#tc-subtlety
1 messages · Page 3 of 1
it triggers around 4-5 times per sim
i don't think when you refresh rupture is important because of pandemic
e.g. "let rupture run out" vs. "refresh it within pandemic" does nothing for duration
Now gotta test GB > Rupture > build to 6 > Rupture
you mean in the opener?
i am still not sure i understand it
What armin described
do you want to apply a rupture during a running one?
tldr: you want it to play like it always have Finality up
I don't think GB is of importance. I just think they want to minimise the amount of CP spent on the non-finality rupture.
Exactly that
basically do the small rupture into big rupture for the Finality amp
oh, finish for rupture earlier
Only when current rupture is low duration
So basically losing as little CP on a "throwaway" rupture that only exists in order to get the finality buff up for the main-line rupture
Yea
Yeah, except its rigid as GB > Rupture not CP based
Evis > GB > Rupture > build to 6+ > Rupture
Average CP gained from GB is 2 anyway, if what armin said is true
Best case scenario would be to have ShT tick for 1 cp if you want to minimize waste
As long as it wouldn't cause energy capping to occur
This describes how he wanted it to behave
With rupture extensions his napkin math may make it worth it 
So in essence it would be almost permanent finality buffed rupture
Outside of opener which would be a normal rupture
You'd only have to refresh in pandemic there because you'd have finality for regular pandemic
is it meant to rupture 20 times
and spend 90+ cps on it
in the idea that armin had
genuine question, idk what its meant to do
he said 12 cps or some shit
this does look fine
Basically need to check if active rupture is finality boosted
If it is you refresh at 2-3s with low CP rupture into big CP rupture
If its not finality boosted you refresh for finality (in pandemic if possible?)
in the thing i linked above, its also letting finality drop before doing the 7 cp rupture into 7 cp rupture
ye i think i can fix that easy
Only time a rupture won't be finality boosted is from opener too
this refresh is too early as well, its less than 30 seconds after the previous finality rupture was applied
20 seconds later
to be exact
which is weird by itself
So it should be opener dance > refresh rupture with finality in pandemic (or with as little time left on finality buff as possible) > dance > next time out of dance and that previous applied finality rupture is 2-3s left > low CP rupture > build > Rupture > repeat
Not sure if the rupture timing will ever fall in dance with the extensions
May be two dances between each reapply
just to state this, this goes against the usual and tries to be as true to the thesis as possible
took me quite a bit to make the sim do exactly what was asked for
less strict rules would lead to less loss, but also would not fit the described pattern
i do have some other ideas from the testing, just to give a rough idea i had 9 test cases in the end
What’s the main loss coming from
Inefficient cps?
Is it possible to sim it with never refreshing during symbols, to make use of the evis buff
i haven't looked much else than checked if:
- cooldowns get used correctly
- ruptur gets refreshed/applied correctly
but my hinch is it is not one specific thing, but a combination of multiple
if you see something that seems wrong in the sample sequence, lmk
We often see a loss in playing around finality, and I have a hard time wrapping my head around it
How is it that it’s better to drop a 30% buff to a spell over just using 7 cps for it
Is the evis that strong in comparison
this way you also shift things
you introduce inefficiencies in energy
and you potentially have more finisher during high lingering stacks
nothing is without trade offs
but maybe i missed something that leads to more loss
You get slightly lower uptime on dance because you "waste" globals
Well now I just meant actually making use of the finality buff
Not doing the smol into big gameplay
But simply using a rupture before finality ends
That’s also a dps loss
And idk why
It’s at worst like 4-5 seconds off pandemic
Because you overwrite effectively half of your rupture
Half?
28-29 seconds can go before you rupture again
I don’t think the sim plays that way tho
or like 48
7 cp rupture is 32 secs, in pandemic 42, you have 2 dances during that makes it 52 secs
I think it’s just refreshing in pandemic
Cus it was a loss to play around the extension last time we checked
during that time you get another dance so 58
etc.
rupture use is rly low with tier
what we do is introduce additional casts of rupture
so it is not just 30% more damage on rupture
but the lost value of evis
and the things mentioned above
Again, not talking about doing the 2 cp into 7 cp
additionally, you need to do some decisions, because rupture does sometimes line up weirdly
I am not a fan of that gameplay either way, so i hope it’s never good
btw
just to mention it
i do like the idea of
strike-rupture-storm-dance
will test it later for default apl
Rupture just doesnt seem to do enough damage to warrant being inefficient with cps and energy to buff it by 30%. Even going of DPE eviscerate is roughly 30% of a rupture cast by itself (obv evis does less outside of dance and symbols but still)
Even refreshing it before dance so 12 sec or less?
That also assume refreshing in pandemic, if it falls to an early refresh to consume finality it should cause the dpe of rupture to actually go down even with finality buffing it
Isnt it odd that the DPS of rupture is noticeably lower with the finality one? Regular build has 10701 dps from rupture while the one that applied more finality have 9582. That seems incorrect does it not? Ye the apl seems to do what we think it should do but how could it possibly lead to so much lower dps when uptime is only 1.9% worse
Having 30% more damage on most rupture ticks should lead to a very clear increase in rupture dps shouldnt it? Atleast not a noticeable decrease
Is it doing a 30% dps reduction instead 
I'm a bit busy, but I took out the log with just rupture stuff and you can see if things line up as it should buff wise
Pastebin.com is the number one paste tool since 2002. Pastebin is a website where you can store text online for a set period of time.
ye thats what im saying, its doing what we want it to do
But somehow rupture does less damage with finality up almost all the time
Pure dps
I don't know how rollover damage is handled for rupture. Might be correct as is
Or maybe it’s always active
https://www.raidbots.com/simbot/report/8Dk9Mz6VsKWwvvJ8SRZJ5m/simc
Here is a sim without db and then one without finality and you can see finality does increase rupture overall dps
Although i find it a bit sus that it gains about 20% rupture damage despite it only ever using one of the buffs (after the opener)
So my guess would be that finality amp stays in the sim when you refresh rupture in pandemic
Ye
If its been this way since shadowlands, even?

Cos I would assume implementation would be identical
Well it’s a talent
Vs legendary
Doubt it’s just a copy
This might have happened in 10.1 even
Cus stealthi found a 1k dps gain by just ignoring the min/max
In a 10.1 sim
That would make sense if it suddenly just gets all the dps
Without min/maxinf
The min/maxing would never have been a gain if it was always like this or since sl
Different effect between finality from Legion, SL and DF
Ye but the min/max was a gain in 10.0
But implementation wise, no clue
Why would refresh early ever have been a gain if the number on rupture never changed
Seems like a 10.1 thing to me
T29 4p?
But that didn’t snapshot
Ya but that would explain the gain in 10.0
Don’t think so
@rugged solar @coarse laurel If you read through some of the messages above here we highly suspect the damage amp from finality stays on the rupture in the sim if you refresh it and not let it completely fall of. Just having a profile with finality vs one without makes us gain ~20% rupture damage despite actually not using the buff more than a single time, and Fuu's apl optimizations to maximize rupture damage with finality just lead to lower rupture damage overall.
Since it’s just a passive rupture buff
Why would spam refreshing be a gain
Unless the tier set was implemented as a snapshot
did you compare uptime?
just to be sure it isn't a consequence of uptime changes
Yes
also the old finality rule does see a increase without tier
Yes
what could be is
It lost 2% uptime and like 15% pure rupture dps doing your finality optimizations
These seem suspect however
Ye losing uptime is expected, but having significantly less damage isnt when the whole point is to get the 30% damage buff on all the time
what exactly do we look at
if anything, it couldbe that the tier set does change the damage?
log=1 from simc
and just filter out all rows that's not rupture_finality or rupture ticks
It's doing the same damage with and without finality
After applying a finality rupture
Likely shadow dance used there, or SoS, or proc etc
SoD*
I stripped all of that away
oh,
ye thats what i assume
maybe the tier set extension
does not keep the pmult
i see what you mean
the one after should be higher
so the bottom one
because it is without dance
but it seems to be the same
Ye since it doesnt refresh rupture inbetween those
They should be higher since they should be affected by finality
But theyre the same
this is why i assume
that the tier set implementation might not keep finality on rupture
let me quick check the code
I'm in an onboarding meeting with a new employee, so can't really be of much help
It’s the opposite if anything
As in
Finality is always active
Once it’s been applied once
no hurry
that seems fine
Nvm it doesnt seem to be the same without tier when i dug a bit deeper
is the report taking all interations or one specific
https://www.raidbots.com/simbot/report/1GNywXTNkcFHfp1go2bmeQ/simc
Here is fuus old sim with the rupture shenanigans except without tier. Here we can see rupture doing around ~4% more damage with the finality optimizations instead of like 15% less like with the tier
this can happen
it seems
so finality wouldn't be up during that dance on rupture
but overall its not the norm
But also fuu if you check this sim i linked https://www.raidbots.com/simbot/report/8Dk9Mz6VsKWwvvJ8SRZJ5m/simc
You can see that rupture does about 20% more damage with finality specced, despite it only getting the benefit of the buff for a single rupture after the opener
hmm thats a lot ye
but also keep in mind that symbols will lead to higher dps with tier
Ok so how do i get the data rosvall posted earlier after running a sim with log=1
Ty
I don’t believe my implementation should change the snapshot. Adjust duration shouldn’t change the pmultiplier. Not sure if you guys are saying this is different from the game or not in any way though.
not sure, so far it just seems like trying to pin down what could lead to differences in damage output.
#1065728795455266888 message
This post that rosvall linked has a finality rupture ticking for 17536 at 112, then he applies a new rupture without finality at 115 and it still ticks for 17536, then he applies a rupture with finality (but pops dance and cooldowns right afterwards) at 119 but after those cds end at 139 it again ticks for 17536.
Which implies that the finality amp somehow stays on regardless since all ticks seem to do the same amount of damage
Or doesnt get applied at all
cant you look at rupture numbers pre-finality
like the first finality rupture values
i mean
rupture numbers*
should be easy to see if its a finality thing or a rupture thing
my guess is that the sim applies finality on every single rupture post 1st one
and not the other way around
So even finality: ruptures appear to do the same non-crit damage per tick?
12k without finality
17k with finality
and then 17k forever
is what i'm seeing
coincidentally
12 on 17 is exactly 30% diff
Youre comparing hits and crits
The 17k from before is crits
Those 12k are hits so
But it also has pots n cds up early
oh they are crits
I was too dumb to clean the data i got from log=1 properly to have a log where its easier to see (ie one log with finality and one without, perhaps without inev too if symbols has anything to do with it)
But there definitely seems to be something thats weird with rupture finality in sims rn
cant you just do the same shit
and remove pots
nothing else should be buffing its dmg
that early on in the fight
you can, just do potion=disabled
Cleanest would probably be to just make an apl that only uses rupture and a builder. No cd's or anything. And you'd see what happens there. No proc trinkets or so either
https://www.raidbots.com/simbot/report/iKrMcGfUBgVh9SD6u3qKY3/simc
The sim you linked has the base profile doing 10701 dps with rupture (98.7% uptime) and the "rupture shenanigans" profile doing 9582 dps with rupture (96.9%). Obviously the DPE goes down because you cast more but the overall pure dps number of rupture should be higher when you have finality up almost all the time and not lower as it currently is. Having a 30% buff up for a significantly longer period shouldnt result in the spell doing less damage.
Other spells doing less damage because you cast more rupture would make perfect sense, but there is no reason why rupture itself should be doing quite significantly less dps
Is there an easy way to make it not use any of the APL besides just rupture and gloomblade so its easier to see how finality interacts (like rosvall suggested)
yes, just use a custom apl, moment
with shadow dance sync
just using dance on cd
In this first one it works just like expected from what i can see https://i.gyazo.com/ae09c23d4220ace42dde173e5b898db1.png Ill check the other ones in a sec
Well that one drops finality buff.
Just ran one where it ruptures more often and finality works as expected.
Also to note is that Dark Shadow and Finality has the same modifiers. So a non-finality rupture during dance would have the same values as a finality rupture outside dance
Now check one that uses Shadow Dance
then that'll rule out some weird t30 extension bug?
the sims i linked are with 4p
so it wold extend them with tier
it could be just that
events like this happen
even tho i tried to prevent them
where you lose value during the high impact of finality
ya but the one he used doesn't use dance
so if it does it with dance being pressed
that's the culprit
So the sim is saying the dance extension deletes finality amp
And replaces it with a normal rupture?
Or am I misunderstanding
Pretty sure once dance (and/or symbols) fades the finality buff seems to drop
or the amp rather
sec
This is what I would suspect as well
Not very clear but 6s after dancing (the blue stuff) the damage drop back down to red (non-finality) instead of the green finality number
Also dance and finality seem to be additive, is that how it is in game?
No. Once dance (or symbols, they have the same duration in this sim) drops the finality amp is removed

wish we could present this information in wlogs format
would make digesting it so much more manageable
I'm really not totally sure what the jist of this conversation is. Dance duration dropping has nothing to do with the Finality persistent multiplier. Are you not considering the fact that you are losing the Dance modifier itself because it isn't persistent?
if the sim applies a finality rupture, then dances
Also all this stuff is very easily seen in the full debug log
The rupture ticks after dance/symbols have ended are not buffed by finality
Maybe the 4p buff getting removed also removes the finality amp somehow, idk
@coarse laurel We are trying to figure out why what im explaining in this post happens #1065728795455266888 message
Why optimizing and having finality up all the time leads to lower rupture dps (not DPE) than ignoring it and letting the buff fall off
If you give me a reproducible single APL of a case you think is incorrect, I can take a look
I see a number linked above but unclear which is being discussed
@past fable keep in mind that the finality does overwrite it immidiatly
so there is little time with no finality
Ok so what happnes in the logs me and rosvall linked is this:
Non-finality rupture is ticking for 100
You apply a finality rupture and now your rupture ticks for 130
You dance (+symbols with tier) so your rupture ticks for 185
Your dance and symbols end and rupture ticks for 100
The last line should be rupture ticking for 130
Since you didnt apply a new rupture, so it should still have finality
Is this also fuus sims or are these yours without tier?
Think I found the issue, just some very odd SimC core code in the extend logic
nice
Should be fixed now
saw the commit, but needs a bit to be on raidbots latest
alright, thats a 0.3% win now
which one is this
the one i spend more than a hour hacking together
the armin change
yes

why is this sim so low tho
its hard to keep up when we have multiple different profiles being run
because it is without the ovrewrites
can you do it with the overwrites
it is?
the sim we had before was above 136k
i overwrited both
probably before the inev fix?
oment
oh
*moment
Thanks for testing it fuu btw, kinda weird to me why it's still worse than no finality but 

what talents
Oh, how we do it now? Okay nice
you drop inev
same as before it seems
i can try the one that simmed slightly higher
moment
hmm
can you link the sim
BUQAAAAAAAAAAAAAAAAAAAAAAAAABAAAAAAAECkkICJgESCSSSSCJhikElAJlIkkkkWAAAA
uhhh
fuck
i did probably use wrong config
sry
1 boss, will take a bit
BUQAAAAAAAAAAAAAAAAAAAAAAAAABAAAAAAAECkkICJgESCSSSkQSokkEtAJlIkkkkWAAAA
fairly standard
Loool. I'm pretty sure we drop VT and take improved sht after changes and end up around 131
maybe there are energy threashld optimizations
but thats a 1-2 day task to update
so nothing i can do immidaitly
also only usually for 0.2% dps or so if even
yes but lethality fix is another 0.5%
its not in this sim?
no lethality fix is not in to my knowlage
but i double check in a minute need to do something quck
*quick
rly quick loking at wowhead, don't see strike
but will look real spell data soon
Shadowstrike still isn't in Lethality afaik
^ i didn't see it in the hotfixes
https://i.gyazo.com/6a2e7dc4bcd7645bad32713cdce46186.png
When the amps are removed theres pretty much no way this doesnt go above that build fuu just linked though
^ I tested flag. Even with. Lthing to amp it's about 1.2% behind db
Removed rings and trinkets and enchants that would otherwise get amped to test
i know it was in the hotfix notes, but talking spell data
sure
but we are simming NS as if its fixed
so why not lethality
both are not fixed
Well one is an existing whitelist and the other is just something that was broken so the implementation path is different.
Until they update the spell data for the whitelist it's not really practical to update
ye idk
btw. thanks for everyone to look into the sims, nice bug we found and koji fixed
I fell asleep shortly after my last message, nice work ladies 😄
I read through convo
Are we still doing the same thing with finality as 10.0 or ignoring it or what
we would go back to our rupture gameplay from vault
only rule changes for 10.1 so far is symbols use
Alright
most of the apl optimizations else are around trinket use
or talents we don't use
tho as always there migth be more optimizations
Probably just rotten left tbh or 1 dance flag apl
If all our shadow dmg amp talents tank then flag might become best
i don't mean something specific
just in general that there is always something to find/improve
Ye
actions.cds+=/use_item,name=irideus_fragment,if=buff.cold_blood.up|fight_remains<20
@ivory kestrel should probably use this if you dont want it to completely scuff irideus
With cold blood might not be 100% optimal, but its a lot closer than using it like the default apl does, sims around 0.5-0.6% higher in my tests
yea getting it to use it just as you use sec tech might be the thing
couple of things 1) looks like the sims are over riding existing symbols
- not sure if its worth it or not, but with the flag stiletto inev build, is it worth it to hold blades for flag?
Yes, use the apl in the pins
Thats how it works ingame
Unless it was fixed, cant test cus ptr is down
^ no as in, it uses symbols when theres still 3-4 seconds of symbols remaining from dance + inev
should I also change this line?
shadow_blades,if=variable.snd_condition&combo_points.deficit>=2&target.time_to_die>=10&(dot.sepsis.ticking|cooldown.sepsis.remains<=8|!talent.sepsis)|fight_remains<=20
like ill add &cooldown.flagellation.remains<=2
to kind of sync them together
just copy paste entire apl
in advanced
then add a copy with what you wanna change + the block it's in
yeah adding |cooldown.flagellation.remains<=3 is roughly a 300 dps gain
maybe its worth then idk
never mind I tested all variations, syncing flag and blades seems like a dps loss. just send on cd
Can just copy the entire profiles tab from the sheet
I expect more trinket optimizations Is irideus off gcd?
Trinkets usually dont
yea syncing blades+sepsis+flag is a win but only blades+flag seems meh
we might want to see if saving always 1 charge of vanish if possible for flag is a win
Quick question that probs need sim to solve
Okay so with 2p, If Shadow Dance is like 5 seconds till off CD, but SoD is ready to go.
Do i hold SoD, otherwise id be overwriting the SoD buff with the free SoD from 2p?
Hold and see how it goes. 😏
@rugged solar this is a relevant question
you delete symbols uptime by sending it, but you have to hold it for quite some time
if you dont
good question
the apl is not considering it atm, but i can check in a moment
need to do something first
Can check quickly. Just instinctively it sounds like a loss to hold it for half its cd
You would hold it for 9-10 seconds in this situation i think
Even if you override it with 2p?
rly quick test says its a loss
with flag is it a gain to hold vanish for every flag cast possible
to smooth out cp generation
or guarantee it, even
You mean to only vanish when flag build time is up? Or when flag buff time is up?
To build i assume
Toi build yeah
Also could be a thing to check if its worth to hold sec tech to 3rd finisher in dance with flag up, to get more flag stacks on it
So tfd-evis-strike-evis-vanish strike-sectech
One way to do it is prob to check for higher flag stacks before using sec tech
And + dance up ofc
BUQAAWulQAmRd/c4Iy2r7VOUFAAAAAAAAAAARAJJiQASIJEJJRCJRKJJRLJkUSCJJhWAAAA
Is it flag or flag persists, do you know?
Flag
Persist is the 2nd buff
@lilac crescent
Not surprisingly, if it would been positive there's no chance it could be more than an extremely small gain
ya its the difference between 28 and 30 stacks
Problem with the sd line I wrote there is that the "x" seconds are so dynamic. Sometimes 3 seconds SD cd is 3 seconds. Sometimes it's practically 0 seconds
due to cdr
Since SoD buff uptime is consistent, you'd want the same amount of irl seconds before you press SD which.. is hard to model esp since SoD generates energy in itself. Like if you're at 4cp and <X energy then SoD for energy + finish + SD would likely in the situation be better than holding SoD. These types of rules that "on average" is a small increase. While in reality becomes a larger increase once every 10 pulls and a small decrease all other uses I feel is bad practice to implement like that and suggest people doing
is it worth sec tec early in 2nd dance during large aoe if u can fit it into flag buff
as in, a 2nd dance later on
You shouldnt need to
^ in big mass AOE u should be able to fit 2 secetch within flag window
If you do flag>dance and have constant aoe youll get 6 seconds of flag during your second dance
Im still curious about Push
's questions tho
during flag
is it worth it, to push sectech to third finisher when vanish is up?
for essentiallt 14% more mastery
probably not
u lose 6% damage on strike, and evis but gain 14% mastery for a sectech, yeah definately not worth it
And you might fuck your next sectech
yep very possible
the one thing that is true is that in aoe with tornado you can use two sec techs inside flag
@rugged solar whenever you get the chance over the coking days 
So does this mean you would even press SoD in the opener right before shadow dance? or hold it in the opener till <=3 secs remains on the 2p SoD
you hold until 3 secs left of SoD inside dance with opener on normal PE build for 2p
for rotten you do it on pull likely
ty 🙂
does not seem to be worth in a quick test
Dslice?
Is the application I meant not single target
Sorry I wasn't clear
same, but less bad
Damage gone down in Dslice?
yep
small fix for tier set, marginal damage impact tho```diff
- actions.cds+=/symbols_of_death,use_off_gcd=1,if=buff.shuriken_tornado.up&buff.shuriken_tornado.remains<=3.5&!set_bonus.tier30_2pc```
^ change is within margin of error
@rugged solar can you check if it’s more prio damage over a fight with st build to only st compared to only funnel during flag window or possibly even very low lingering stacks
it is almost certainly more damage to pure st
if you want to do the most damage to one target, you just spec full single target and ignore everything else
if you compare funnel to aoe
funnel is 15% stronger than aoe
but pure st is 131k
it highly depends on the talent builds tho
the problem this tier is inev, the synergy between it and funnel is not there
a dungeon build can beat the pure st in heavy funnel possibly, but this just highlights how talent dependent it is
I guess there's some optimization there
I made the assumption that buff.lingering_shadow.remains returns the duration of the buff and not the stacks
~18s
mh results don't make a lot of sense
actions.build=gloomblade,if=spell_targets.shuriken_storm>=3&(buff.lingering_shadow.remains>=13.5&variable.priority_rotation)
Just added that, with different breakpoints
oh nice
yeah I've turned something around, sec
That should be correct, and results make sense
@rugged solar
One or more of the 44 spell target checks needs a poriority rotation variable
"skip all spell targets" is just ST with a lot of residual damage
But the top one uses shuriken storm
All it does is makes every other of the 43 spell_target checks be >999
oh
So one or more of those.. has quite a decent impact on priority target
It's super niche.. maybe not worth investigating
interessting
probably would need to go through them and debug
can probably do later
ill give it a few min before meeting
@rugged solar almost certain it's SnD uptime. So related to this line
slice_and_dice,if=spell_targets.shuriken_storm<cp_max_spend&buff.slice_and_dice.remains<gcd.max&fight_remains>6&combo_points>=4
(Swapping cp_max_spend to 999)
3 casts vs 0.
i assumed it was slice and dice
welp swapping that to
slice_and_dice,if=(spell_targets.shuriken_storm<cp_max_spend|variable.priority_rotation)&buff.slice_and_dice.remains<gcd.max&fight_remains>6&combo_points>=4
was only part of the problem
probably the 2nd snd condition too, moment ill check
that only triggered 0.1 times
now both has <99% SnD uptime
er
99
https://www.raidbots.com/simbot/report/ndvVu9zZXwScS9n5joVEVA/simc
weirdly enough... the one without the checks has lower SD usage... and still outperforms in priority damage
^ seems like a easy one
adding a check to the 2nd snd variable does not change much:
https://www.raidbots.com/simbot/report/qYbHcSrEJf67puGCQnuLgn/simc
yeah it gets hit so seldom
It's the shuriken storm while stealthed
shuriken_storm,if=spell_targets>=3+buff.the_rotten.up&(!buff.premeditation.up|spell_targets>=7&!variable.priority_rotation)
mh
i wonder, moment
just temporarily
shuriken_storm,if=spell_targets>=3+buff.the_rotten.up+(variable.priority_rotation*10)&(!buff.premeditation.up|spell_targets>=7&!variable.priority_rotation)
Makes it better but... still not all the way there
that actually lowers prio damage
I think I might have swapped flasks there and forgot to swap it back
the changes I did, I did not do to the skip one
That peak though
moving priority_rotation up and adding it to slice and dice was the only dps positive change i found quick```diff
- actions+=/variable,name=priority_rotation,value=priority_rotation
Apply Slice and Dice at 4+ CP if it expires within the next GCD or is not up
- actions+=/slice_and_dice,if=(spell_targets.shuriken_storm<cp_max_spend|variable.priority_rotation)&buff.slice_and_dice.remains<gcd.max&fight_remains>6&combo_points>=4```
haven't found much else
the dungeon slice one
https://www.wowhead.com/talent-calc/rogue/subtlety/DAOUVERBARQakWIkRBUORUUVERBQpEQpJEA
BUQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAARAJJiQASEJJxBSSSCJRIJJtAJlIkkkkWAAAA
Mmmh.. I guess we have different approaches here.
My use case was the one Kush was in on Echo/Sark where adds spawn in just one of the phases and they want to maximize boss damage
But does it
Inev isnt much better than say VT on pure st
If you only utilize inev to 50%
Then i dont see it being better
I feel like this point is lost
You'll utilize inev to like 95%
Firstly, the % of the fight on Echo/Sark that there even exists adds is very low.
And then you only storm when you're under like 20 stacks of lingering
and there's adds up
don't rly see a prio increase by playing around lingering/inev
on 8 targets
but maybe my conditions need to be refined
I just added a gloomblade before storm on builders
actions.build=gloomblade,if=spell_targets.shuriken_storm>=3&(buff.lingering_shadow.remains>=2.25&variable.priority_rotation)
Where 2.25 you could replace with (0.25*18) or 0.5 or 0.75 etc etc
that shows a loss too
What if you swap to regular ST talents?
i did
Still not above your test there, but something in the setup is different
ah it's different gear
it is using the m+ gear ye
possible that its related to the high amount of vers
the st gear profile has a lot of vers, while the m+ profile has a lot more mastery
also, what did you change to make it a gain?
I mean, more shuriken = finishers. More mastery = better finishers.
Which makes shuriken higher value
You'll cap the buff either way, so no idea
I meant more so for the extra evis dmg
The one I did there is just proof of concept. I slapped 999 on all target requirements. Except shuriken, because I overrode that with the gloomblade with lingering modifier. Didn't have time to finish going through each of them. But positive ones I saw I added to the profiles not called "skip all target * "
But as you can see, it's still a gain to just skip all targets all together so... there's some more gains to be found.
now I really gotta go
Pastebin.com is the number one paste tool since 2002. Pastebin is a website where you can store text online for a set period of time.
oh, you did just copy that one over
i assumed you modified my one
fair, so there is more room for improvement
#1065728795455266888 message
Just bumping this again for the default profile to include something similar, not that i think it will make irideus fragment a lot more desireable, but certainly a bit better
oh ye
good idea
can probably just make it a item action like for beacon
i think i did for the sheet
Stealthi did ask me to look at Bomb dispenser use, to maybe shift it as a priority to not waste bombs:
https://www.raidbots.com/simbot/report/nWx3u9dKk8qjrhtBfXCS2v
I tried some variations, lmk if anyone finds something
you would trade off the optimization to use it in dance by doing so
but should be easy to test
Its also probably more worthwhile when we have dark brew specced
this one is more towards the idea
but also slightly worse
the only vanish that is used regular is the first one
so it seems not worth it
but also not that big of a diffrence if played somewhat reasonable
will also depend on trinkets if you have things like harlan or neltharion's rupture has more dmg per cast then strike + snd likely and it will line up eventually
Strike from stealth also gets 25% bonus from stealth
yea but not affected by mastery(harlan procs) and agi procs benefit rupture more for the full duration. i mean it being neutral is already a telling it won't be much of a win anyway
the RS thing does make me think tho for 2 target
if it is already neutral for st more or less
about RS the thing that is not affected by finality is the 20% extra from the talent the dmg itself i think does get the finality dmg increase
I believe Fuu said that cleaved Ruptures do get the NS bonus but the Shadow damage additional ticks do not
Yes
Which makes sense as they are triggered separately and have their own coefficient
Is it possible to check if changing symbols logic for dust rewrites is a gain with new tier, for example sending a symbols in opener or other places where portions of it will be overwritten by tier symbols but the cd is being reset by dust regardless, also dances in dust builds seem to be really wonky especially since they’re not using macabre stuff like snd storm etc in dance
needs to be worked on for sure
yes absolutely, also it would be good to get a bit more specific input
i do read repeatedly "it is scuffed", but i can't relate to what part of the sequence needs change or has potential for improvement form this expression
the first change was a sepsis fix, not rotten specific
but also should help rotten
the sequence now is
thats what i changed with "fix storm"
the first slice and dice cast is at this mark in the sim above
ofc, there is no rule to not snd in dance, but it isn't exactly a high impact spell
given the low frequency of use
It seems to be snd in dance because of sepsis proc / the storm issue so that might just resolve it with the storm fix
Push wanted to line up sepsis more with blades. here for example blades misses 3 sepsis ticks so about 30% of sepsis. which is about 0.5* 0.3 % of your sepsis damage over the course of a fight.
yes from my exp using it earlier can be good
if you have wft it syncs well tho
there are times when all 1.5 min cds are coming in 10 secs and dance comes that one is also a point of thinking
^ the thing is its always going to come up ebfore sepsis and flag
with stilleto that is
you would end up holding it on average 10 seconds per cast.
but either way whatever happens, i can guarantee rotten is going to be outperforming current build on farm since it will always do more damage than current build for the first 2-3 minutes of the fight.
i need to sim with 100k iterations
as most apl chanegs seem fairly low in impact
this does not seem better, but worse:
https://www.raidbots.com/simbot/report/gaCubSsVdDd6nRZZxEs5nr
its not doing it correct tho
so moment
yeah thats what I thought. blades is a weird cd, almost never worth it to hold is what i got from my testings too.
ah
Where are you moving shb to right before you sepsis?
Oh whoops lol
Can also check shb before sepsis on opener then shb on cd but line up the 3 min sepsis with shb
Since this builds really only ideal for sub 4 minute fights
i think this managed to do it
but
is a big loss overall
sry, need to go afk for 10 minutes
Oh wow
my power adapter has issues and cutting out randomly, need to see if i can fix that
i talked it a bit on sub normal channel but also saying this here : could be getting 4 rotten stabs inside dance>sepsis uptime
i didnt think about that one with new 2p
although syncing them has done better for me always
because you can use the extra strikes and PV stacks well
How would you get 4 rotten stabs in? You cant dance while dance is active
after using vanish you get dance back
so you dance again
those are 2 more
but first lets fix other things
Oh thought you meant fitting 4 into a single dance
no lol ahaha
just in the 20 secs blades last
i have to try these st builds a bit but im still solid on the camp of subterfuge and tfd+ 1 point in improv dance instead of swift death and 1 point in lingering
theres definitely some cursed things with the current sim (mostly using SND inside dance), but other than maybe doing little optimizations for PV I dont think its actually that bad.
back
same i just have issues with the build in general but the dancs themselves and other things are solid
i also have some arguments about it but will try to explain them better when i have time
the easiest fixes can be as talked before and easy to put as lower fruit is using SoD on pull
did solder together a cable to a old power adapter, because the biggest issue is that the cables always break
i mean the build sims almost identical to current build using NO APL chanes, with BIS gear at 4 minutes and sims almost 1% higher at 2-4 min. and about 2% higher at 1.5 min
so if my power dies again, i try if that worked, it is janky but if it works thats fine with me
that sounds really janky yikes
as long as power does not have an easier way out it should work
no worry, i did 5 years shool for electrical engineering
perfect excuse to do janky stuff
It also seems to sometimes vanish and not strike
alright, trying to catch up
Like vanish into a cd such as flag or sepsis but I’m not sure if that’s just iteration error
i do think this one should be better for ST at least will explain why doing some clips/video and updating excel but will also try the one we have and see BUQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAARSQSiIkkkkEkQQSSCEHoIJRLQSJCJJRKBAAAA
thats the build from the sheet or?
and tfd instead of swift
it is because of things i also put on the excel that can krangle
"I found that using sepsis with blades at 4-5 cp into flag finisher then using the free strike from sepsis to always enter with a PV stack for backstab to be a sure win as it gets to use
the power of tfd and also get the strike used."
"First of all if we don't use SnD in the first globals it will run out before first strike in dance, we also rupture because we have no FW yet after flag
then entering dance with FW is a must because first global is an evis, so shuriken storm to 100% have FW. Dance is normal for rotten with no tier just SoD
first global then stab. We use those because we get bonus from 2p and 4p from S1 on the first evis and the stab that follows with a PV stack.
"
For the subterfuge part
with new tier set this would not change much as you just SoD on pull
Why would snd run out before you strike in dance
with subterfuge it takes 10 secs to first strike
^ too much gap between the two. you have to reapply snd
so it runs out
it is not a loss either because it also takes 9 secs to use first strike after snd
so it does not go above pandemic limit with snd
shb seems fine as is interesstingly
and you are applying FW elsewhere before the first 2 finishers
yea blades sync with sepsis is fine imo
Isnt the opener Strike > blades+sepsis > flag > rupture > storm > dance?
even if you dont get the sepsis ticks in you get strikes in dance and PV stabs
With or without subterfuge anything else just seems not worth no?
different build
i think should be rupture strike strom dance
this is what it does atm
I just assumed its better to waste a strike from sepsis than to waste 4 cp
on opener i ignore first one
^ the strike gives you a pv stack for your first rotten gb
outside of opener i use both
which is quite nice
You already have one from the opener
And you could always just not gloom first builder
And strike instead
hmm yeah youre right
no strike uses both rotten and premed
i dont really like not playing tfd that is why im trying to push a diff one
you also lose 1 global of flag ramp
and talenting swift death i really really
dont enjoy
yeah swift is actually almost 100% a dead talent with this build
but tfd is also really awkward idk
i mean after trying both i will talk more but i still have 200 pages to proofread
and only 4h of day left now
after writing so much
LOL yikes gl
but yea easiest one is just SoD on pull
Rotten is 30 seconds so you can def prepull it
35 second pull timer so i can rotten for pull
okay fair
But maybe you could 25s cd symbols and then swap XD
Nope that doesnt work either
oh thats nice
so 25 sec pre pull is actually just
easy gain
tho you get premed always
What
nvm had a thinking mistake
Fwiw using beacon during flag seems like a bad idea
Not specifically related to rotten but still
i did test limiting that earlier
because kieron mentioned it
but it was in both cases not a damage gain
(not using beacon during shb and not using beacon during flag)
pre pull symbols does sound rly good tho, lets try this
At the minimum can use before opening strike
you get additional damage on strike and a guaranteed crit
so can only be beneficial
and with that
we now have rotten beating non rotten
with potential of more improvements ofc
hmm
think this didn't work as intended
it used symbols pre pull
but 2nd symbols cast is on 25 secs
what seems weird
its prob because of the symbols usage apl
when you vanish you get both dance and symbols up
and it doesnt want to use symbols at that point
i guess
considering its a rotten talent build, its quite bad its delaying the use by 25 seconds
rotten prob needs different symbols apl
than regular
it is fine you use it prepull then after second dance's vanish clutter is done
because it says it ignores the pre-combat option
that is if you use it just on pull
yes, think so too
if you want to game rotten and still have symbols on cd you would still use it on pull
you probably can use symbols earlier with rotten
after first strike
perhaps play more around makgin sure you consume the most rottens possible, instead of keeping symbols up
but dust
cant look at the sim but 25 secs in it is fine
as it would align with the second dance after using vanish and not clipping more or less but if it is not doing that no diea
yea it should use it after the 3 secs of second free SoD on second dance if it is not there something is wrong
and if you're doing the non subter one it does not add up it seems
after the strike finish
Wouldn’t you have symbols for the entire opener / two dances and after
idk with no subter how it really goes it does drop a bit after vanish and as you enter new dance
then using SoD after the premed strike should be fine there
that is if you game with swift death
that im quite against
and 30 secs prepulls seem cursed
i mean on pull
then dance symbols
vanish -it drops off a bit
dance symbols
after premed strike symbols finish stab
basically
atm it does just symbols pre pull
without doing it earlier
but if you do it earlier you basically have symbols up again
Yea I mean even just using it right before the first strike covers a lot with inev
You’ll also have pv+lingering+rotten for second dance gloom no?
Seems like some nutty burst
why set it to 25 sec tho
25 sec is the cooldown of symbols
cant it be worth using it 15 seconds before
ur not using symbols on pull anyway
it will be up mid dance where you'd get it during dance symbols pandemic
you potentially are
you use it on pulll
possibly just use symbols stright away
because vanishing resets it
well isnt the idea to gain rotten on the first strike
because you get it back anyways from vanish
from stealth
yes ofc
fuu means to use the rotten extra cp on the strike from sepsis
but i dont like that build anyway
so idk
you dont get fw up
Yea just symbols strike in opener
i would just symbols strike opener
well i only care about whats the best numerical build atm
but it makes sense in my build normal one is quite different with extra cps and getting fw up
well i will make it the best
sure but
rotten is the best numerical already
my point is
there are some easy gains
with the additional symbols cast
in the opener
yes
lets just try make rotten work with the top simming build rn
and then do hachas stuff
not a problem the SoD on pull i said was the easiest
i personally have little faith in the subterfuge build, but if hacha can cook it up idm
^
one by one things
with the current build using swift we could have a 23 secs pull


