeld line was causing it to waste a good portion of its Shadowmeld cooldowns, especially because the sim doesn't halt autoattacks when it Shadowmelds so the stealth breaks immediately if it doesn't cast an action immediately after s'melding. Incidentally I also made Shadowmeld cancel your current autoattack swing, so the DPS change is negligible (I didn't test this but I figure it has to work this way in-game or it would be rather ineffective, let me know if I'm wrong). The conditional was also missing an equals sign which means it eval'd to false when there was no rake on the target, but that doesn't really have an impact on anything.
ImageImage

Exalted
Posts: 564
Joined: Tue Feb 08, 2011 5:51 pm

Re: Simulationcraft APL Discussion

Postby ShmooDude » Sun Oct 12, 2014 5:13 am

aggixx wrote:
ShmooDude wrote:How did you change berserk specifically? Did you simply change the if to a sync=tigers_fury? If so, I'm only seeing a DPS up in certain specs doing that. Some (like incarnation ones) are seeing a loss.

Berserk is now used any time Tiger's Fury is active (so, immediately after activating TF in all but edge cases) and Incarnation is synced to Berserk. If you have Assurance it instead gets synced to TF.


I'm still seeing that (at level 100) as a loss for BT & CoS Incarnation. Its a small gain or neutral for everything else though. However, if I also desync Incarnation from Berserk and simply add a if=cooldown.berserk.remains<14 instead, it becomes a DPS up for incarnation as well. This is likely a bit of a quirk on fight duration but shows that maintaining the highest incarnation uptime is important. The question then becomes, which is better if the two become desynced too far? This is something that might happen in raiding where you won't have 100% target uptime, but never in a patchwerk sim. For that matter, the same could be said of the Berserk change. It might turn out that the old berserk method is better when in non-optimal conditions. We should probably create a fight with some downtime (preferably similar to some real T17 fight) to test these cases.

aggixx wrote:
Tinderhoof wrote:Just as a note. As we aren't going to even see 2p until we are in the middle of January making any action list favor it right now is way to early. I would focus on what we are going to actually have available for the first 3 months.

While its total impact is quite significant its impact on a single action selection so insignificant that I'm very skeptical it makes any practical difference in the APL. It may cause an energy flooding issue on a very high number of targets, but its not going to shift any DPE breakpoints.


Yeah, I doubt it'll affect the APL. However, he has a point in that we'll literally have 100% different gear until BF opens. I might build a highmaul only gear set to mess around with.

Exalted
User avatar
Posts: 2278
Joined: Fri Nov 25, 2011 7:49 pm

Re: Simulationcraft APL Discussion

Postby aggixx » Sun Oct 12, 2014 5:24 am

aggixx wrote:There's no real reason to FB at 50 when you could wait til higher instead, which reduces SR collision and increases Bloodtalons success rate. Also I think adding an energy conditional on the Berserk half is probably a bit excessive since you're not likely to even get down to 25 energy in the Berserk anyway, at least at most gear levels. I'll give it a shot and see what I can do.

I'm having trouble proving this in practice and I'm not sure why. Both the average damage per execute and average energy spent go down when I change the conditional like this:
Code: Select all
actions+=/ferocious_bite,cycle_targets=1,if=combo_points=5&target.health.pct<25&dot.rip.ticking&(energy.time_to_max<=1|buff.berserk.up)

The SR uptime did go from 99.9% uptime to "constant" (which isn't exactly 100%, not sure what the exact cutoff is) but that may have been coincidence since we're talking such a small change.
ImageImage

Exalted
User avatar
Posts: 2278
Joined: Fri Nov 25, 2011 7:49 pm

Re: Simulationcraft APL Discussion

Postby aggixx » Sun Oct 12, 2014 5:28 am

ShmooDude wrote:I'm still seeing that (at level 100) as a loss for BT & CoS Incarnation. Its a small gain or neutral for everything else though.

Hmm, I wonder why that would be. I don't see any obvious correlation with how the result would change from Inc-LI to Inc-Anything, unless you're just not simming LI, that's the only talent combo I tested it with at 100 before I made the change.

Anyways I need some sleep, I'll work on committing the changes so far when I get up.
ImageImage

Exalted
Posts: 564
Joined: Tue Feb 08, 2011 5:51 pm

Re: Simulationcraft APL Discussion

Postby ShmooDude » Sun Oct 12, 2014 6:10 am

aggixx wrote:
aggixx wrote:There's no real reason to FB at 50 when you could wait til higher instead, which reduces SR collision and increases Bloodtalons success rate. Also I think adding an energy conditional on the Berserk half is probably a bit excessive since you're not likely to even get down to 25 energy in the Berserk anyway, at least at most gear levels. I'll give it a shot and see what I can do.

I'm having trouble proving this in practice and I'm not sure why. Both the average damage per execute and average energy spent go down when I change the conditional like this:
Code: Select all
actions+=/ferocious_bite,cycle_targets=1,if=combo_points=5&target.health.pct<25&dot.rip.ticking&(energy.time_to_max<=1|buff.berserk.up)

The SR uptime did go from 99.9% uptime to "constant" (which isn't exactly 100%, not sure what the exact cutoff is) but that may have been coincidence since we're talking such a small change.


Two things:
1) You're actually getting less total Ferocious Bites because while waiting for that energy to cap, sometimes you end up casting Savage Roar instead. This on average appears to result in less total Ferocious Bite casts (0.4 less).
2) You're no longer guaranteed to use full energy Ferocious Bites in Berserk. Without Bloodlust up its actually quite easy to drain all your energy during a Berserk so any FB on the tailing end will be sub-optimal. It probably doesn't show quite as much with BT builds as that can delay the FB by a GCD (if your CP builder at 3 cps crits you to 5).


aggixx wrote:
ShmooDude wrote:I'm still seeing that (at level 100) as a loss for BT & CoS Incarnation. Its a small gain or neutral for everything else though.

Hmm, I wonder why that would be. I don't see any obvious correlation with how the result would change from Inc-LI to Inc-Anything, unless you're just not simming LI, that's the only talent combo I tested it with at 100 before I made the change.

Anyways I need some sleep, I'll work on committing the changes so far when I get up.


I sim all 9 at once (very slow, but makes sure any changes work for everything). I see a loss for Inc-LI (though actually not quite as much as the other two iirc).


One new thing I found as well is that more aggressively using 5 CP finishers with SotF is actually a DPS up (ie changing the non-execute FB and 5 cp Savage Roar from energy.time_to_max<=1 to energy>=50). Not a huge gain, only ~80-100 DPS but noticeable (dps loss with other two talents though).

Exalted
User avatar
Posts: 588
Joined: Wed Jun 29, 2011 4:49 am

Re: Simulationcraft APL Discussion

Postby Stenhaldi » Sun Oct 12, 2014 9:02 am

Intuitively, TF before berserk (instead of during berserk) makes sense if TF during berserk would waste energy -- a condition that requires a certain amount of haste, bloodlust, T17 2-piece, bloodtalons, soul of the forest, or moderate clearcast luck. In a patchwerk fight, it may still make sense to always cast TF before berserk just because you want to do that for the first cast (since bloodlust is up) and you don't want to delay cooldowns later. But in general (at least without bloodtalons/sotf/2T17/haste), if TF ever gets delayed it's likely good to keep it behind berserk. Might not be worth entering this kind of thing into the APL, but something to consider when you're wondering why certain APL changes have the effect they do.

Also, an alternative to casting TF before berserk is to just berserk below full energy. This could be better than "TF before berserk" because it allows you to subtract an arbitrary amount of energy from berserk, rather than just 0 or 60.

Exalted
User avatar
Posts: 2278
Joined: Fri Nov 25, 2011 7:49 pm

Re: Simulationcraft APL Discussion

Postby aggixx » Sun Oct 12, 2014 12:44 pm

ShmooDude wrote:Two things:
1) You're actually getting less total Ferocious Bites because while waiting for that energy to cap, sometimes you end up casting Savage Roar instead. This on average appears to result in less total Ferocious Bite casts (0.4 less).
2) You're no longer guaranteed to use full energy Ferocious Bites in Berserk. Without Bloodlust up its actually quite easy to drain all your energy during a Berserk so any FB on the tailing end will be sub-optimal. It probably doesn't show quite as much with BT builds as that can delay the FB by a GCD (if your CP builder at 3 cps crits you to 5).

I did it with a Berserk conditional as well and it wasn't significantly different, and that line never happens during Bloodlust because Bloodlust always happens at the start of the sim.

And my impression was that the "lost" ferocious bites should be zero-sum: the sooner you Savage Roar the later you have to do it again, which means more time for Ferocious Biting. I suppose its possible that it refreshes SR to be longer than the amount of time the target will be alive for and loses a fraction of a bite, but otherwise it shouldn't make a difference.

Edit: Nope, its not that.

Stenhaldi wrote:Also, an alternative to casting TF before berserk is to just berserk below full energy. This could be better than "TF before berserk" because it allows you to subtract an arbitrary amount of energy from berserk, rather than just 0 or 60.

This was essentially the behavior before, it was casting it "on cooldown" but there is a line for Improved Rake above it so it would always open with that, and then past that point its generally not going to be at full energy again.
ImageImage

Exalted
User avatar
Posts: 2278
Joined: Fri Nov 25, 2011 7:49 pm

Re: Simulationcraft APL Discussion

Postby aggixx » Sun Oct 12, 2014 12:56 pm

It occurred to me that I could add an option to Ferocious Bite that would restrict it to being used with max energy, this might clean up the action list a bit while still being fairly comprehendable. Something like this:
Code: Select all
actions+=/ferocious_bite,max_energy=1,if=combo_points=5&target.health.pct<25&dot.rip.ticking

and of course the option would be line-specific so you could also have something like
Code: Select all
actions+=/ferocious_bite,if=combo_points=5&cooldown.tigers_fury.remains<3

if you wanted to (now of course that may or may not be optimal, just an example).
ImageImage

Exalted
Posts: 564
Joined: Tue Feb 08, 2011 5:51 pm

Re: Simulationcraft APL Discussion

Postby ShmooDude » Sun Oct 12, 2014 1:09 pm

Yeah, that'd be nice.

Currently there is a actions+=/pool_energy,for_next=1,extra_amount=25 as well, but that doesn't quite work since its different when berserk is up.

I'd pick something other than max_energy as that actually seems to already be used? (I think it meas the same thing as energy.max)

Exalted
User avatar
Posts: 2278
Joined: Fri Nov 25, 2011 7:49 pm

Re: Simulationcraft APL Discussion

Postby aggixx » Sun Oct 12, 2014 1:30 pm

Options and expressions aren't the same thing, I can basically name it whatever I want. If you think there's something clearer I'm all ears but I think "require_max_energy" is a bit longer than is ideal.
ImageImage

Exalted
Posts: 564
Joined: Tue Feb 08, 2011 5:51 pm

Re: Simulationcraft APL Discussion

Postby ShmooDude » Sun Oct 12, 2014 1:55 pm

aggixx wrote:Options and expressions aren't the same thing, I can basically name it whatever I want. If you think there's something clearer I'm all ears but I think "require_max_energy" is a bit longer than is ideal.


Ah, then yeah, use_max_energy or just max_energy would be fine.

The other thing I was thinking of doing was to split up the action list.

Currently (if you take out the thrash and move it up higher where it needs to be) the bottom 6 are all combo_points<5 and then the next 5 above those are all combo_points=5. I was thinking something like this:

Spoiler: show
Code: Select all
actions=rake,if=buff.prowl.up|buff.shadowmeld.up
actions+=/auto_attack
actions+=/skull_bash
actions+=/force_of_nature,if=charges=3|trinket.proc.all.react|target.time_to_die<20
actions+=/incarnation,if=cooldown.berserk.remains<14
actions+=/potion,name=draenic_agility,sync=berserk,if=target.health.pct<25
actions+=/use_item,slot=trinket1,sync=tigers_fury
actions+=/berserk,if=buff.tigers_fury.up
actions+=/potion,name=draenic_agility,if=target.time_to_die<=40
actions+=/blood_fury,sync=tigers_fury
actions+=/berserking,sync=tigers_fury
actions+=/arcane_torrent,sync=tigers_fury
actions+=/tigers_fury,if=(!buff.omen_of_clarity.react&energy.max-energy>=60)|energy.max-energy>=80
actions+=/shadowmeld,if=(buff.bloodtalons.up|!talent.bloodtalons.enabled)&dot.rake.remains<0.3*dot.rake.duration
# Keep Rip from falling off during execute range.
actions+=/ferocious_bite,cycle_targets=1,if=dot.rip.ticking&dot.rip.remains<=3&target.health.pct<25
actions+=/healing_touch,if=talent.bloodtalons.enabled&buff.predatory_swiftness.up&(combo_points>=4|buff.predatory_swiftness.remains<1.5)
actions+=/savage_roar,if=buff.savage_roar.remains<3
actions+=/pool_resource,for_next=1
actions+=/thrash_cat,if=remains<=duration*0.3&active_enemies>1
actions+=/run_action_list,name=finishers,if=combo_points=5
actions+=/run_action_list,name=generators,if=combo_points<5

actions.finishers+=/ferocious_bite,cycle_targets=1,if=target.health.pct<25&dot.rip.ticking&energy>=50
actions.finishers+=/rip,cycle_targets=1,if=remains<=3
actions.finishers+=/rip,cycle_targets=1,if=remains<=duration*0.3&persistent_multiplier>dot.rip.pmultiplier
actions.finishers+=/savage_roar,if=(energy.time_to_max<=1|buff.berserk.up|cooldown.tigers_fury.remains<3)&buff.savage_roar.remains<42*0.3
actions.finishers+=/ferocious_bite,if=(energy.time_to_max<=1|buff.berserk.up&energy>=25|cooldown.tigers_fury.remains<3&energy>=50)

actions.generators+=/rake,cycle_targets=1,if=remains<=3&combo_points<5
actions.generators+=/rake,cycle_targets=1,if=remains<=duration*0.3&combo_points<5&persistent_multiplier>dot.rake.pmultiplier
actions.generators+=/moonfire,cycle_targets=1,if=remains<=duration*0.3&active_enemies<=10
actions.generators+=/rake,cycle_targets=1,if=persistent_multiplier>dot.rake.pmultiplier&combo_points<5
actions.generators+=/swipe,if=active_enemies>=3
actions.generators+=/shred,if=active_enemies<3

Exalted
User avatar
Posts: 2278
Joined: Fri Nov 25, 2011 7:49 pm

Re: Simulationcraft APL Discussion

Postby aggixx » Sun Oct 12, 2014 2:01 pm

Hmm, yeah that could work. Something to keep in mind is that when you call run_action_list (and the conditional evals to true) then it runs the action list but does not return to continue down the parent list of none of the lines in the child list end up triggering. Not an issue with your suggestion but I originally had something like that in the WoD list and ended up having to remove it because of that.
ImageImage

Exalted
User avatar
Posts: 588
Joined: Wed Jun 29, 2011 4:49 am

Re: Simulationcraft APL Discussion

Postby Stenhaldi » Sun Oct 12, 2014 6:53 pm

aggixx wrote:And my impression was that the "lost" ferocious bites should be zero-sum: the sooner you Savage Roar the later you have to do it again, which means more time for Ferocious Biting. I suppose its possible that it refreshes SR to be longer than the amount of time the target will be alive for and loses a fraction of a bite, but otherwise it shouldn't make a difference.

If you wait to savage roar, you might cast it with fewer combo points, and that gives you more ferocious bites.

(But overall, the shred-bite system values 1 CP at very nearly 12.5 energy, making this essentially break-even. I'm getting 1 CP ~ 12.3 energy.)

Exalted
Posts: 564
Joined: Tue Feb 08, 2011 5:51 pm

Re: Simulationcraft APL Discussion

Postby ShmooDude » Sun Oct 12, 2014 7:37 pm

Stenhaldi wrote:
aggixx wrote:And my impression was that the "lost" ferocious bites should be zero-sum: the sooner you Savage Roar the later you have to do it again, which means more time for Ferocious Biting. I suppose its possible that it refreshes SR to be longer than the amount of time the target will be alive for and loses a fraction of a bite, but otherwise it shouldn't make a difference.

If you wait to savage roar, you might cast it with fewer combo points, and that gives you more ferocious bites.

(But overall, the shred-bite system values 1 CP at very nearly 12.5 energy, making this essentially break-even. I'm getting 1 CP ~ 12.3 energy.)


That seems to pretty much match what I'm seeing in the sim. For snicks, I took out the 5 CP Savage Roar line (meaning SR only refreshes at <3 seconds with whatever combo points you have) and the dps stayed virtually the same. BT got hit the hardest (but even then only 200 DPS at most) while the rest stayed equal.

Next

Return to Kitty DPS

Who is online

Users browsing this forum: Convery, Yahoo [Bot] and 5 guests