can you try it again with vision minor?
#tc-research
maiden cargo
knotty oriole
Possible it kicks in at 3x Deadshot but these differences are pretty small so could really be anything
golden briar
as per the comments in assassination channel, i'd like someone to look into why this is happening in a sim 🙂 https://gyazo.com/118b3786a9d12f05635be749cca57986
it's clearly wrong and i don't have to explain why
maiden cargo
probably best to link the sim where this happens
golden briar
the formless void minor was just to test its uptime as i can't look at buff uptime in top gear, but it shouldn't affect the APL at all
golden briar
i thought the apl was changed to cast rupture before TB if <9 sec or smt and to prevent casting it in TB (as it proved to be a gain)
knotty oriole
What is it you are concerned about in particular?
golden briar
but even if you ignore rupture, there's late BotE and vanish outside of TB (even though it's off CD)
several things
rupture in BotE crit window + TB, vanish at the last global of TB, BotE desynced from TB
golden briar
yea but that's definitely not the way to optimize, you want bote just before or just after TB so it's up for the whole duration and also vanish during that time for 100% crit in the bonus crit dmg window from bote
like vanish is literally being used on the last global of TB, no sane rogue would ever do that
and it's up before that, it's not a CD issue
knotty oriole
Rupture is not force used at 9s, not sure where that suggestion is from. Haven't seen it before.
APL will avoid using Rupture if TB is up as long as Rupture has at least 2s remaining
golden briar
there was talk about it and it was custom simmed and proved to be a gain (a month or 2 ago)
wasn't sure if it was implemented into the default sim
either way
ignore the rupture
knotty oriole
Have seen no such proof before
golden briar
wtf is with that vanish
knotty oriole
Vanish won't be used if Rupture is refreshable
And with BotE won't be used until BotE is up
golden briar
bote is more than ready at that point (1.5 min CD and used a few seconds into the pull)
knotty oriole
Most likely just a combination of bad CP luck (double mutilates needed for both sets) and bad Rupture/TB alignment
It happens
BotE major makes the sequencing very strict. Won't really see it without BotE since it will use Vanish earlier
golden briar
that just seems extremely wrong... i guess it might be hard to code it properly but this is just wrong in many different ways
prime terrace
you're looking at a single sample sequence
golden briar
yes, i'm not saying it does this all the time
knotty oriole
Unless you can prove it out as wrong I wouldn't assume it is wrong in a generic sense
golden briar
but this shouldn't happen in a sim
knotty oriole
You have a bunch of things going on here
One is TB alignment and usage
One is trying to line up pre-Font and Coral
Another is trying to get BotE up before Vanish between those things
And last is trying to make sure Rupture doesn't fall off
Sometimes the timings are such that they don't really align perfectly
Or you don't get the CP gen you need
But that doesn't mean that delaying the sequence is actually better either
golden briar
i'm not talking about delaying the sequence
knotty oriole
Font timing is already locked in well in advance
So doing a delay somewhere in the middle, or letting Rupture drop, delaying TB, etc. could turn out being worse in most cases
golden briar
but in that exact same scenario, if BotE was used just before TB, and vanish with first envenom in TB, nothing would be wrong
knotty oriole
BotE tries to be used prior to a finisher
Which is why it was after the TB
debuff.toxic_blade.up&combo_points.deficit<=1|debuff.vendetta.remains<=10
"Always use Blood with Vendetta up. Also use with TB up before a finisher (if talented) as long as it runs for 10s during Vendetta."
I didn't implement that line personally but I'm sure given that it has a specific comment that Mystler tested the alternative
golden briar
also, an unrelated question
why does the sim give me a sample sequence of 4 min instead of 5 when i sim for a 300s patchwerk?
torn dome
variance of 20%
golden briar
it does that every quick sim i do tho
knotty oriole
Yeah default settings are +- 20%
golden briar
does it always choose the shortest one?
torn dome
No, sample should be totally random
golden briar
i'll try a few more times but 3/3 so far 4 min samples
torn dome
If you open up the html
knotty oriole
At the end of the day, this is most likely just a very odd alignment. Like the first time I re-ran it, it does this:
golden briar
yea i reran it too and it played properly
knotty oriole
But changing the APL to force BotE after the TB in Vendetta make no difference
Even with like 10000 iterations
Like literally exactly the same DPS
So this likely doesn't matter
golden briar
it just baffled me how the sim managed to do all of those things in such a short sequence but i guess that's like 2-3 iterations out of 10k
knotty oriole
Ideally, we could throw Font and Coral in the scrapper and it would vastly simplify the timing issues.. but oh well 😛
If you can find what you were mentioning about the Rupture refresh window from before, feel free to re-link it. But I don't recall anything like that and it's not in the APL afaik.
golden briar
oof that might take a lot of scrolling or very specific searching
well, i found it, it happened in assa channel on 10/04/2019 but proved to only be a 0.1% overall increase, the sims are expired though so the links don't work
i'm bad at discord so idk how to allow you to go to the message directly from here 
either way i guess it wasn't big enough to implement into the default APL
knotty oriole
Yeah guess that's why fuu never made the change. I can ask him about it later.
regal agate
It was a very specific change with very low win that was why i did not push it (likely).
Probably also needs to be re-evaluated by now if we want to push the change.
runic garden
When are you planning on adding the corruption changes going live this upcoming reset ?
lost echo
reset i think
boreal patrol
I have added my overrides to HeroDamage so they should be represented in the upcoming update by tomorrow. For personal raidbots sims, they should arrive once we have the actual spell data which is with/shortly after reset.
runic garden
thanks !
jagged river
Anyone happen to have the overrride for Gushing Wounds 15% damage nerf available?
Trying to test something, saw the other corruption nerfs/buffs had overrides available pretty quickly.
lusty swift
correct me if im wrong but
i think it's
override.spell_data=effect.318187.base_value=10281```@jagged river'
yeah i think that's wrong
lol
boreal patrol
override.spell_data=effect.803979.coefficient=3.7087
royal hare
where do you put the override string in on raidbots?
maiden cargo
in advanced sim at the top above your simc info
royal hare
ty
regal agate
hotfix changes should be in the nightly now (raidbots nightly build)
spring parrot
Danke
tawny ore
Cool, so even after nerf Gushing Wound still beats all at 445+
lost echo
did gushing even get nerfed
copper glacier
15%
boreal patrol
looks like it didn't, pssht
tawny ore
Gushing had a flat 15% cut and something about the proc when it’s already up on the target
boreal patrol
yes, that's what they said in the bluepost, however, we're lacking the actual hotfix from spell data, so it looks like it didn't go live (yet)
jagged river
so gushing is not showing up as nerfed in the sims then yet
regal agate
yes, it behaves like ingame
boreal patrol
looks like they shipped a second hotfix pass now that included the announced GW changes, raidbots is being updated right now
muted osprey
With the removing of detection working for EV, is it being slightly overvalued in sims?
knotty oriole
I didn't see the proc flags change in the hotfix
empty vortex
in game detection no longer adds stacks
knotty oriole
Might be a new hotfix we haven't pulled yet
Because that would be reflected by the proc flags
(And as such, SimC doesn't really need to be updated if it's in the spell data)
Will take a look
You can always just do a custom APL without it for the time being. Will likely see a big drop for Assassination
Not that it's really worth using for Assassination right now anymore anyway 😦
muted osprey
I'm still going to be rocking it afaik, just need to see how shit I actually am now 
knotty oriole
I think it may be my worst possible trait right now, it's pretty miserable lol
muted osprey
That bad?
knotty oriole
If Detection no longer procs it, it's 110% ded 😄
muted osprey
I knew it was gonna lose a lot but I don't actually know how much detection gave for dps/corr
knotty oriole
Detection was about a 3-4% gain
Out of whatever it was doing before.. 12-13% or so
so 25%+ of the value
knotty oriole
Well Assassination has a lot of downtime
A lot of specs get full value from it due to being GCD capped, but that kind of change is pretty hard on Rogues in particular
muted osprey
Yeah I get that. Didn't honestly think it was that big though.
Welp, EV dead truly then.
knotty oriole
For my own EV profile it goes from like 5.3% of my damage to 4.2%
So yeah, loses about 20% more value
I have made the code fix to SimC
empty vortex
what custom apl can I include to exclude detection use
knotty oriole
Talking to Seri about getting it on Raidbots now
maiden cargo
remove actions+=/detection,if=equipped.echoing_void|equipped.echoing_void_oh from https://github.com/Ravenholdt-TC/Rogue/wiki/Assassination-APL-Research
knotty oriole
But yeah, this more or less kills the trait unless it's heavy AoE (although should probably use TD for that anyway)
With my profile it's down to only around 4% of my damage on ST for EV3 for 60 Corruption
Which is not even close to an acceptable rate
muted osprey
Holy shit that's fucking atrocious
knotty oriole
Most of the mid-tier traits will return around 7% for that much Corruption
vagrant quail
its still not worth it for M+?
knotty oriole
Honestly think stat ones would be worth better, but Twilight Devastation will be a lot better anyway I think
vagrant quail
Torment in a Jar is really underperforming fro me and other friend .
we used it in underrot 15 to test and did really low dmg.
is there any chance that this trinket is oversimming or just bad luck?
knotty oriole
It's always possible. Probably would need to see logs though
I don't have it myself so can't really verify
vagrant quail
did not log it normally don't do for dgs =/
but might worth to ask for the community feedback.
if more people that have it could test we would have more data
knotty oriole
That doesn't seem overly lower than I'd expect though
That's 3.7% for one and 4.2% for the other from the proc effect
ST sims seem to show the proc at around 2%
maiden cargo
if the proc is hasted then it will proc more for DH's naturally
vagrant quail
can't really compare but Harlan's Loaded Dice feels better.
i might be wrong but always thought trinkets contributed more than 5% to your total dps.
knotty oriole
Well it's stats also
willow knoll
I've noticed for outlaw that Breath has gotten a lot better as a major, before it was even with CLF, now its a lot better, what specifically happened to make it better? just the corruption nerfs, or was the apl changed in some way?
knotty oriole
I thought Breath was always basically BiS?
There was a while when it wasn’t critting properly but that was quite some time ago.
willow knoll
the only 3 images I have saved from the graphs on HD are this (which is when corruptions were a lot stronger)
this one, which doesnt have CLF in the image (i think it was above it)
and then the combination page from a few weeks ago, which doesnt even include breath as a major at all
but if you look now, its "viable"
im just wondering what changed to cause this specifically, if you know of any reason
willow knoll
yeah i guess that'd explain it then
torn dome
Most likely
plain coral
Hi folks ! I found out that with heavy crit setup (Coral + BotE + lot of crit) and when using MA talent, it is best to not line up MA usage with Blood buff.
_First suggestion was to use it AFTER blood buff end and it shown an upgrade of 0.7%.
_Second suggestion was to use it BEFORE blood buff start and it shown an upgrade of 0.9% with my profile.
The idea is to use this kind of rotation during Vendetta : TB Vanish Env Mut Env Blood Mut Env Mut Env Mut.
To make this into the APL I replaced the condition for vanish to be used outisde Blood buff insteed of inside.
Then I just added a condition for Blood to be activated only if vanish is on-cooldown.
There is probably a better way of coding it, but this is just the general idea.
Note : It is possible that this change is an upgrade only for heavy crit setup (>40%-45%) and not esspecially an upgrade for lower crit setup (<35%)
https://www.raidbots.com/simbot/report/29xymhWCwGKHee57LXf4x9
knotty oriole
Interesting. Thanks. I will take a look.
golden briar
problem with this being the default apl is that not every1 has high crit that overcaps during bote
also font+coral timings on patchwerk are such that you don't have coral during the first 2-3 globals of TB+bote so you can use MA there
knotty oriole
We can probably put in a conditional for crit rating or crit% if needed
If I can figure out a solid breakpoint for it
plain coral
I think it's best not using BotE sync with Vanish anyway, like even with low crit setup. Basically crit have diminishing returns. A 50% crit buff will worth more if you have only 25% crit base than if you already have 50%. Now BotE also give crit damage and the question is is that buff enough to compensate? and I think the answer is no.
Quick maths :
2.25/(1+1.25*0.5) = 2.25/1.625 = 1.385
1.75/1.25 = 1.4
This is relative gain out of the 50% crit buff from MA with 25% base crit, first with BotE Crit buff + Crit damage buff, then without. The full plot here show that in fact the MA crit buff has always more value when not used with BotE crit+critD buff.
https://i.imgur.com/OdaZQY9.png
Theese are just quick maths based on a single scale factor, it does not take in account any other "could-be-lined-up-with" buff. Its purpose is just to illustrate the whole thing and cannot replace real simulations.
torn dome
This only seems to apply to font+coral
knotty oriole
I think there are other factors though which is why it sims less in the lower base Crit cases.
Primarily things like variance reduction and being able to fit in more finishers in the timing window due to the CP gen without having to double Mutilate as frequently
plain coral
Hello, I found out that when using VoP minor and with my own profile, using Vanish on cooldown insteed of synchronizing it with vendetta (every 3mn so) is better.
This is the line I changed to a get a 0.7% increase which place minor VoP more in line with other alternative.
Default
actions.cds+=/vanish,if=talent.master_assassin.enabled&!stealthed.all&master_assassin_remains<=0&!dot.rupture.refreshable&dot.garrote.remains>3&debuff.vendetta.up&(!talent.toxic_blade.enabled|debuff.toxic_blade.up)&(!essence.blood_of_the_enemy.major|debuff.blood_of_the_enemy.up)
Changed
actions.cds+=/vanish,if=talent.master_assassin.enabled&!stealthed.all&master_assassin_remains<=0&!dot.rupture.refreshable&dot.garrote.remains>3&(debuff.vendetta.up&(!talent.toxic_blade.enabled|debuff.toxic_blade.up)|(!cooldown.vendetta.up&debuff.toxic_blade.up))
https://www.raidbots.com/simbot/report/a6CTkzNnhnm8piSNjy9Qy8
knotty oriole
Did you happen to check Subterfuge?
west prism
What about subterfuge with 1 SS
and more NP traits, i wish i could sim it myself. But i tried and came out at 71k dps (7k below my normal sims) so i very likely did something wrong.
knotty oriole
SS is not in a great position right now. Subterfuge is a viable option even with a triple DD setup though.
For my current gear Subterfuge is within 50 DPS (virtually identical) of MA for single target with 3x DD/NP/HoD
west prism
Yea i just gambled a head with NP and SS, wanted to see how it performed with this build.
astral trail
@plain coral What about font in that case? It wouldn't line up anymore 🤔
knotty oriole
I suspect any kind of adjustment like this would need a lot more logic attached to it. But it’s an interesting idea. I’m going to investigate in depth tomorrow.
plain coral
No font wouldn't line up and APL will use it once every 3mn (just like it did with vanish). Subter is for me an upgrade in any case :/
I may need to check but I'm pretty sure default APL use vanish on CD while using subter
knotty oriole
I improved the Subterfuge logic a bit last week.
It is not strictly on cooldown but should be closer now.
Font+Coral syncing is a major problem for all Vanish stuff unfortunately. So it’s always a set of trade offs.
plain coral
yes all the sim I did earlier 'bout VoP was without font
A good 475 vita charged might be better than a 445 font theese day anyway
It's worst when it comes to playing VoP BUT sometime delaying it like on Mythic N'Zoth (which is why I'm actually wonder for VoP, playing Font is fine because you vendetta every 2mn for the 6 first mn, but then vendetta on CD so it's a big mess ...)
knotty oriole
Yeah. I have a bunch of time tomorrow so I will dig into these VoP cases based on what you said and try to come up with something that doesn’t affect non-VoP. Maybe also check for Font Coral conflicts. I just need to test a few configurations.
plain coral
https://www.raidbots.com/simbot/report/qRT3NDrHGABK8RvxjW3G7F
If you ever manage to understand wtf is wrong with this sim, please let me know because I still don't understand. The "new" stuff is just an exact copy of the default APL (copy pasta of Github)
But for unknown reason and with Vita + Font setup only, it does show such a big difference
plain coral
About Coral + BotE + MA thingy, I found that using vanish AFTER BotE debuff faded is a more stable way of dealing with the crit overcapping. I make sure that I didn't used Focused Resolve potion and tried with Vita + Coral and it still shown an upgrade of around 0.5%.
Note that you will always overcap crit during MA buff when playing BotE (unless <25% base crit) but the first vendetta does not have Coral crit buff, so you still get a large amount of crit out of MA buff during 1st Vendetta. Yet I didn't try to make the APL overlap MA + BotE only for the 1st Vend and not overlapping for others.
boreal patrol
Ohh, good catch, if you took the APL copy that is on our wiki, I ran a diff and it appears to be slightly out of sync and missing a few changes (mostly around Font) Koji made a while ago. What makes this interesting is that the seemingly outdated Wiki copy sims better for you.
knotty oriole
@plain coral I did narrow down the issue. Looking into it. It seems like a very specific edge case with Toxic Blade cooldown alignment causing a late Font cast.
Not sure what profiles it would affect but the way your casts align in your profile seems to make it really bad.
I should be able to find a solution to it
fervent onyx
Been testing out drestagath trinket since it dropped from mythic with socket. Was looking at the APL and it uses it in M+ for assassination after 3rd Garrote (4th cast in action list). I think it should be held till after BotE since it has chance to crit per enemy hit. Was wondering if using the trinket as 4th spell was just a default thing
When using BotE before Writhing on use it seems to be a 3%+ gain to overall damage for the trinket through +15 dungeons
Doing more than Torment in a Jar at same itmlvl with socket
knotty oriole
@fervent onyx That is likely just the default use_items behavior. We haven’t put a specific line for this one in.
Should be simple enough to add a BotE sync line for it.
knotty oriole
Ok, I figured out the issue Mikihisa. It only applies to non-Coral setups with Font and only in specific timing cases. So thankfully this should be pretty rare. It's nice you caught it.
Basically just need to change the Font line to ignore Toxic Blade's cooldown when calculating the offset because Vendetta with Font and no Coral also ignores TB syncing.
(I did confirm it was still correct to do that for Vendetta, and it does seem to hold.)
We should be able to get the default APL fixed pretty easily tonight or tomorrow.
plain coral
Ok pretty nice you made it, I'm still surprise this made such a big difference
maiden cargo
For @lean talon 's outlaw profile with 2 Brigand's Blitz, not using Blade Rush during ADR is a ~600 gain. Also tried changing the current time to max threshold on Blade Rush to different values https://www.raidbots.com/simbot/report/e1WEHmGXWgLMN4LcHSFNjo
For a profile with 1 Blitz changing the energy thresholds is a smaller but similar gain, perhaps worth looking into https://www.raidbots.com/simbot/report/qgqmFpVmQQNJCp3z8JZe9P
knotty oriole
I think changing the time to max threshold is probably more generally applicable
Probably why the AR check works ever so slightly differently is because the time to max doesn't consider the increased max during AR now so that Energy does effectively overcap
Once it goes away
maiden cargo
ah good point
knotty oriole
Could maybe have both
Or maybe check against the normal max Energy
I think maybe leaving the AR condition is might be the easiest, but I'm going to put in a fallback of energy<45 just in case you get really low Energy for some reason and can't afford a SS
But yeah this seems like a decent general-case optimization. It's at minimum 0.3%, which is small but seems like a clear gain
Almost 1% with multiple traits, so will get this added. Thanks
Going with
if=variable.blade_flurry_sync&energy.time_to_max>1&(!buff.adrenaline_rush.up|energy<45)
@plucky charm noticed the feral rogue APL tries to execute mobs with the Breath of the Dying major which give it a significant boost in dungeonslice.
It adds these variables:
actions+=/cycling_variable,name=reaping_delay,op=min,value=target.time_to_die```
And changes the reaping action to this condition:
```actions.essences+=/reaping_flames,target_if=target.time_to_die<1.5|((target.health.pct>80|target.health.pct<=20)&variable.reaping_delay>29)|(target.time_to_pct_20>30&variable.reaping_delay>44)```
Copy/pasting these changes gives it a ~3k increase in dungeonslice with my profile <https://www.raidbots.com/simbot/report/dLFjsBh8PpAtwhAz6e5FPp>
I'm not sure if it can be better adjusted for rogues or if that mattersYeah that’s on our list to bring over. It’s pretty new functionality.
teal forge
I've been thinking if it's worth keeping TB with mfd in aoe during AR with 2 BB traits.
Could this be tested somehow?
verbal latch
MfD typically favor TB because I remember simming my reroll logic with mfd and it showed me TB being top. Having said that, I typically hold TB during AR and reroll. This may be one of those feel craft feeling but it's what I do sometimes if I need cd.
maiden cargo
cooldown reduction for marked for death is misleading in aoe unless you get enough CDR that you can cast it again on another target before it resets
because once it resets all that CDR meant nothing
should be an #outlaw discussion, not here
lost echo
since sims are only doing one garrote in my vanish-subt windows, wouldnt it be a slight increase to vanish 3 seconds before doing that one garrote, since its still within the subt window while putting vanish on cd sooner?
knotty oriole
That’s not a bad idea in the opener. I’ll check it out.
cedar urchin
how much of a loss is it to hold vendetta for vanish so you can get off a stealthed garrote for the entire duration of vendetta?
i would run my own APL but programming noob. I'm guessing this has already been checked thoroughly as well, so I assume there's an answer
unkempt crescent
You mean with supterfuge talented?
I think the apl holds vendetta to align both after the 2nd use if you won't lose a cast
knotty oriole
I’ve tested this one a fair bit and has not really been shown to be a gain. I’ll probably recheck it again at some point.
lost echo
@knotty oriole dont mean to pry but any update on the vanishing before garrote to put vanish on cd earlier? if you do it two seconds early it lets you get another envenom in tb before you refresh the garrote if that matters
knotty oriole
I tried it two days ago and got it working but didn't seem to net any DPS differences
lost echo
sick
maiden cargo
Hey guys, I have freshly arrived on this Discord so don't be too harsh :d but I was wondering if any of you still got some spreadsheets for 4.3.4?
regal agate
4.3.4 is cataclysm what could potentially have a simc version .
But this channel is focused on the retail game and spreadsheets are not used anymore.
maiden cargo
alright, thanks for the quick answer bud
tough hornet
https://www.herodamage.com/demon-hunter/azerite-stacks/ds-t25-havoc where is this gear set? looks like its using 8.2 talents and gear
id like to update it
boreal patrol
I don't think DH profiles are on topic for this channel or even discord but, yeah, it's the DS profile and i think nobody updated them aside from Rogues.
pliant kelp
hey not sure if this is supposed to be here but i've noticed that simc doesn't accurately reflect vop major behavior compared to ingame
https://www.raidbots.com/simbot/report/fexcDXn1VjiSX51aDpBAfz
in this sim at the 3:03 plater cast vendetta simc procs a vop major at 3:20 and the actor gains the haste buff but it does not extend the np buff and most likely neither the vendetta debuff
https://www.warcraftlogs.com/reports/DvRpC7ybfAdPV29G#fight=last&type=auras&pins=0%24Separate%24%23244F4B%24casts%240%240.0.0.Any%24129342535.0.0.Rogue%24true%240.0.0.Any%24false%2479140^0%24Separate%24%23909049%24auras-gained%240%240.0.0.Any%240.0.0.Any%24true%24129342535.0.0.Rogue%24false%24289467^0%24Separate%24%23a04D8A%24auras-gained%240%240.0.0.Any%240.0.0.Any%24true%24129342535.0.0.Rogue%24false%24303344^2%24Off%24%23DF5353%24expression%24nothing^2%24Off%24rgb(78%25, 61%25, 43%25)%24expression%24&hostility=1&spells=debuffs&view=events&start=8946703&end=9077772
in this log i cast vendetta at 5:43 and get a vop major proc at 5:58 and it extends for the full duration where nothing personal falls off at 6:10
https://www.warcraftlogs.com/reports/DvRpC7ybfAdPV29G#fight=last&hostility=1&source=46&type=auras&spells=debuffs&pins=0%24Separate%24%23244F4B%24auras-gained%241%240.0.0.Any%240.0.0.Any%24true%24158041.0.0.Boss%24false%2479140^0%24Separate%24%23909049%24auras-gained%241%240.0.0.Any%240.0.0.Any%24true%24158041.0.0.Boss%24false%24286581^2%24Off%24%23a04D8A%24expression%24vend^2%24Off%24%23DF5353%24expression%24felori&view=events&start=8986514&end=9064297 (same log)
here it shows the vendetta and nothing personal debuff on nzoth applying at 5:43 and falling off at 6:10
just thought i'd bring this into someones awareness
the potential loss of vendetta debuff and nothing personal dmg would probably be somewhat significant i think
knotty oriole
I'd have to double check the logs but as far as I know the sim behavior is correct
"and most likely neither the vendetta debuff" in your first note is not correct, the debuff is extended
pliant kelp
ah
knotty oriole
Last time this was tested the NP buff did not get extended in-game
If this has changed, it would be recent/ninja-change though
Which is definitely possible
pliant kelp
maybe i didn't start playing this until recently
but for sure the np buff on players is extended
that log shows it
pliant kelp
anyway to get debuffs to show in simc though?
knotty oriole
Can run the log for a single iteration or you can look in the target
pliant kelp
i'll take a look thanks
knotty oriole
Fluffy_Pillow : 0 dps -> Buffs
pliant kelp
yeah i see it now
knotty oriole
But here was my last note about it
// 6/22/2019 - Vision of Perfection procs trigger Nothing Personal for the same duration
// This only applies for the initial cast/trigger, extensions do not trigger NP
// NP dot is logged as 'refreshed' after a true Vendetta cast, but adds no duration
It is definitely possible they fixed this without saying something in one of the patches though
It would unfortunately be somewhat common with BfA patches 😦
pliant kelp
i'll look through my log some more
that log i linked has almost every possible condition though
knotty oriole
It is obviously a bitch to test
Since getting a proc when Vendetta is up is fairly rare
Which is why this being fixed probably wouldn't actually be a huge DPS gain, but if this is the behavior now it would be good to adjust
But cool, I will check out the logs in a couple hours
Have to spend some time banging my face against N'zoth
knotty oriole
But it does look like you are right that this behavior has changed based on the logs
errant tinsel
yea the np buff on player extends aswell
knotty oriole
Something not obvious in the logs
knotty oriole
Do you know if you got a re-pop of the energy regen buff?
The 3s base Vendetta one
pliant kelp
uh
knotty oriole
It is not showing on WCL
errant tinsel
dont think so
errant tinsel
i just tested on dummy and it extended
pliant kelp
if it is inside of a player cast it does not
errant tinsel
but didnt notice the other buff popping again
knotty oriole
Ok so the 3s energy thing is not on refresh, just on trigger
knotty oriole
Right now that is what SimC does and what I had seen also
The main difference here is the NP behavior
It will be fairly easy to change that to work like Brigand's Blitz does
errant tinsel
why cant we see np energy gains in logs?
errant tinsel
hm okay
knotty oriole
Restore Power auras work different in logs than Periodic Energize
They just add to the natural regen ticks
But the NP buff is refreshed on the player so it should work
In those logs
Last I checked, casting Vendetta when a VoP proc was up was still kinda janky though
At least it definitely is for Adrenaline Rush
Since it fully resets the Brigand's Blitz buff instead of extending it
knotty oriole
Yeah
Anyway yeah those logs look solid
I will make the change tonight
Thanks for the report
pliant kelp
np
knotty oriole
(I really wish stuff like this was in patch notes... lol)
pliant kelp
same
pliant kelp
@knotty oriole one last thing i noticed,, on the sim i linked you shouldnt the uptime of the nothing personal debuff listed in the damage breakdown matchup with the vendetta uptime in the flufft pillow debuff breakdown? The np uptime is 31.9% and the vendetta debuff is 35.91%
knotty oriole
It's the same issue
They are applied at the same time
I have a fix done now, just finishing testing it
pliant kelp
Ah, so sims were maybe missing a few ticks of np dmg too?
knotty oriole
The difference does seem quite minimal on average, something like 0.3% damage from NP, plus 30-40 Energy
But it's something
pliant kelp
Probably a couple hundred dps at most then?
knotty oriole
Something like that
But I didn't check every combo
So will have to see
Should be on Raidbots tomorrow
pliant kelp
Nice, thanks for looking into it
pliant kelp
Looks like the update is in, its a 1k inc for me much more than i was expecting
knotty oriole
Nice!
tough hornet
when you guys post an update like the reaping flames usage in #tc-updates does that apply to any class using reaping or is it just in the rogue apl
or in general essence/trinket changes that could apply to multiple classes, are they rogue updates only?
knotty oriole
Unfortunately there isn't a great way to roll it out to all classes generically
Personally, I updated it for all Rogue specs and Havoc
Feral already had it
But I will make a note in the SimC channels about it 🙂
maiden cargo
triple garrote after vanish with 3ss seems to be a tiny dps increase (372 in this case, with raz' gear, around 300 with my shitty gear): https://www.raidbots.com/simbot/report/6nTX7VzyBwJAXxnGzytfTt
knotty oriole
I don't think it's so much triple garrote as much as the last garrote at the end of vanish giving another couple empowered ticks
Since typically it will use Envenom or Rupture in the 3rd GCD due to capped CPs
Possible the better solution would be just to try to get to 2 CP before Vanishing but I'll take a look
Possible the wasted CPs don't really matter all that much
maiden cargo
forcing blood after exsang seems to be an increase as well (which is basically how people play it, looking at logs)
and i dont know how or if that makes sense but maybe you could also check out if using this exsang better would be an increase so you get as many ticks into that vendetta&blood starting at 2:22 -> https://i.imgur.com/6XH6fEN.png (or delaying the 2:18 vanish for it even)
knotty oriole
Seems a little unlikely on that one. If anything would just be shifting blood rather than Exsang.
maiden cargo
im just randomly removing stuff until i see a bigger number 
knotty oriole
That timing is very non deterministic
It depends on when you have a refreshed Rupture
knotty oriole
Delaying Exsang to sync up better likely not a good idea
maiden cargo
ideally you want to squeeze everything you have into those dots
knotty oriole
It’s gonna fit inside Vendetta anyway
Ticks before or ticks after makes little difference in that regard
Blood is the only timing factor here that doesn’t overlap as well
maiden cargo
the more ticks get blood+vendetta+trinket+whateverelsewecancontrol the better, no?
and exsang seems to just do that
knotty oriole
Sure but Vendetta and Coral will both affect all the Exsang ticks in that example above regardless
Still 14s left of Vendetta there
maiden cargo
are that buffed garrotes from the vanish at 2:18?
knotty oriole
So weather you compress the start and have more normal Garrote ticks at the end of vendetta. Or compress the end and have more normal ticks at the start. It’s all the same.
Blood the relevant because it’s such a tiny window.
So shifting that may make sense.
maiden cargo
as long as it's the buffed garrotes from the vanish in the blood and vendetta window it should be fine right
and then exsang them to squeeze as many into blood as possible
knotty oriole
Your blood line could probably just be set up to check the Exsanguinated/pmultiplier state rather than the cooldown so that it works mid fight
maiden cargo
i dont know how 
maiden cargo
thanks! 
knotty oriole
@maiden cargo yeah just actually forcing Blood to sync up with Exsang specifically seems to be the solution
That is easy enough to add, I just need to clean up the expression so it's a bit easier to read
maiden cargo
awesomeee 
knotty oriole
I want to test some other stuff though with a few other essences..
maiden cargo
raz played 3ss+exsang+deeper+clf on xanesh if that helps: https://www.warcraftlogs.com/reports/3fq1J9yBk7GxD8jh#fight=52&type=damage-done&source=11
and this guy with breath on destagath: https://www.warcraftlogs.com/reports/78RdbzaTF4tKhgVc#fight=55&type=damage-done&source=204&translate=true
knotty oriole
Also for the triple SS thing I think mostly this seems to be a side-effect of one of the vanish conditionals just occasionally doing it at an inopportune time
Just disabling the Exsang-specific Vanish line in favor of the existing Subterfuge line seems to fix it
Basically the new Subterfuge line I added a few weeks ago is just better even for Exsang, and gets used most of the time. But sometimes the Exsang-specific line kicks in
The F vs. the G there
Just eliminating the F branch in this case makes up that 0.3% difference
knotty oriole
that bit there is to ensure it always overwrites a non-buffed Garrote in vendetta
Even if it isn't in pandemic
maiden cargo
ah okay just seemed weird
knotty oriole
It applies primarily to the second Vendetta where Vanish is potentially delayed due to the opener
knotty oriole
@maiden cargo Ok I think I have close to all the changes I want to make that work for all the various spec combos without a loss. I will check it in tonight, should be in tomorrow's nightly.
Overall it's about 1% or something in total
mystic jolt
I was just messing around with talents and I noticed in the opener that the APL is casting mutilate at 5 combo points (with DS) to get to 6 right before vanishing, rupturing, garroting, and then using exsang - would it not be better to get that last combo point with garrote?
knotty oriole
Could just be an artifact of the rare case I'm removing as part of the changes today. I would test it tomorrow.
I'm changing a large chunk of the vanish/stealth dot logic right now
To split off some of the Nightstalker and Subterfuge logic to be cleaner
The new logic should yield something closer to this most of the time:
Alright @maiden cargo it should all be in SimC now, so should make the builds soon I presume. Ended up going in a slightly different direction, and found some more gains. You can find the actual APL changes here. https://github.com/simulationcraft/simc/commit/f13369909d79533eeefd92abc991316955b463f2#diff-708721a44294e76f0f7f2275360bd3bd Will update the wiki tomorrow.
Had to test this with every combo of SS/non-SS/TB/Exsang/Subt/NS so it took a bit to verify. 🙂
maiden cargo
it's a 1.6% increase for me so i'm more than happy, thanks again 🙂
also, what wiki?
knotty oriole
Now I'm having to fix up the Galecaller's Boon logic for Exsanguinate since it's pretty terrible with Coral getting in the way now.
remote granite
I decided to look into it myself after seeing koji's post in updates - @knotty oriole it seems that Diver's Folly damage is also increased by the damage buff from Ghostly Strike. Have a short log here after i realized I was getting consistent massive proc's on single target. https://www.warcraftlogs.com/reports/6NGPaHr27VjKvX34/#fight=last&type=damage-done&source=1
I just compared the stated tooltip damage (the buff says how much itll do when it procs after the intial 10s window has faded), and how much it dealt within ghostly strike.
might be worth including in that nightly if that interaction is also a thing, which would make it even stronger with the right set up
knotty oriole
Ghostly Strike is implemented as a generic composite_player_target_multiplier in SimC, so I believe that should already be accounted for.
It is one of the few things that is not whitelisted right now
Increases all damage afaik
remote granite
yeah, that's why i went to check, which actually makes it incredibly strong
260k follies on ST without any rediculous buffs
maiden cargo
is there a chance the outlaw APL wiki could get updated? 
oh also, because killing spree memes were brought up in outlaw, we did see that not using it during ADR was a decent gain on single target for this profile https://www.raidbots.com/simbot/report/ch1xBH4KJxAKv3pdorTSW1
knotty oriole
Yesterday and today were my raid nights so tomorrow is when I will catch up with Wiki work.
I suppose it's just the energy<15 part that allows it to happen
I can't imagine there is actually a time during AR where energy.time_to_max>5 would ever be true
But that energy condition was written in days before Lucid so and such so not too surprising
Too bad I found out about the indirect KS nerf that happened at some unknown point where it only generates one Wits stack per teleport even though it has MH+OH impacts
Gonna offset all your sick gains 😄
(Love that KS is so bad that using it 4.5 times instead of 6.8 times is a DPS gain)
(Also love that that the 1.7% was a bigger DPS gain than the total amount of damage KS does of 1.6% haha)
keen horizon
suppose i wanted to build a simulation of a specific dungeon where would i start?
i can't seem to find the dungeon slice code
knotty oriole
The base DungeonSlice events are in code
But map to this
raid_events_str +=
"/invulnerable,cooldown=500,duration=500,retarget=1"
"/adds,name=Boss,count=1,cooldown=500,duration=140,duration_stddev=2"
"/adds,name=SmallAdd,count=5,count_range=1,first=140,cooldown=45,duration=15,duration_stddev=2"
"/adds,name=BigAdd,count=2,count_range=1,first=155,cooldown=45,duration=30,duration_stddev=2";
keen horizon
ah thanks
tough hornet
any chance we can add a ThiccAddPull of about 10
knotty oriole
I was thinking about it. I did some analysis of the "denser" dungeons a while ago and made a framework for it but never got implemented
When I looked at analyzing some routes last season I ended up with something like this
At an initial glance, I don't think it'd be unreasonable to generalize this as 5 pulls between the boss, equal parts 1T -> 2T -> 3T -> 4T -> 5-6T, with a 25% random chance on every entry to spawn a short-duration minion wave of some reasonable size.. Probably 6 or something.
Although it is really worth noting at 6 of the 10 dungeons have no "minion" waves of note and almost all of the minion stats are due to AD (7), Shrine (15), Siege (10), and Waycrest (8)
Obviously some groups pull mega big, but something like a default fight style is going to be aimed at a broader group
I think the main barrier to this was implementing support for the RNG chance at the trivial mob waves since we don't have support for random raid events atm I don't think
But it is kinda on my radar
knotty oriole
@maiden cargo @remote granite I've got some bad news for you
Turns out KS is not even worth the loss from AAs during the channel anymore
So RIP
(for ST)
remote granite
lmao what a garbage talent
knotty oriole
I'm fixing the time_to_max stuff in SimC now and simplifying the expressions
For both this and Blade Rush a little bit
Probably will be something like
killing_spree,if=variable.blade_flurry_sync&spell_targets.blade_flurry>1&energy.time_to_max>2
blade_rush,if=variable.blade_flurry_sync&energy.time_to_max>2
Based on some testing with my local build
KS only worth using on AoE atm
This is 2k higher than the no-AR one for AoE situations
2.6%
Although that doesn't help much lol
lost echo
anyone ever consider an addon that lets you grab the simc topgear string for a given piece you can shift click
that'd be sick
maiden cargo
you can do /simc shift-click-item and it adds it at the bottom of the output
maiden cargo
yea i just found out 2 weeks ago its amazing 
knotty oriole
@maiden cargo wiki should be good now
unkempt crescent
2 talents worse than no talents, what a time to be alive
golden briar
to be fair
at least with killing spree you have the option of not casting it
slice and dice straight up removes your good ability and replaces it with a bad one
prime terrace
https://www.raidbots.com/simbot/report/aMXhjzk2ce8e22Mf6kNHWR
copying over the time till adds spawn check from iris to blood seems to be a gain with blood in a dungeonslice environment, didn't check any other fightstyles since hac is the only other one i know that has similar raidevent add spawns but it makes no change cause of the target count i guess
might be a slightly bigger gain post-prey since it'll effectively be another prey on the weak blood aswell though idk on that
(not checked several profiles as of yet, just my own)
knotty oriole
Prey won't impact DS damage all that much. The impact is only limited to the primary target
Since it is a damage taken multiplier it does not affect Blade Flurry
@prime terrace also keep in mind that is slightly redundant inasmuch as
variable,name=blade_flurry_sync,value=spell_targets.blade_flurry<2&raid_event.adds.in>20|buff.blade_flurry.up
Basically is already doing the same thing, just with a 20s threshold instead of a 60s threshold.
I'm a little surprised it is worth holding BotE for 60 seconds given that it is only a 90 second cooldown...
But suppose it's possible
Given that the add waves in DSlice happen every 45s though I suspect it's not actually testing the maximum there
prime terrace
i actually didn't check the blade flurry sync variable at all for some reason which is pretty silly of me
knotty oriole
Did you happen to try any other values? 45s, etc?
prime terrace
i checked 30 in dungeonslice when i was trying to make sure it wasn't some massive loss on hac or something and the gain was still a gain but much smaller (0.3%~ instead of 0.9%), i've not checked a bunch of numbers beyond that though
knotty oriole
45s seems sufficient
Mostly just for the timing of the second use at the end of the "boss" before the first add wave
Updated
maiden cargo
Is there a way to sim an increased uptime on Vita based on how many are in your party?
solar swift
i tried to check build variations for 3:20 boss fight and find weird sync with blood+vop+exsang: https://www.raidbots.com/simbot/report/mwqWBfTGbuH2eXua9nRkvf/simc (check Blood + VoP / Exang), sim doesn't use bote+vendetta with exsang on 1:38, but w8 empowered garrotes on 2:28 exsang (and waste 1 vendetta and bote), also doesnt use last exsang on ~3:15. And why vanish + triple garrote (in Blood no VoP / Exang)?
solar swift
https://www.raidbots.com/simbot/report/oDrJyZZ9ZisWq1S6JLoFeq using vendetta + bote on CD
knotty oriole
I don’t think the syncing is a problem here in general because that has been tested out as being reasonable to desync.
This is likely just a fight length issue
Not all of the cooldown lines have special cases for dealing with fight length offsets so I’ll take a look at which is causing it in this case.
To some degree you are going to have to accept some odd timing desync if you are attempting to use VoP though because it doesn’t affect Vanish. That’s most likely the cause of this.
I believe I put some condition on VoP vanishes/vendetta for the TTD issue but probably is not on the Exsanguinate lines.
solar swift
Of course I understand this. But why 3 garrotes?)
knotty oriole
Since they have different conditional sets
3 Garrotes I would need to check on. You want to use 2 Garrotes anyway
solar swift
knotty oriole
Could just be an energy threshold issue. Usually it should be trying to use Mutilate there but looks like it doesn’t have enough energy
Minimal DPS difference though
That type of thing washes out over the iterations
Mutilate doesn’t really contribute much damage in this build
Quite possible if you ran this again you would not see it. This is just a single iteration.
I have a 1.3s window on the “last second” Subterfuge refresh to make it less timing sensitive to energy issues (more realistic) but this is just an awkward energy amount since it can’t afford any filler
VoP is kinda prone to fight length barrier issues because of the Vanish/Vendetta desync issues and even in this case the fixed version performs pretty poorly (far worse than non-VoP) The using on cooldown approach likely doesn’t work for other fight lengths so would need “wasted charge” sync logic similar to what we do with Guardian of Azeroth + Vendetta.
(It tries to sync up with Vendetta as long as delaying it wouldn’t cause you to lose a cast over the remaining fight duration)
VoP minor in general has pretty limited value for us outside of specific encounter timing issues.
Even then it’s a little iffy.
solar swift
with 3 NP traits vop stong even in minor
knotty oriole
Sure but in that case why Exsanguinate? Surely a 3x NP setup would prefer TB?
solar swift
ye, sure, but if i want play 3 np 3 ss?))
knotty oriole
Specifically referring to the Exsang setups here because the Vanish sync logic is a big part of it.
golden briar
@knotty oriole if not enough energy in subterfuge, why not just pool for a few sec before the 2nd garrote?
knotty oriole
Yes that would be slightly more optimal but I didn’t see any particular gain when I tested a larger pool value for the Vanish.
Was something I looked at but it’s quite rare
golden briar
probably because it doesn't happen often enough in sims with a lot of iterations
but on a single iteration could make a significant difference
if you could somehow filter the iterations that had that energy issue
knotty oriole
Well I wouldn’t say significant. At worse you are losing 2/3 of a Mutilate on a build where they don’t do much
golden briar
hmm true
but casting that 2nd garrote is literally just a waste of energy
if there are 3
fervent onyx
is it because vendetta gives back energy its trying to dump it before vendetta?
knotty oriole
It’s best not to focus on super edge cases because that’s what iterations are for. And also because the impact is very low compared to other forms of RNG.
Over a normal length sim this sequence would only account for like 50 DPS at best.
When you consider RNG with procs can swing iteration DPS by up to 50k damage this starts looking a little less important. 😉
And when I say 50 dps I mean that iteration itself. Over the full sim it is nothing.
knotty oriole
By the way, looking into this issue some more from above @solar swift it's definitely a time boundary issue since it is a -0.6% loss on specific fight lengths but a gain in others. I'm definitely going to have to investigate more specific syncing logic for specific fight durations with VoP.
Although only really seems to be an issue with SS+Subterfuge
solar swift
Yea, need sync vendetta/vop with TTD, you right, but why sim doesn't use last exsang in 3:20 fight in some iterations (or use it on ~3:08), i don't find answer
knotty oriole
I have some expression code I use for Guardian of Azeroth that I can likely apply here too
actions.essences+=/guardian_of_azeroth,if=floor((target.time_to_die-30)%cooldown)>floor((target.time_to_die-30-cooldown.vendetta.remains)%cooldown)
Which likely would become something like some variable to track VoP waste like:
value=floor((target.time_to_die-20)%cooldown.vendetta)>floor((target.time_to_die-20-cooldown.vanish.remains)%cooldown.vendetta)
or something along those lines
Basically attempts to sync them if possible
Unless it will result in wasting something
And toss it into a variable
But I will test some stuff later
solar swift
also if delete subterfuge condition, sometimes sim used bote before exsang coz vendetta debuff less than 10 sec
need sync vendetta with exsang too i think
i used exsang cd < 3 sec
knotty oriole
I'd have to actually test to see if that is a gain or not
harsh basin
I noticed the APL was delaying GCB in my sims quite significantly, so I decided to mess around with it to see how much of an increase it was. Changing it to also use GCB if Exsanguinate was on cooldown for more than about 20s ended up being a decent damage increase.
https://www.raidbots.com/simbot/report/tqbNVDj5ZsdiTMLYMCwpAZ
I wasn't sure if it was just my gear, so I tried running it on the base T25 profile with the same essences, trinkets and talents. It ended up being a small decrease.
https://www.raidbots.com/simbot/report/et5WwfLJCpykpBR198eea1
But if you include 1 Shrouded Suffocation it ends up being a decent increase again.
https://www.raidbots.com/simbot/report/dE55ccbK7NaT3KQQdRTbM6
I'm not sure exactly the reasons it does, but there's probably a more optimal way to use it than the APL is currently with SS.
knotty oriole
Yeah I did an optimization pass with GCB a little bit ago but could be some gear setups where it still could be improved. Although if it’s just one usage shift it could just be a fight length thing.
But I will take a look at this and see what’s up. Thanks!
(0.3% does seem small enough it could just be an end of fight thing though. The Exsang syncing was a pretty huge gain when I added it. Think it was 2% higher in a general case.)
harsh basin
yeah my guess it'd be just being able to fit an extra GCB cast in the 5 min window, but figured it was worth mentioning since it made GCB a lot more competitive among trinkets for exsang for me
knotty oriole
Looks to me it’s 4.2 vs 5 uses so probably just fitting one in at the end of the fight length. I think as a general rule the current logic probably is more DPS but I can look to add a check about if it will fit one more in. Also could be due to the ilevel of your trinket since it’s 475 that this becomes a little higher.
harsh basin
it's still about the same increase if I check it with a 465 trinket so I don't think it's just the ilevel, but yeah if it's not too hard to add a check for it
solar swift
@knotty oriole exsang logic still need update, https://www.raidbots.com/simbot/report/ocSqJuWU1eyP6Wh8AegmLU/simc
1: coral can use with first garrote, not on 4sec, its possible gives extra stacks
2: garrote-muti-garrote after each vanish (I think it's better to wait for energy than to use muti)
3: total desync vanish/exsang/bote after 4+ min fights (vanish on 4:28, bote 4:41, exsang 4:44)
regal agate
#1 seems to be related to subterfuge, its def. a good find
knotty oriole
@solar swift Not really sure about some of these notes. Vanish and Exsang don’t always need to be synced. It actually doesn’t affect the number of Subterfuge ticks in this case. Also Garrote Mutilate Garrote was tested as being a gain. If you think waiting is better, I’m not totally sure why. And it would be tiny at best. BotE and Exsang not overlapping perfectly on the last one is due to Vendetta duration (10s remaining) and the total length of the Crit buff, not just the 5s buff. Doubt shifting this would make a DPS difference for this particular timing offset.
The issue here if anything would be that potentially you could shift Vendetta to be later in an end of fight situation possibly.
Although looking at this sequence I don’t actually think the DPS difference would be that significant.
Razor Coral I think can be fixed. It’s mostly just an artifact of most CDs and trinkets being on the GCD down shifting the CD list after the Stealth list. However since Coral is off-GCD as an exception I can probably move it out.
But for say the Garrote Mutilate Garrote sequence, I’m not totally sure why you would wait. It’s just a waste of Energy to do nothing. You can see in all these cases that the player in this case is at 95-100 Energy after the following Vendetta GCD so there’s no advantage to pooling there that I can see.
solar swift
But gar-mut-gar lose 2-4cp, mb better use vanish with high cp (vanish - envenom - gar - gar - vend - rapture)?
knotty oriole
That is a possibility but generally not worth delaying things for.
I tested it a while ago and didn’t see a gain
Since it makes the timing far more strict
In an ideal world that would be fine though
Although the damage probably offsets in potentially fitting an extra Envenom into Vendetta.
golden briar
@knotty oriole btw, i know it's not too relevant, but could you look into font+gcb usage for exsang? i think prefonting at 14 sec so you have gcb for exsang is a dps gain, but i'm bad at apl
knotty oriole
Quite possible
There is only really special logic for Font + Coral but I'll check GCB
golden briar
at least on pull, just force a 14 sec prefont
plucky mural
you talking about font prepull so that you can exang the vanish?
golden briar
i'm talking about font 14 sec before pull so you can use galecaller for exsang
drifting shore
Could you test whether it's worth having 2 envenoms in bote window when running exsan bote? Instead of gar rup gar ven mut rup exsan bote mut mut, gar rup gar ven mut rup exsan mut mut bote? Trade off would be 1-2 ticks not having increased crit chance for having one more empowered envenom into bote
knotty oriole
I can check but seems counter productive.
Envenom isn’t a huge source of damage for this build and the difference of 20% of one vs. a Mutilate seems pretty tiny. I think it should be easy to check though
drifting shore
Muti is around 3% of the damage, Envenom around 11%
knotty oriole
"Trade off would be 1-2 ticks not having increased crit chance"
1-2 Vendetta Exsang ticks doing ~25% less damage vs. (20% of the damage of an Envenom crit - 20% of the damage of a Mutilate crit)
Seems unlikely to come ahead
drifting shore
Is it 25% less damage in theory?
knotty oriole
I mean it's simplified since I'm not adjusting either side for crit chance. It's a 15% relative increase vs typical base crit
But your Envenoms aren't all going to crit either
drifting shore
Ye true
knotty oriole
Also 3% for Mutilate doesn't seem right to me? Mutilate should be around 8%?
Looking at Raz's profile for example
drifting shore
I was using https://www.warcraftlogs.com/reports/CrJGxcDbWA2wPy4n#fight=14&type=damage-done&source=91 a patchwerk hc maut as reference from yesterday
mut is around 2.68 envenom 14.26
drifting shore
Oh shit.
knotty oriole
But anyway, the DPET on Mutilate is around 20k vs. 60k for Envenom generalizing a little bit
drifting shore
Then It's probably not a gain
I thought mut is this low
So it felt like shit casting it during bote
knotty oriole
So basically you are trading for about 20k extra damage during blood from a crit, times crit chance (let's estimate 65-70% during Blood)
So about 13-15k extra from Envenom vs. Mutilate
But potentially lose crits from SS/Rupture
I suspect it probably will wash out to be pretty neutral
I think probably on average you'd lose ~6k expected damage from every SS tick due to not having the crit buff
Anyway I'll try it later when I have some time, it's a simple APL change
drifting shore
Alright 🙂
west prism
Is it possible to get any sim on the dps loss/gain on holding BotE for high rank tendie procs
(t6+)
maiden cargo
@west prism this is rank 5 TA but it's already a small loss https://www.raidbots.com/simbot/report/oW8K8tVKsrVZ2iaajZoox
west prism
Alright, thanks
knotty oriole
Seems pretty high variance also to hold a cooldown for an RNG proc that may not come. Like technically if you wanted to mega fish for ranks it could be some idea to hold CDs for some triple overlapping TA proc or something, but the average outcome of that approach would be very poor
And other negatives there with missing out on other preferential alignment
muted osprey
@rigid meteor That's not really what this channel is for, you can ask such questions in spec channels this is more about APL/rotational changes or math based discussions about the spec(s) 🙂
golden briar
@knotty oriole btw forgot to bother you about this, but can you look into g-r-vend-g opener for SS+subt?
i doubt it will make a big difference but it's an extra global (hasted) of garrote in the opener
knotty oriole
I'm not totally sure I follow
golden briar
in short - vend global is hasted, with hero on pull it's ~0.8 sec and 2nd gar is pretty safely in subterfuge window if you're not running around having trouble hitting the boss
knotty oriole
Right, without lust it's gonna be a lot tighter right?
I mean it's fine for the sims with lust on pull but probably worth keeping in mind right?
For the fights without lust on pull
I can take a look after raid
golden briar
yea, i'm just curious if it's a gain with lust on pull
i would test it without lust too but i'd probably be too scared to try it in practice
knotty oriole
Some really funny things happen when building corruption sets up from the ground up to test things
Casual 16% jump from adding one trait sometimes 🤣
zinc pecan
hod 
knotty oriole
yeah it's just amusing 😛
It's actually a real problem we'd run into with pure GW stacking if that was still athing
Since we'd drop below 25 pretty soon. Would have to add fillers back in anyway eventually
solar swift
gar - rup - vend - gar opener possible even with 7% haste (haste only from neck + goblin passive, w/o neck - can't do this opener), w/o any haste procs, so don't need lust 🙂 but idk how check this in custom, sim use only 3 abilities in subt window 😦
vagrant quail
Im not a pro simming mainly using raidbots but it's possible to make a script that tests every possible combination of corruption on the equiped gear? no gear change only what you have.
something like the top gear from raidbots. TOP corruptions.
setting max resist to 135 and corruption to 39/59
unkempt crescent
Sim with 160 corruption and subtract the 135 in your mind
warm roost
@solar swift the way that works is you actually just need more haste than 0
cuz vendetta's gcd is shortened by any amount of haste
knotty oriole
Technically yes but there is still latency to factor as to if you get the 4th global in before the server thinks Subterfuge is expired.
The potential DPS loss here for overwriting a buffed Garrote is quite high unless it’s a comfortable window.
regal agate
200ms sounds very short esp. when not hitting every gcd precisely
golden briar
idk, i've done it 100% consistently over an entire raidnight
knotty oriole
I'll test this tonight when I'm done grinding for Expedient IIIs...
But I'm still unsure where the gain is supposed to come from here exactly
As far as I can tell with my Haste levels anyway, the only thing that happens in that single global window is 1 tick of Rupture and possibly a melee swing
tiny storm
yeah i'm with u, i assume you still want the first 3p rupt for pandemic on the rupt before exsang
just saying its fairly possible to execute as long as u have 100% haste on pull
i suppose it can allow for 4 subt garrotes in aoe pulls tho
knotty oriole
yeah that could be pretty interesting for sure
That is actually possible sometimes if you get aura lag
plucky mural
@knotty oriole yeah i tried for that 4th global a long time ago, couldnt hit it but this was probably 96% haste
not worth imo.
bust that subterfuge and cry.
golden briar
you can't get 4 non-vend globals in subterfuge without lag
the idea of putting vend into subterfuge is because that delays 2nd garrote, thus increasing the total duration of empowered garrote by probably 1 tick
in the standard opener, you get a full 23.4 sec garrote at 2 sec into the fight, whereas with vend as 3rd global, you get it ~2.8 sec into the fight, and the number of globals between garrote and exsang is reduced, so exsang's overall effectiveness is increased slightly
@tiny storm rogue globals (other than vendetta) aren't hasted, they're all 1 sec
tiny storm
whelp sucks if thats true, i thought in legion they made .75 the minimum global when 100% hasted but could be wrong
raw knoll
rogue are hard set at 1
golden briar
that's for other classes, rogues have always had a 1 sec gcd
not just rogues, i think all energy classes have it
golden briar
i think ferals too
knotty oriole
Yes Ferals also
golden briar
either way, i'm fully aware that this is a minor gain if any, i just wanna see if it's a gain at all 😄
knotty oriole
Had to make a code change to test this since current code wasn't letting Vendetta go below 1s, but testing it locally now
Preliminary test seems to show it being within error margin at 0.3% error, so running at 0.1% to see for sure
DPS Error: 105.2 / 0.099%
So if anything this is a slight loss
304 DPS behind with a +/- error margin of 210
So, sorry @golden briar 😄
wispy badger
frosty pebble
"if u trust"
knotty oriole
With Exsang we ideally Guardian prepull so that you get a 5 stack before your first Exsang
gusty ridge
can someone check if using formless void major would be a gain for some fight lengths. With a good amount of expedients that you can now get formless void minor starts being better than lucid as well as guardian major and with some fights lengths that you don't miss a cast of guardian you can get formless void as major and get a better minor than clf (something like CnS for example)
knotty oriole
Formless Void is not simmable right now in any usable way because of the difficulty of implementing it
I suspect there are definitely some fight timelines where it makes sense if you can get another CLF user in your group
There are definitely some fight lengths where you could get Blood+CLF when otherwise you would only get CLF
gusty ridge
I see, I was hoping that something custom could be made like just 'disguising' it as a clf with a longer cooldown and changing the clf minor to formless void one or something like that
knotty oriole
It's possible I can look into it again
General consensus among SimC devs seemed to be that it would be a lot of work
But I can take a peek
gusty ridge
alright
having no idea how the essences are implemented I would imagine one implementation being when you pick it as major it changes to all the major essences you have chosen in the sim with the longer cooldown
knotty oriole
But yeah it's definitely possible it becomes better at all sub-3 minutes since technically the minor is a little stronger
Although the difference would be really small
Have to have someone in the raid to copy though
knotty oriole
Yeah true
golden briar
the problem with formless and bote
is that bote's base cd is 2 min
on rank 1
and it interacts with the base cd
so you can't just steal blood and align it with vendetta every time
knotty oriole
Ah it doesn't get the CDR?
golden briar
nope
knotty oriole
Ok, wasn't aware of that. Haven't actually tried it in-game 😛
golden briar
i've tried it on n'zoth
knotty oriole
This is another reason it would be really hard to implement in SimC 😦
Would need to test all the rank interactions
golden briar
well, generally just take r1's CD and use the essence's effect at r3
knotty oriole
Sounds like a bug tbh lol
But alright yeah, that sounds less useful then unfortunately
The short CD ones probably are not worth copying because we get pretty GCD saturated with high Haste
And nobody runs the niche ones which would be kinda interesting to toss in as short fillers. (Unbound Force or Iris)
Thinking about it, WV Major could actually be the most abusable if anyone actually ran the Major.
Could easily run WV Major for first Vendetta then CLF for second Vendetta.. that would actually be quite strong if you could talk people into supporting you 😛
golden briar
hmm, that's actually a pretty good idea
gonna give it a try on thursday
if any1 in the guild wants to try it ofc 😄
knotty oriole
So fun-fact looking at the higher haste builds. It seems like they can reach a haste level that actually allows Crimson Tempest to beat out Poison Bomb in ST. By about 0.4% or so, so it's minor... but considering it's quite a bit stronger in AoE/cleave, that is a pretty interesting breakpoint. I'm not quite there with my personal gear yet but seeing it probably around the 100k DPS mark levels of Haste.
Will be updating the APL probably tomorrow to account for this so that people can check with their gear.
solar swift
0.6% gain only with CT if <2sec and rupture ticking and cp >5
mb can do more by optimizing opener
knotty oriole
Line I am testing atm is
actions.dot+=/crimson_tempest,if=spell_targets=1&combo_points>=(cp_max_spend-1)&refreshable&!exsanguinated&!debuff.toxic_blade.up&master_assassin_remains=0&!azerite.twist_the_knife.enabled
solar swift
need test with mfd) like mfd prepull - rapture - g1 - g2 - vend - ct - exs - blood 🙂
plucky mural
Mark is not going to be a st trait.
solar swift
same as CT)
knotty oriole
Deeper is a pretty big gain for Exsang
Don't really think MfD will be able to be competitive
solar swift
@knotty oriole so can u link edited apl for ct?
drifting shore
that's already 1.5k dps upgrade for me lmao
knotty oriole
😮
It has some other minor changes to the Exsang setup as well
I'm finding that it is likely a gain to sequence it as Garrote->6 CP Rupture->Exsang as long as a Garrote cooldown is available
But I'm still trying some minor variants
inland herald
Garrote-> max cp rupture regardless of garrote duration into exsang
But rupture only when in pandemic?
knotty oriole
As Exsanguinate comes off cooldown you would use Rupture at max CP, doesn't need to be in pandemic.
feral bobcat
When talented into vigor, we want 5 cp rupture to refresh. What when mutilate set us on 4 cp, another muti or? Garrote on cd.
inland herald
@feral bobcat
- Pre-Exsanguinate it will attempt to Garrote -> Max CP Rupture -> Exsanguinate regardless of existing Garrote duration, if Garrote is available in ST and adds won't spawn within 6 seconds (minor 0.4-0.6% gain for ST Exsanguinate to do this when it aligns)
if you see exsang coming up most likely want to mutilate before you refresh garrote I'd guess, or might even be worth overcapping, will see what koji comes up with
harsh basin
Posted about this about a month ago but Galecaller's Boon being used more frequently is a decent increase (1%) at least for me, probably worth getting looked at again.
https://www.raidbots.com/simbot/report/wDDjshFReZu3ojT7NVizPd
Line I was using was
actions.cds+=/use_item,name=galecallers_boon,if=(debuff.vendetta.up|(!talent.exsanguinate.enabled&cooldown.vendetta.remains>45|talent.exsanguinate.enabled&(cooldown.exsanguinate.remains<6|(cooldown.exsanguinate.remains>20&target.time_to_die>60))))&!exsanguinated.rupture
but probably can be optimized further than that, it also sims equal or better on every fight length I've tried so far.
opaque sky
Does anyone have an image for opener using CT?
knotty oriole
@harsh basin That is a different line than the one you posted before isn't it? The one from a few weeks back seemed like a loss at some fight lengths and talent setups. But this looks a bit different?
I did not put in the previous one because it was a loss in some situations
harsh basin
@knotty oriole yeah I added the time to die check because it was a loss at 2 minutes otherwise, I could mess around with different timings for it but I figured someone else would have better knowledge on how to optimize it.
maiden cargo
exsang affects ct, right? did you test casting ct before exsang? atm it's casting after exsang it seems
knotty oriole
Yes and yes
It's not worth adding to the setup
If it happens to catch it, that's fine
plucky mural
@knotty oriole with all the energy loss in the high haste openers, do you think Max cp spenders will be a gain? Even if that means mut on 5 (with DS). It would be hard to condition Egain per second but maybe within the first x seconds of vendetta.
Idk on 5cp env. Vs the global delay of the mut env
Stuff probably gets weird with infinite Energy windows.
plucky mural
It's probably insignificant compared to a single corruption proc.
knotty oriole
@plucky mural hard to say but I think the CP waste would be a concern. It's probably only fine to "waste" Energy if it doesn't also mean wasting CP. (e.g. a low CP Rupture refresh is not a problem for this reason, since it's not CP-inefficient)
Using non-max CP finishers is generally just energy inefficient, while being CP neutral or sometimes even CP efficient
solar swift
Why now sim use CT before exsang in opener? its better than default exsang opener? i think its bad idea delay exsang for 5sec
drifting shore
@solar swift it does it if the first mutilate doesn't reward 6 CP
5 CP -> CT, 6 CP-> Rup Exsan
solar swift
so mb better always using CT before exsang on opener, even with 6cp? or vice versa never use on opener, coz not all empowered bleeds get in burst window
drifting shore
im not sure if there's a way to filter the cases out (probably Koji knows) but it'd be interesting to see the descriptive statistics of the distributions with 5CP -> CT and 6CP -> Rup Exsan sims
knotty oriole
I tried forcing one over the other and neither was a gain over the current behavior
solar swift
@knotty oriole can you please update default apl on git wiki page?
solar swift
i tested double use trinkets, like fire mage, font + gale, with 17 sec precast and condition with vanish/exsang instead with vendetta, and its better for me than raden
but need new apl for clear test
drifting shore
How big of an upgrade?
drifting shore
445 font?
solar swift
455
knotty oriole
I tried precast font last week and really didn't see any gain
With 445
It was neutral with basically anything else
solar swift
and second use?
knotty oriole
Yes, with GCB and Plumage for example
It's possible it's a gain with 455 specifically
I only looked at base ilevel since most people won't have 455s
solar swift
in default apl font used before vend, but need wth vanish exsang
knotty oriole
I don't believe I can check item ilevel in the apl
solar swift
so font, after 16 sec vanish -g-g-vend/gale-exs
golden briar
you can pre-cast font at 14 sec and have galecaller come up for exsang
i did ask you to test this @knotty oriole but you never did 😄
knotty oriole
I did
I tried precast font last week and really didn't see any gain
With 445
It was neutral with basically anything else
😛
golden briar
i meant you didn't do it when i asked like a month ago or so 😄
knotty oriole
Well, mostly because I didn't think it would be a gain with some quick napkin math 😉
The precombat time there has to be pretty high, it unfortunately makes the trinket a lot less valuable
Mythic Vita is so strong, it's gonna be hard to beat especially if you consider the group buff
prime terrace
hey, i was hoping someone could help me understand simc reports a little better:
https://www.raidbots.com/simbot/report/2mxLz8w5SkmVoq1nsMHLU2
i'm not really curious on the dps gains themselves (because i do not think the changes indicate good play, the entire 1k gain is purely bote cast damage) but rather why the report for the default blood shows so weirdly - the modified version more or less aligns casts with iris as expected but the default is doing.. damage randomly even before bote could possibly be off cd?
maybe it's not weird and i just don't understand the report, i know if abilities are condensed for example it wouldn't bottom out but that shouldn't be possible given botes lengthy cd, but i'm sure someone can clarify
boreal patrol
These graphs should not be interpreted as the curve for one fight, meaning there's no random damage while on cd. It's based on all iterations, i.e. big bump more iterations count into it and smaller bumps are distributed with more variance.
In this case, the modified version aligns it more consistently while default seems to have some breakpoints, probably related to maintenance stuff and TTDs.
knotty oriole
@prime terrace yes basically because of your spell_targets.blade_flurry>=2 condition it's basically gonna be used on CD in DS
Instead of waiting for Flurry to be up at the same time as BtE which is going to be a non-determinate timing
This is the type of APL change that people will hate though because they will post sequence screenshots of "I can't believe it uses Blood without fitting BtE in" 😄
prime terrace
i see, makes sense thanks, i'm not sure why i initially interpreted it as a single iteration breakdown
i guess using these as a base if i wanted to improve it on my own, the best approach initially would be either to find a way to align it similarly the no-maintenance version to preserve cast damage gains with some basic checks to get better 'uses' without delaying far enough into a 'worse' pack OR add something way earlier on the bf sync to prevent drying up and causing weird timing issues
my original thought was that the small 15 second packs were causing larger than ideal delays due to convoluted setup where it could chain into a single big add pack on a decent chunk of iterations causing it to be delayed even further (https://www.raidbots.com/simbot/report/uo18W9P7WhuiLqvvr5TBru example sim, bigadd increased to 2 minimum and cycle extended a bit to allow breathing room for the setup - it makes sense after the explanation that the bumps here are more spread than firing on cd as there is still some variance in when the cast goes off iteration to iteration, but way more dense than the original because the delays aren't as long)
i'll continue to tinker around with it and see where i end up, which is probably the same place i started but.. learnings! thanks again
knotty oriole
I don't think it's getting chain "delayed"
The use count is almost exactly the same
I think it's more that this just ends up aligning it on all the big packs more reliably
Although I'd say the APL logic does this rather accidentally
Since it's not specifically trying to do it
But just due to the cadence of the DS waves ends up doing it with those particular 60s vs. 2+ target conditions
(e.g. if you reversed the wave spawn ordering I'm not sure the results would be the same)
But I think more to the point is likely what this is showing that the AoE damage is potentially more impactful than min-maxing the crit bonus
Which is not really too surprising to me
Because Seething Rage is way weaker than people think and the Crit% buff is pretty much weaker and weaker by the day with Severe setups...
The AoE damage is the only component of the Major that scales with neck level as well
As a note, (and maybe this is what put you on this if Guy mentioned to you,) was investigating over the last week something similar with a small gain for removing some conditions on Blood for DS but still was trying to track down something repeatable/generic and not just it overfitting the fight timings.
prime terrace
i've tried a decent variety of pack sizes/lengths and generally my findings were that the end result was roughly the same, but i do agree that a large part of it is cast damage related and the actual gcds inside seething aren't too impactful (though it'd be better to have them good than bad, if there are no losses elsewhere for it)
and yeah i know any ''''change'''' i make is usually very narrow on what i'm looking at specifically and never very cognisant of other styles/profiles/lengths, i'll leave the complicated stuff up to the smort people i'm just here to break things
knotty oriole
That's cool, just wanted to match up my findings on it also for your reference
I honestly don't think it's unreasonable for AoE fight styles to consider just optimizing more towards the AoE damage component itself
Even if it means people will post crying snippets of "scuffed" Blood usage every day or two 😉
This likely is going to get more impactful the higher neck levels go
This whole thing is definitely a moving target in that regard
Iris vs. default Blood logic for example has something like a 2.5-3% DPS delta in the last 10 neck levels (from Iris being behind to being ahead)
So this probably has to be revisited like once a month or so lol
prime terrace
perfect!
knotty oriole
Suppose we could make some profiles with 100 neck level and see what it looks like in advance
maiden cargo
100 Neck Level without AK is gonna take ages to reach.
With how few people bother to grind AP even now, and interest in the Expansion dropping off (especially once Beta comes around), I'd wager that this would be largely wasted effort
Didn't do the math on how long exactly 100 Neck will take though for average AP farmers like me. Might be wrong
knotty oriole
Gonna take a while for sure although a lot of people at 99 now so it's not too crazy
In a month or so it won't be too uncommon
Too far beyond that probably will slow down a ton though since it's not linear
My raid is mostly in the 92-94 range atm
finite lodge
There's no more artifact knowledge so if you ain't actively getting AP you're probably stuck where you're at
hardy kindle
does anybody know how to multiplie the mastery amps work? is there a dimishing return?
noble pendant
They’re additive iirc
finite lodge
is the exsang opener still 5 cp -> CT / 6 cp -> rupture?
tardy current
Thats what raz said on stream 1.5 weeks ago i think
think it should still work out that way right?
Since he was already rocking as good as bis
knotty oriole
Yes there is no difference there
rustic ibex
Did some1 try out aligning galecaller with exsang and vendetta? Because it seems to not do it right now and I find it odd.
Like I'm running galecaller, worldvein and exsang build and it seems reasonable to me to allign those since you don't lose an extra cast on exsang considering the wait for vendetta anyway
raw knoll
scroll up and go back to end of may discussions.
knotty oriole
actions.cds+=/use_item,name=galecallers_boon,if=(debuff.vendetta.up|(!talent.exsanguinate.enabled&cooldown.vendetta.remains>45|talent.exsanguinate.enabled&(cooldown.exsanguinate.remains<6|cooldown.exsanguinate.remains>20&fight_remains>65)))&!exsanguinated.rupture
is the current line and has been recently tested
solar swift
https://www.warcraftlogs.com/reports/82CpRJZLK4X7wgGb#fight=37&type=damage-done formless void major, possible, but not perfect boss fight timer
mb possible steal breath/flame between wv and golem
knotty oriole
That's pretty neat 😄
drifting shore
i'd guess something like replicating bote for first vendetta and clf for second one would come out ahead on 3 min kill timers
Maybe even on 5 min kill timers, bote -> bote -> clf
golden briar
bote has 2 min base CD tho
drifting shore
does it take the 2 min or the 1.5 one at rank 3 ?
golden briar
2
it should still be good on a 3 min fight tho
if you yoink it >6 sec before the fight
you'll have it at 2.30
so you can use CLF then
drifting shore
yep
It'd overlap exactly with the 2nd vendetta
gonna try it today on some bosses, gotta check for ppl who run either of em in my raid
6s would be fine if you can skip CT in the opener, else I think you'll have to overcap
golden briar
ye actually dodgy 🤔
but idk, i like just stealing CLF and then having an extra good minor
finite lodge
If in the opener, it goes 5 CP -> CT, does it still try to get to 6 for rupture-> exsang?
Been spamming quick sim but I only ever see 2 muts in the example, and it doesn't show CP of course
knotty oriole
Wouldn't BotE go on a 3 minute cooldown with Formless if it uses the 2 minute base? Seems a little pointless since that just makes it the same cooldown as CLF anyway. I think BotE would be good in this situation if it used the 90s base because it would be up soon enough to be useful at 135s
If it's not, probably WV is the only real strong option to fit in before the 2nd Vendetta
@finite lodge yes it will always go for a 6 CP Rupture in that condition with Exsang off cooldown
weary scroll
I found a possible weird action in dungeon slice and restealthing for outlaw https://www.raidbots.com/simbot/report/azosvEnn7RF6ktDeu6uifL
In the no restealth sim, I removed actions="stealth"
knotty oriole
Looked into it briefly and it looks like AAs are just delayed getting restarted coming out of Stealth and that is the only cause of the difference. So it isn't "real" in that it won't really affect anything other than the final number. But I'll see if I can identify the root cause.
Not really an APL issue, just some SimC code issue of some sort
Most of the difference is just in AA count
Which has other side-effects on procs and such, basically just removing combat time
So you can just think of it as one profile has like 1s more combat time every pack than the other profile. Doesn't really change what it is doing or any calculations, just gets some more combat time which adds up.
empty vortex
@knotty oriole if you land on 5 cp again after opener would you fill last point with pk fok or just mut and then 6 cp rupture
knotty oriole
Just Mutilate should be fine, I mean no reason to change anything too much there, Energy during the opener isn't usually a concern
Pretty much capped on Energy most of the time
knotty oriole
@weary scroll yeah, thanks for the report. I have a fix for the issue in code however how it works with Stealth in the APL is actually correct
Removing Stealth from the APL was causing it to be allowed to auto-attack against the invulnerable main target between waves
Which lead to better proc uptimes, and more energy regen via MG
I am changing it to correctly cancel the AA events when the wave completes
Regardless of if Stealth is specified
golden briar
@knotty oriole formless r3 increases CD by 30%, not 50
so it's 2m 36s for bote
so if you steal it >6 sec before the fight on a 3 min fight, you can steal CLF for the last 30 sec
but you really have to know your killtimes
knotty oriole
Ah ok, I don't know how I missed that R3 reduced the penalty
Haven't used it much
Makes sense
knotty oriole
Gotcha, my bad 😄
golden briar
@knotty oriole reminder to try hacking bote 🙂
gusty ridge
I was trying to sync up bote on 3 min fights with second cooldowns better and came up with this https://www.raidbots.com/simbot/report/ozCiZXq2yJ7GM7jvwjtBgP
I don't know if I skewed the results somehow but I added a condition for vendetta and vanish like this
actions.cds+="/variable,name=ss_vanish_condition,value=azerite.shrouded_suffocation.enabled&(non_ss_buffed_targets>=1|spell_targets.fan_of_knives=3)&(ss_buffed_targets_above_pandemic=0|spell_targets.fan_of_knives>=6)&(cooldown.exsanguinate.remains<5|debuff.vendetta.up)"
actions.cds+="/variable,name=vendetta_subterfuge_condition,value=!talent.subterfuge.enabled|!azerite.shrouded_suffocation.enabled|dot.garrote.pmultiplier>1&(spell_targets.fan_of_knives<6|!cooldown.vanish.up)&cooldown.exsanguinate.remains<5"
also added target.time_to_die<20 for vendetta
seems to be neutral for 5 min for my profile
and I don't know if the vendetta exsang condition is necessary cuz it doesn't seem to do much if I remove it
and its probably not the best way to do it, but maybe something could be made off that if its any good
golden briar
definitely worth syncing everything on a 3min fight cuz you don't lose any casts, but you kinda have to sim with fixed times
gusty ridge
it was neutral for me on 5 min
golden briar
well you don't lose much on a 5min either, but i think the apl needs to be applicable to all fight lengths 😄
gusty ridge
yea, I was specifically trying it for 3 min, but maybe something could be done so it doesn't screw the other fight lengths
golden briar
i'm all for syncing all 4 CDs around 2.25
but you gotta know your fight lengths to do that 😄
knotty oriole
I don’t think I would frame this as “BotE sync” just to be pedantic
This is more gains for shifting CD in general. The gain is identical without BotE major looks like.
More just a general short fight optimization
gusty ridge
exsang sync then I guess 😄
knotty oriole
E.g. replaced BotE major with Formless major and the gain was the same
As a percentage
So yeah this is cool I was wanting to look into this over the weekend anyway
In terms of a shifted CD alignment if it wouldn’t result in a usage loss
gusty ridge
yea, and what I did was probably not the best way to do it
knotty oriole
I’ll take a closer look in a bit
gusty ridge
so I assume it could be optimized further
knotty oriole
Doing some Father’s Day stuff 😄
Logically it makes sense to try to capture that Exsanguinate in CDs as long as you won’t lose a CD cycle. In practice it will take some knowledge of fight lengths but that’s fine.
I need to try to do something similar for CLF alignment on fight lengths between 3-4 minutes for example.
(Right now CLF will delay for Vendetta if it can but not the other way around)
gusty ridge
yes, I was thinking about that as well
but I felt like more fights are around the 3 min mark or less
knotty oriole
It was only recently I was able to get things set up to allow delaying Vanish without everything puking. There used to be some Exsang setup lines that were very connected to Vanish CD state
Although to do this properly probably means extracting out the timing to a variable
golden briar
also, on the topic of CD alignment
knotty oriole
I suspect delaying could do some funky things to the trinket lines still
Like Font and such
Don’t think most people use Font any more but I don’t like breaking things 😛
gusty ridge
I've seen some people use it, mostly on shorter fights
golden briar
when wearing GCB, you can delay 2nd one to use it 8-9 sec before 3rd exsang and then use all cds at ~2.30 again without losing any casts, thus having GCB on 3/4 of your exsangs
knotty oriole
How much do you end up delaying there?
golden briar
2nd GCB goes around 1.27-1.30
so you have it up for 2.30
basically if you cast it 8-9 sec before 3rd exsang, it comes off cd 6-7 sec after 4th exsang, so you just delay the vendetta/vanish/exsang for the gcb as you're not losing casts anyway
the idea is that you have the very end of the gcb buff for your 3rd exsg
knotty oriole
Yeah that starts getting into pretty tricky territory but might be possible if I set up the Vanish delay logic the right way.
golden briar
but i feel like it's extremely specific to fights between 2.50 and 3 min 😄
also, try hacking bote for the formless minor buff 😄
knotty oriole
Don’t worry we will have to start optimizing for 1 minute fights at this rate 😄
Maut getting down sub 2 minutes now
golden briar
maut goes sub 2 for groups that don't transition him
knotty oriole
Xanesh down sub 3 minutes pretty easily too. Getting kinda dumb lol
Yeah no-shield on Maut isn’t that hard now honestly. With full vendor gear comp.
Just have to make sure he doesn’t get any mana lol
knotty oriole
@gusty ridge Reviewed this more at my PC. I think the general idea is fine but I'm gonna change a couple things so that I can toss a projected Vendetta/Vanish time into a variable for other things to check
Otherwise the other CD alignment checks (like for CLF) will get confused
gusty ridge
aight
knotty oriole
But it's good to see what I can aim for as as a gain for this in terms of my version regardless
gusty ridge
I did it to get a base idea
knotty oriole
Yep, the research is appreciated!
I should have time to work something up this evening I think.
maiden cargo
smn checked if it's worth holding first exsang in the opener for the haste proc from blood minor?
with lust and focussed resolve it usually seems to be like 2sec after i press my first exsang i think
knotty oriole
I’ve tried waiting for all manner of procs and generally doesn’t seem worth it. Likely because the Garrote gets reapplied so quickly anyway it just becomes CD loss.
maiden cargo
Holding 2sec for guaranteed 820haste (I think? Before expedient) sounded worth to me at a glance, but I If its been checked thats unlucky
Ty for the effort
knotty oriole
Well it's not totally deterministic, it's 8-12s on average. I think you'd have to delay at least 2s as you said. Perhaps it could fit in with forcing CT as the initial cast instead of Rupture with BotE minor, that isn't something I tested specifically with Blood minor. I can check that.
warm roost
https://i.imgur.com/IlXSQ65.png can someone explain what is going on here
this looks a little broken to me
how is it randomly getting a restealth on this boss after casting 2 abilities on it
golden briar
is it meld by any chance being shown wrong?
knotty oriole
Possible. I've never really seen that personally but I wouldn't worry overly about it since this sequence would really make no difference in the overall DPS. There is a slight delay in when Fluffy Pillow is invulnerable and when the Boss1 is spawned so that could be contributing but dunno at a glance.
Not seeing anything else strange with it so almost certainly due to the initial spawn. Might debug it later just out of curiosity but this basically won't affect anything so wouldn't be concerned with it
lean talon
So i just noticed that "rolling" your coral buff doesnt work correctly in dungeonslice sims. The way it uses coral there is it stacks it really high on the boss fight part of the sim (14 stacks in my case) and immediately goes into AOE with it. it then extends it in AOE by applying it to a mob and having it die before the buff runs out. In my case it got 4 more stacks on the mob, meaning it got a crit buff of 18 coral stacks when it died for another 20 seconds. In the actual game, "rolling" your coral stacks doesnt work this way. When you have 14 coral stacks and kill a mob that has 4 stacks on it before your buff runs out, you get the equivalent of a 2 coral stack buff (so 240 crit instead of 2160 crit with my 455 coral). I tested this in game before, its not just a tooltip error, you actually always get 2 stacks worth of crit when you "roll" your coral buff like the sim does it.
Even if this was how it worked, I think the way it uses it would be super different from how you can use it in actual game scenarios.
you see it going from 14 coral stacks to 18 here
knotty oriole
The way the stacks are represented in sims aren’t the same as in game
The stacks are used for tracking the buff multipliers. The sims actually have zero support for the in game stacking bug.
As such the sims are under representing the potential power a little bit
But I’ll look into this particular case
I don’t think you are quite correct about the in-game stacking behavior though. If you kill a mob with 4 stacks and the buff refreshes you get the equivalent of an 8 stack iirc. The buff Crit value will refresh to the last known killed value of 4, multiplied by the buff stack (of 2 in your example.)
In-game the behavior should basically just be stacks on the last mob killed * visual buff stacks on your character.
lean talon
so even when it says 18 stacks there im not actually gaining 18x 120 crit? i am pretty confident that the way it works in game is how i described it. No matter how much crit your current buff had and no matter how many stacks the mob you killed had, you ALWAYS get 2 stacks worth of crit when you extend it
knotty oriole
No that is not correct as far as I last tested it, which was about 3 weeks ago
You should always get the Crit Rating relative to the stacks on the last mob you killed, which is then multiplied by the buff count
Now it may be that was 1-2 stacks because of the limited time-frame bit in your case
But if you do build up 4 stacks on the 3rd mob for example you will get a visual buff that has 3 stacks, with the crit value from 4 stacks, and thus have 12 stacks worth of Crit in-game
(Although this still can reduce your crit rating if you had a large initial buff)
lean talon
ah so if i had say a buff of a 10 coral stack and refreshed it by killing a mob with one coral stack i would get crit worth 2 coral stacks? since its 1 coral stack times 2 because its a 2 stack
knotty oriole
yes
lean talon
that might be it, maybe i always tried refreshing with just 1-2 coral stacks
knotty oriole
Yeah it is kinda hard to test assuming that the tooltip can't really be trusted lol
But I think I was able to confirm the tooltip was correct
At least in terms of the raw crit
I may dig into the SimC code for Coral and just try to implement the whole thing
I didn't implement is so I need to get up to speed on it