#tc-updates

1 messages ยท Page 4 of 1

rain lance
#

still though a WA could just track white hits mod 5

chrome sun
#

definitely

#

because you can scrape the combat log

#

i basically want it to play a small ding when it fills in a combo point (or two)

#

and let you know when you're likely to see one

#

slash guaranteed to see one

crisp basin
#

technically youre never guaranteed one

chrome sun
#

that's not what fiery's stuff looks like?

rain lance
#

because every next white swing could miss?

chrome sun
#

i mean, yeah, i guess you'd want the weak aura to say "next white swing hit will guarantee a proc"

#

that event could take arbitrarily long

rain lance
#

yeah, my plan is make a simple counting driver and then let people do what they want with it

chrome sun
#

i think it's like 16% chance to miss

#

from behind

#

right?

#

so you've got some geometry series

#

converges to

#

1.19 swings per hit?

#

1 / (1 - 0.16)

rain lance
#

yeah that looks right

chrome sun
#

*geometric, ugh can't type

acoustic lake
#

Problem with tracking ST is how do you track the procs that occur when you're full combo points with your weakaura so you can reset your swing count. Last I remember there was no combat event for it

chrome sun
#

there is; the shitty weakaura i wrote still pings if you are at full CP

#

you still get the spell_energize event

acoustic lake
#

must have changed since beta then

plush relic
#

is there currently an optimal way to use sprint with the treads?

#

ive been using it on cd right before my next run through

icy mango
#

@chrome sun do you have a link to it?

chrome sun
#

link to what?

#

just a test, i wrote a shitty weak aura that makes a noise when shadow techniques procs

#

dCJ(caGAarVeaSlQuVgazMukMlOy2cUjGWTbzNsP9I2nH9tj6NasdJsLFlQHcegmO0WfXbbiDmq1XbsTqQOLsPQftP0YLQhcq5PkpwspxLjcq1uPyYsmDOlcuxLQQlt66eDyH2mvLTlfFKsyzusnnauFhq1FfPtRQrtLmEaItsj5wakxda68uHNbeTwGKptvMWPHdIRWvOHlwXVrn93Zt7arUOSfyWTJd7VNN2PHRi7r8Zco7CdNROh6x8ixsi6q7CxIgcTriLBUILmWTewJwDjEql1JtEA6LL9Q2gHuAlxvEygk11xuo8H0c3jTvBuboPpFpGWLcUfS9wyfOaZjfAJkqo8H0cN2OcKdWYq2grlH1MShh6ir50gvGCYttVeneAJqkDYPnQapUbcW2ytk4wW2BHvGcmGZjpn9YYEv9lEKo5QYdZqPOJeLtBubYjpnTOh6x8iDYDzzVQ(fpYLeIo0orUqowOHdJbvG0Wbjd4tdrICxIwPH7EHxqzlijYjpn9s0kDYvEF((QmGoOHdsgWNgIe565GsdhKmGpnejYL0KGXGkqA4GKb8PHiroFXk(zbnCqYa(0qKixneVJgU7fEbLZEGAHFWe5KNMwdX7OtUESQ0Wbjd4tdrIe56Q3Nfghjk3xZcUVWnxXsg4wcRrRUepOL6XfR4NfhnChg7inCxwkkCwjqT7lwvya0sbOIdgq0JiWaPB7GzzPOCWaaXQRxaqUZEECLC6LLIcDsKiNqcz6VNN2p2cNR8hqkfv8BuUyiqK74S2T1UH7ga5AylCa2A4ejb

upper wave
#

Thanks for the WA, will test when i get home

icy mango
#

@chrome sun Yeah that's what I was asking for, sorry for not being clearer ^^

manic bramble
#

I'm pretty sure this was updated

#

It was higher before

#

as well

#

just not by much

untold laurel
#

Why does Master Assassin have a value of 0.00

#

Wouldn't multiple MA relics be really good with From the Shadows and in general for the burst energy etc

acoustic lake
#

this just in: stat weights are personal based on your gear

untold laurel
#

@acoustic lake Is that a response to my question?

acoustic lake
#

no

#

the guy above you, but someone deleted his useless question

crisp basin
#

hm looks like all combo point generators now have energize events even when used at full CPs

#

mustve been fixed since beta, definitely wasnt the case back then

stuck sapphire
#

there is no static stat priority, like you mentioned it is highly depending on your gear, so if you want to have them use simcraft

torn pond
#

did some of you drop Leg Cloak ? If yes how is it performing especially for Sub ?

stuck sapphire
#

Aethys did sims on legendarys, its in the faq

chrome sun
#

fun fact, that's the legendary that aethys has ;P

stuck sapphire
#

rng only gave me neck so far ::(

solid helm
#

does casting nightblade while master of sub 10% buff is up do more dmg or no?

fathom bramble
#

yes

#

until it fades

dull vapor
#

Until the Buff fade or the dot ?

fathom bramble
#

the buff

dull vapor
#

I miss bc for that :p

solid helm
#

is there any reason to try and cast nightblade during the buff?

#

or would it be stronger even if already applied on the target

elfin remnant
#

how i can simc trinket... like 3-4 trinkets what i have?

icy mango
shrewd pasture
#

do you still need to manually add artifacts to simc or is that fixt?

brazen tartan
#

fixed

verbal valley
#

Is there a benefit to using Potion of Old War outside Outlaw CD's?

cerulean plaza
#

So I just installed SimulationCraft and ran my numbers. Does the current version reflect the recent spec changes?

icy mango
#

703-03 has the most recent hotfixes

gilded stag
#

Anyone have any idea of the optimal usage of the dreadlord's deceit for sub in single target/ aoe, like what amount of stacks you should use it at and so?

upbeat bear
#

from what i gather from simc, you (and i) use shuriken only during shadow dance and at or above 29 stacks

#

or during stealth rather

#

for st

cobalt torrent
#

is the simc model for outlaw broken ? why does hast have such a high value.

real crag
#

Mine's always valued haste pretty well

acoustic lake
#

probably because you have low haste on your gear

#

simc probably values haste higher than you should in a realistic scenario because its a simulation that doesn't suffer from lag. In reality too much haste will have you capping energy in scenarios simc wouldn't

icy mango
#

I thought there was an option to set your lag on simc

acoustic lake
#

there is, but its not as simple as that in reality

snow terrace
#

The point where versatility passed haste for me was at 9%

icy mango
#

Why not @acoustic lake? It seems like it'd just add a simple delay to when the abilities are pressed based upon your latency?

acoustic lake
#

it doesn't work like that in game either, there is some innate queuing of abilities even without the lag tolerance turned on iirc

#

Even so, think about it like this. in game, with just over 10% haste you should be able to do 2x ss 1x eviscerate 2x ss without any energetic stabbing procs

#

But you can't

#

simc can

#

I haven't really played around with the simc lag setting before to know if changing that would then prevent it from doing that combo

brisk saddle
#

are the updated APLs live with the simc now?

lethal oasis
#

Anyone in here tonight?

acoustic lake
#

always people around

strong iris
#

any reason why I would sim higher with the fan of knives cape, than with the vendetta boots?

#

that makes zero sense

brisk saddle
#

What does remains<tick_time mean with the nightblade condition?

acoustic lake
#

tick_time is time between ticks of nightblade

#

so without seeing the rest of that line I would assume its telling the sim to refresh nightblade when there is less than 2 seconds remaining

brisk saddle
#

ticks are dynamical right? shorter ticks with haste, right?

#

well, pauses between ticks rather

acoustic lake
#

only for casters far as I know

brisk saddle
#

hm interesting.

#

while you're here, i took a look at the condition exanguinated. but i can't read c++ for shit. would you be so kind to explain to me how the APL evaluates it?

acoustic lake
#

what parts specifically

#

theres a lot of exsanguinate things in the apl

brisk saddle
#

sec, let me fetch a line

#

actions+=/call_action_list,name=exsang,if=dot.rupture.exsanguinated

#

as an example

acoustic lake
#

well thats going to call an action list if you've used exsang

brisk saddle
#

so it just tracks if we used exang?

#

exsanguinated <= this is what I'm interested in

acoustic lake
#

if you scroll down you'll see an action list called exsang

#

so when you've got exsang up

#

it will use that action list

brisk saddle
#

yea, but there is no exsang "buff" right? it just accelerates the up dots

acoustic lake
#

well its just tracking that rupture is exsanguinated

#

it can do that

#

just in a mythic+ hard to give good answers

brisk saddle
#

I think we're misunderstaning each other xD
I wanna know how the function works, as in how does the code behind the exsanguinated work

acoustic lake
#

oh well

#

i can look

#

after this mythic+, almost done

acoustic lake
#

@brisk saddle Looks like you need more than a decent understanding of simc to really get how this works in the background

#

bit beyond what I know personally

#

Is there a reason you want to know how this works beyond knowing that it works?

chrome sun
#

grep in the code base of engine/class_modules/s_rogue.cpp for exsanguinated and you'll see the implementation

#

*sc_rogue.cpp

#

correct, there is no buff, but it's easily trackable

#

ends up looking like

#
  {
    timespan_t duration = rogue_attack_t::composite_dot_duration( s );

    if ( cast_state( s ) -> exsanguinated )
    {
      duration *= 1.0 / ( 1.0 + p() -> talent.exsanguinate -> effectN( 1 ).percent() );
    }

    return duration;
  }```
acoustic lake
#

if the original question was just to undestand how the .exsanguinated apl expression worked it seems to me the jist of it is that in simc rupture has a struct containing its state. When exsanguinate is cast, any active garrotes or ruptures have their struct changed to evaluate true for exsanguinated

strong iris
#

hey guys quick question. For some reason, I sim much lower with one legendary than another, how would I go about working out why?

acoustic lake
#

which legendaries

strong iris
#

dawn walkers are simming lower than dreadlords deceit

#

vendetta reduction boots

#

are worse than fan of knives cape

#

What I don't understand is that if I look on the sim, fan of knives is about 2.8% of my damage

#

but I would expect much more than 2.8% dps increase from having vendetta on an average 50 second cooldown

acoustic lake
#

how often does it cast vendetta in the sim with the boots

strong iris
#

55 seconds

#

and 89 seconds without them

acoustic lake
#

pm me the sims you did and I will check

plush relic
#

bump

echo wagon
#

i assume its an issue of not having vendetta reduc relics, since then the cloak desyncs vendetta from every 2nd exsang/kingsbane

#

try with agop maybe

strong iris
#

wouldn't think so

#

I have one of those relics

#

means vendetta pretty much sincs up with my first 4 exsgs

echo wagon
#

how can it sync when you say it casts it every 55 seconds

brisk saddle
#

I just wamted to understand how it evaluates that a bleed was exsanguinated

#

but if it just flips a flag then thats fine ๐Ÿ˜ƒ

strong iris
#

well thats on average

acoustic lake
#

Check pins before you ask questions please

hushed heart
#

what makes memento so strong for assa?

outer rapids
#

Lol looks like outlaw's relics got broke again on the Simc front page rankings

fallen agate
#

Dont get it, when i sim Outlaw i am around 355k and as sub 260k :S Cant be that much diff just becouse i got higher ilvl relics as outlaw`

#

?*

vast bobcat
#

Stat priority is different, is probably your gear

hushed heart
#

Can anyone explain why memento is so high ranked for Assa? Haste and Mastery are really bad to have and thats 2/3 of the trinket basically...?

torn flume
#

did anyone ever find out if sub t18 2p is a dps gain

untold laurel
#

mastery is not bad for assassination

#

at all

#

its really good

digital sluice
#

Outlaw damage in simcraft is final atp?

#

It's was few days ago 10k higher and now back down

tight solar
#

If you were to have the Legendary boots (reduces vendetta cd by 1 second for every 50 energy you spend), would it be better to have the vendetta relics vs the rupture ones?

rain lance
#

I'm not sure anyone has explored that, worth simming to see

outer rapids
#

I doubt more than one or two would be worth it because the CD of vendetta would be lower than that of the CD for exsang and kingsbane.

karmic drift
#

how do i change my food buffs in simcraft?
just add this? actions+=/food,type=seafood_magnifique_feast
is that correct?

rain lance
#

look for the actions.precombat+=/food line

#

food is defined as a precombat action not an incombat action

karmic drift
#

dont have that using simc mod

#

so i can just add that?

rain lance
#

actions.precombat+=/food,name=the_hungry_magister

#

put that line in

#

I think that should work

karmic drift
#

under artifact line?

rain lance
#

and replace the food type with whatever you want

#

yeah

#

I think that will force override the food but I'm not sure

karmic drift
#

ill let you know

#

nah didnt work adding actions.precombat+=/food,name=drogbar-style_salmon

#

still has me using vers food

rain lance
#

try this then, load up the default profile and change the food line in there

outer rapids
#

@chrome sun Relics look like they broke again.

#

armory import, then sim works. But if you try and load the default profile or use the simc addon. It breaks.

summer prism
#

guys why simcraft shows me haste is the best stat

chrome sun
#

thanks Kaz, I'll cherry pick the commit again when I get home from this training ride

elfin remnant
#

btw... how i can change food for simc?

#

food=,id={id} ?

fathom bramble
#

look at the pre actions and food

untold laurel
#

anyone looked into angerboda/nature's call weights yet?

#

seems like a good trinket (combo) for sin mainly

pulsar escarp
#

what's the bonus id in simcraft for a six feather fan 835 ?

icy mango
#

trinket2=,id=141585,bonus_id=1497

#

disregard the trinket2 part

pulsar escarp
#

ty

tight solar
#

Based of my current sims, it's a dps increase to have 2 vendetta relics with the legendary boots

#

the 3rd is only a couple hundred dps extra

#

For those wondering...

torn flume
#

whats the proc chance on energetic stabbings

acoustic lake
#

25%

onyx kindle
#

Weaponmaster procs on Goremaw's bite are currently hiting for around 800k-1.2m and critting for 1.6-2m. Is this a knwn bug, artifact ability and is it modeled at all in dps sims?

rain lance
#

@onyx kindle why would that be a bug?

summer umbra
#

did my message get deleted?

#

wat

onyx kindle
#

It's just something that it seems nobody knows or cares about yet provides huge burst when it happens. I used simcraft this morning to model weaponmaster with my gear and the sim reported not using Goremaw's bite at all in the damage and spell breakdown.

rain lance
#

Weaponmaster mechanics have not been well mapped out, in many cases simc uses a guess on the mechanics

#

can you link a log of this happening

open hawk
#

So I play Sin atm.. using Hemo/Exsan.. however looking at logs the top sin rogues seem to be running Elaborate/Agonizing??? am i missing something do you swap at a certain mastery or wha? i want my dps to be as much as possible... was also thinking because i was looking on ursoc.. is that maybe why??

rain lance
onyx kindle
pseudo grove
#

if you want log of me

#

where it happened

#

(was yesterday on M Ursoc)

#

that's likely a bug and it is not implemented on SimC

#

aswell as some others guesses for weaponmaster that we need to find out

#

(for example, what exactly does a weaponmaster nightblade with finality ?)

#

is deepening shadow correctly working with weaponmaster

#

does the weaponmaster roll is done on both mh and oh hit of goremaw's bite

#

what we knows currently is, weaponmater roll is per target and not per cast

#

hence the talent is the best for AoE

#

my only guess for the weaponmastered big dmg of goremaw's bite

#

is they messed up the MH+OH merge

#

instead of 1000% it's 500%*500%

#

at least, something like that

#

@everyone if you have logs of this bug, please send it to me here or in pm, I'm trying to found out how they messed up, I think I got it but I need more samples

indigo elm
#

Your logs are private @pseudo grove ๐Ÿ˜ฆ

#

btw, how do you see if weaponmaster procd' ?

#

nvm found it ._."

unique carbon
#

I'm trying to remove potion of Old War from my sims APL, but... where do I find the APL? ._. a tut I found directed me to the profiles tabs but those seem pre-made characters with set stats etc?

unique carbon
#

I'm using the /simc command and don't get those lines, but when I sim it does use them

#

So I figured they're in the default Rogue profile?

indigo elm
#

Don't know if it helps but weaponmaster seems to proc on nightblade ticks

green zinc
#

Yeah @unique carbon when you go to simulate, all your apl things will be there. Just look for the ones with potion and deadly grace. They're in default rogue prof not /simc

unique carbon
#

So I have to simulate the /simc, then remove them again? ๐Ÿ˜ฎ

green zinc
#

so you just use /simc to pull your gear/talents/etc, the rest of your apl for actual rotation is in the simc program under the defaulted profile.

unique carbon
#

OH

#

Derp

#

Thanks a ton

green zinc
#

No problem

pseudo grove
#

Oops @indigo elm , check my twitter I put a pic

indigo elm
#

Ty ๐Ÿ˜ƒ

crisp basin
#

Yeah the weaponmaster procs on Dot ticks is whats labeled as "weaponmaster" damage in combat log

stuck sapphire
#

are we assuming weaponmaster to be better with the bug?

crisp basin
#

curiously im already simming higher with weaponmaster, but guess that could possibly be the influence of either the legendary cape or Tirathons trinket, or my specific traits

#

but even then they are both really close most of the time

untold laurel
#

are there any values out there for natures call and possibly in combination with the maw trinket?

#

i saw some rogues stacking them

crisp basin
#

theres shouldnt be anything special since both are more less just static secondary procs, they dont seem to be overly powerful based on SimC

#

their implementation in SimC is another thing, but I dont think anyone here knows the specifics

#

Natures call is probably slightly undervalued because it seems to be missing the Drake attack but thats like 500k dmg over an entire fight

ember wigeon
#

] Senorpwnd: Weaponmaster procs on Goremaw's bite are currently hiting for around 800k-1.2m and critting for 1.6-2m. Is this a knwn bug, artifact ability and is it modeled at all in dps sims?

#

it's a bug

#

known for like 3 or more weeks

#

people started noticing they would randomly one shot people in pvp

crisp basin
#

@ember wigeon Dys its been known and reported since alpha actully :/

ember wigeon
#

ah

crisp basin
#

it was really visible during a brief period where they made weaponmaster 12% chance

high rapids
#

Does anyone know if there's something wrong with the default APL for sin? I'm using default APL as sin and getting around 270-290k dps at anm 858.. meanwhile people in the general channel are reporting around 340k dps aat an 860 as the same build

plush relic
#

you may have unoptimised stat weights, suboptimal trinkets/talents, etc

high rapids
#

still though... an 80k dps difference seems like a lot

#

to just be stats/trinkets

#

I mean, simming 270k with 135%/32% mastery/crit as the poison sin build seems low

rain lance
#

If you think there is something wrong with the assn APL link some logs so we can take a look

versed thicket
#

make sure your artifact imported correctly?

high rapids
#

going to sim again to see what is up

outer rapids
#

You can PM your character if you want and I can take a loot at maybe the difference. I sim to about 330k right now.

fallow pelican
#

what are we talking about team? ๐Ÿ˜„

pseudo grove
#

#weaponmaster

versed thicket
#

Oh boy

#

is it happening?

plush relic
#

so is weaponmaster > MoS cause of this wierd bug?

pseudo grove
#

not only

#

it's better overall

#

does someone have others logs ?

#

my intial guess isn't correct if I take more sample

#

so I need more to try to understand what's going on ๐Ÿ˜„

plush relic
#

i send you some after todays raid if you want to

pseudo grove
#

yes pls

plush relic
#

will be my first raid as sub so dont be to harsh ๐Ÿ˜„

pseudo grove
#

lol

rain lance
torn flume
#

@pseudo grove you want logs with weapon master?

pseudo grove
#

yes to try to breakdown the numbers

torn flume
#

I'll post some later tonight for mythic nythendra If it's not too late. Or works you rather wailing on a dummy

#

Why do leg boots not reduce the value of energetic stabbing relics

fallow pelican
#

are we talking about viability between WM and the Shadow damage talent for sub? t1?

versed thicket
#

WM vs MoS

fallow pelican
#

cool cool

#

As I'm currently putting AP into my sub so I can write a guide once I get acclimated to it, I'd love to hear the results as well

pseudo grove
#

until you'll have near 0% downtime

#

energetic stabbing will always be worth

#

since ShD despite giving dmg gives also you energy back ๐Ÿ˜„

stuck sapphire
#

btw. did anyone test if old war works with t18 2p?

torn flume
#

Figured the extra 15 energy with leg boots wouldnt matter as much as nightblade damage

#

Since it doesn't increase the chance

#

But maybe I'm overestimating how much energy boots give

sage timber
#

@pseudo grove are u still searching for WM logs?

pseudo grove
#

yes

sage timber
#

i hope it's helpful

torn flume
#

Can you see people's artifact path on wcl

sharp heath
#

You can see which traits they have if you select a player and go to the Summary tab

#

Talents, traits, gear

river lagoon
#

@pseudo grove are you still looking for weaponmaster logs?

pseudo grove
#

yes @river lagoon

river lagoon
#

Hoping these help then

#

used WM for the entire raid

hoary bay
#

I need help with simulation craft if thats possible, for some reason i am getting a huge dps for outlaw and i feel like there is something wrong

crisp basin
#

Define huge

hoary bay
#

that huge

#

sims were done on single target as well

crisp basin
#

nothing too out of the ordinary, make sure you included correct artifacts and have up to date simc

manic pulsar
#

That does not seem too surprising

crisp basin
dim pendant
#

@pseudo grove I'll try and use weapon master for sub when we Raid wednesday night for you as well and link logs when I have them

pseudo grove
#

prob artifact import/export

#

btw you should use it

#

even without the bug it's strong

torn flume
#

which bug?

dusty elbow
#

guys, I have 2 legs higher ilvl with +115agility and +730crit and other leg less ilvl but will give 700 versa instead which is better

dim pendant
#

Conformed stronger than mos?

#

Sim it? @dusty elbow

dusty elbow
#

how ?

torn flume
#

aethys whats the bug?

dim pendant
#

Simc was over valuing mos apparently @torn flume

torn flume
#

was wondering if he was was refering to the goremaw bug

dim pendant
#

What's that one?

torn flume
#

apparently it's doing a lot of damage randomly

dim pendant
#

I'm going to try it out Wednesday so I'll see if I notice anything strange

torn flume
#

aethys what were you refering to being stronger even without the bug

pseudo grove
#

weaponmaster prob > mos

#

in ST

#

and besides ST it's always better

icy mango
#

How'd you come to that conclusion?

torn flume
#

didn't use only do 700k damage over the course of a fight

#

yours*

tame sluice
#

How would WM work with Insignia? I'f assume it would proc it again, but how about the ruling on procs proccing procs. And MoS doesn't doubledip on Insignia, right?

unique carbon
#

Is there any reason I'm not makig it to my simmed dps the majority of the time, as Outlaw? I feel I'm majorily on the lower end of the iterations even with decent buffs but can't seem to figure out whats wrong... Is it a known thing in Simc or should I head up to General to have my logs checked for user error?

#

Before the nerf I was getting fairly close, but now...

slow oxide
#

anyone have any links on this mastery assassination build? having trouble finding anything

rain lance
hushed comet
#

I hope this question is appropriate for this section: For outlaw, we have a general guide for rerolling. TB, 2+ buffs, or Shark with cooldowns. However, during curse, would it be optimal to roll away something like BS/Buried Treasure that doesn't do much of anything for Curse/AR?

tame pulsar
#

Is Weaponmaster the ability that gets recorded for the talent? It seems kind of underwhelming. I knocked a dummy around for 5 minutes and it only has 2 hits, max damage being 209K.

upper wave
#

Well, it's a 6% chance to proc. I imagine in that case, every 50 shadow strikes you should see 3 procs, for example. How many times did you eviscerate/ss?

tame pulsar
#

Does nightblade proc it too?

#

If it does, then in total I did 67 hits/ticks

upper wave
#

Not sure if it procs it or not. People are still unsure of how it works, but if its meant to be a replacement for multistrike, it's quite possible that it can proc off nightblade ticks. I'm going to run it on our farm night, and see what logs tell me.

acoustic lake
#

the recorded weaponmaster damage is just when weaponmaster procs off a nightblade tick

#

when abilties weaponmaster, they just cast again in the log, so they appear under their own damage

upper wave
#

Ah

hot radish
#

Spontaneous appendages seems to sim really low, and yet when i see it used in logs its a solid 5% or so of someones damage ,doing on average like 3.8-5 mil, are sims borked for this trinket? And is it any good for sub/poison sin build due to the mastery?

manic pulsar
#

@Rife#4323 You can keep a single SiW during the burst

#

But every other single buff is a DPS loss

#

@tame sluice

00:04:45.151    Roflerie Shadowstrike Lumbering Mindgorger *288155*
00:04:45.151    Roflerie Insignia of Ravenholdt Lumbering Mindgorger 60904
00:04:45.151    Roflerie Shadowstrike Lumbering Mindgorger *280154*
00:04:45.186    Roflerie Insignia of Ravenholdt Lumbering Mindgorger 59213
#

Insignia does not proc on Goremaw's Bite though

tame sluice
#

Too bad with Goremaws, but gonna run wm tomorrow on hc clear. Also doesnt work with SStorm, but might be a bit op if it did. Do we have data wether its better to SS instead of sstorm on two targets with insignia and 2nd shuriken trait? Prob storm without shd, but with itthe?

#

*autocorrect fml <.<

manic pulsar
#

@pseudo grove knows

pseudo grove
#

add 1 target for premed, boots, insignia

#

so from 2 to 5

#

if you have everything

#

(not possible yet since 1 leg restriction)

#

only during stealthed tho

acoustic lake
#

If I did my simc right taking pistol shot off the gcd would give maybe a 14k dps increase at t19m gear. Might be a nice way of bringing up outlaw again

#

not sure my override was perfect though, think I made it so that it wouldn't cause a gcd when used but you couldn't cast it during a gcd

manic bay
#

What if Pistol Shot critical strikes did 100% more damage?

#

Would also buff the value of gunslinger slightly

acoustic lake
#

could also be nice

#

I'm not sure thats an easy thing to override into simc to check

#

but making the traits be a bit more balanced in terms of dps would be a good start

manic bay
#

oh, what about giving Oppotunity! Pistol Shot crits deal 4x damage similar to BTE?

#

(buffs things more)

acoustic lake
#

anything to make it a bit more useful :p

#

although you're sounding like the idea of outlaw being buffed is a bad idea

manic bay
#

I like Outlaw and Sub the most atm :x

#

Ouutlaw is just frustrating when you get unlucky and get no SS procs or good RTB rolls. bleh

manic pulsar
#

feelsbadman

pure pelican
#

how much more dps would outlaw do if there is no 1 buff roll, makes the spec 100 times more fun

manic pulsar
#

On a Cenarius down I rerolled 10 times before getting a good roll on pull ๐Ÿ˜„

stuck sapphire
#

does energetic stabbing hit chance increese or is it 25% flat on all ranks?

acoustic lake
#

flat

fathom bramble
#

25% flat

stuck sapphire
#

thanks

#

did anyone test t18 2p with old war yet?

hybrid leaf
#

question: if i have all three ShadowDances on cooldown and i use a finisher, do all ShadowDance Stacks profit from a Finisher?

upper wave
#

No

#

Only 1 stack is ever on CD

hybrid leaf
#

thx

gilded hound
#

hmm... quick question.. how do i best sim what relic is best?... is there a place where id is shown?

small mountain
white trail
#

@pseudo grove so that addition proved to be correct?

#

talking about ring

pseudo grove
#

probably @white trail

#

XD

#

anyway the diff is very minor so

#

with both legs

white trail
#

I think at some point you would want to change some of those "and" to "or"

#

or just make min( premed + leg_1 +leg_2, 2 )

#

something like that

pseudo grove
#

maybe

#

depending on multiple things

#

including relics

icy mango
#

Energetic Stabbing has a 25% proc chance correct?

warped monolith
#

Yes

boreal rune
#

I saw a post on mmo-c a few days ago talking about Sin openings - it said that Rupture damage snapshots the EP buff if you cast Rupture while still having the buff. Does that also apply to Kingsbane? I assume it does

manic pulsar
#

Woot

#

Are you sure about that?

boreal rune
#

I haven't tested it myself yet, no. Still have an hour before I'm done class and can actually go play

prisma tartan
#

If it snapshotted having it, wouldn't it also snapshot not having it?

bold fulcrum
#

something wrong with latest sin api?

manic pulsar
#

If EP snapshots, then yes, most likely

#

I really doubt it does though

gilded rover
#

is it real that sub prio is mastery>vers>haste>crit?

#

Simmed today

manic pulsar
#

And if you simned it, then yes, it's your current stat priority with your current gear, artifact, trinkets, relics and fight style

#

and talents

azure saddle
#

Not sure if this is a question for this sub-chat but, looking at the sub apl table in simcraft, I can't see any waiting to get max uptime on MoS. Am I just blind or aren't we doing that at all anymore?

rain lance
bold fulcrum
#

cheers

manic pulsar
#

Ideally, sim whenever you gear change

#

@azure saddle @pseudo grove I don't know if it is a DPS gain to try and maximise MoS uptime; however, IIRC, Weaponmaster is better than MoS overall

azure saddle
#

hmmm weird, I sim around 8k dps less with weaponmaster. Thanks though

manic pulsar
#

Have you updated simc recently?

#

IIRC a bug with WM has been fixed a couple days ago

azure saddle
#

Oh shit. Well I'll go do that then ๐Ÿ˜›

#

Thank you kind sir.

manic pulsar
#

Anytime

azure saddle
#

Hmm much closer but still simming higher with MoS. That's ST though, I guess that if there;s even a couple of adds, weaponmaster with FoK pulls ahead.

torn flume
#

Does it work with shuriken storm

#

It would be good because your pounding out evisceratesthat can be copied

#

@manic pulsar how does WM work with finality

warped monolith
#

y it work with SS also to my knowlege is target based and not cast based, it copies ticks from dot and not ''full dots'' (im now siming higher with it than with MoS on ST) for finality no idea how it work but im guessing it just copies dmg

manic pulsar
#

@torn flume That's a good question, I would assume it doubles the same type of finisher but @pseudo grove would know better

crisp basin
#

it behaves like a normal evis cast would

#

so if you cast normal evis and get WM proc that proc consumes the finality

torn flume
#

That's pretty cool

#

So if you get a wm poeti

#

Wm proc on a finality finisher it will give you finality back

icy mango
#

I've yet to see that happen though. I've been collecting logs using wm and have only seen it proc off of nb ticks

#

Or ss

crisp basin
#

thats in combat log

#

WM proc on evis in combat log looks exactly like eviscerate

#

"weaponmaster" is dot ticks

icy mango
#

So it behaves as if it casted (insert ability here) a second time?

#

Shit you guys literally just said that my bad

azure saddle
#

Im guessing that the more mastery I get the better WM becomes vs MoS.

dreamy monolith
#

Assuming you have a 60% uptime on master of subtlety or higher, they should be about the same.

#

With weapon master doing ever so slightly better if it procs a combo point

icy mango
#

Why 60%?

dreamy monolith
#

Because mos is a ten percent buff

#

Wm is 6%

#

A six percent chance to double any ability should translate to a six percent overall increase, or slightly more because of combo point generation

#

That said, mos is improving your higher damage abilities, so it's effect is probably more pronounced than if it were a flat % permanent danage increase of the same value

#

Since it's guaranteed to be active on shadowstrike, which hits like a truck, but not always Backstab, which hits like a wet nerf ball

#

Wm has the same chance of duplicating either ability

white trail
#

If wm procs for Evis does it trigger RS as well?

crisp basin
#

nope

#

at least I dont remember that happening

#

can check the logs just to be sure

#

but iirc RS and Deepening SHadows were unaffected

#

the one at 1.10 is an nightblade btw, just filtered it out

white trail
#

okay

#

but combo points are awarded twice, right?

crisp basin
#

yeah

white trail
#

how can I disable effect of Cinidaria, the Symbiote in simc?

dreamy monolith
#

The takeaway here is that if weapon master copies the ability's you don't want it to, you get a little added bonus.

icy mango
#

Is that a Wm evisc crit at the bottom of the image, @crisp basin

#

Just making sure I'm reading it properly

crisp basin
#

yes, its evis cast followed by wm evis proc

deft hollow
#

Anyone know if rupture snapshots with a previously applied EP buff? I've heard different things

dreamy monolith
#

I'm not sure you can tell if the evis of the wm evis crits, but I'm not sure it matterz

crisp basin
#

Someone said earlier it does, but I havent seen any proof yet

#

it used to for a while on alpha, then stopped, and celestalon said its not supposed to

dreamy monolith
#

The implication is the same, that weapon master doesnt copy the ability, it casts a new version

deft hollow
#

I would imagine it only snapshots with vanish. Since rupture applies it's own EP it should treat it dynamically throughout the bleed

crisp basin
#

yes thats how its supposed to work, tho someone said the idea came from a mmoc opener thread which would explain a lot

deft hollow
#

I'll just punch a dummy tonight and check

crisp basin
#

@white trail dont think there are any direct settings for legendaries like how there are for tier sets, since even when APL ive seen check for legendaries they use equipped.item

#

you can try navv I guess

icy mango
#

I did some quick tests on a dummy and didn't see anything significant

white trail
#

@crisp basin I might just find perfectly matching stat item

#

if it's easier for simc

marble kelp
#

quick how does the wither j'im trinket

#

sim

#

seems mediocre

rain lance
hexed oak
#

is it possible that the default profile for assasination uses deadly grace while the profile for the sims on the frontpage of simcraft.org uses old war?

#

and if so...why?

#

because im siming at 293k with frontpage-profile and at 283k with default-profile

plush relic
#

frontpage simcraft profile? what do you mean by that

hexed oak
#

on the simcraft site...there are are simcraft results

plush relic
#

those are pretty old

hexed oak
#

i thought they are updated

#

4th october

plush relic
#

they are t19Preraid which is set to 840 ilevel

#

most people are past that

hexed oak
#

thats not my issue

plush relic
#

are you saying you compare your character sim results to the front page?

hexed oak
#

im using the profile of the homepage results with my gear at get 293k

#

using the default profile with my gear...i get 283k

#

and they are using different potions

plush relic
#

if you know they are using different potions, then i guess i dont understand your question

#

old war is better

#

so whatever you sim your char with should use that

hexed oak
#

then why does the default profile use deadly grace

plush relic
#

is your simc build updated recently

#

i bet the front page one is

#

but maybe yours is old

worthy spire
#

is goremaw + wm bugged?

#

or what

#

sometimes i c goremaw doing mills dmg

hexed oak
#

hm weird. im on 7.0.3 release 2

#

which is the newest version looking at the homepage

#

but homepage sims are on 7.0.3 release 3

worthy spire
#

Anyone else noticing the same?

#

with WM + goremaw interactio

plush relic
#

homepage release version

#

says released september 16

#

which 7.0.3 release 2 sounds right

#

if you go to nightly builds i bet you find 7.0.3

#

release 3

worthy spire
#

just ignore me i guess ๐Ÿ‘Œ

plush relic
#

i dont know enough about sub sims to know

#

would you rather i answer incorrectly? lol

#

from what i vaguely recall from conversation here it might be bugged where the value is multiplicative on goremaw on accident

chrome sun
#

@worthy spire - yes, there's a bug

#

frozen soviet got it in one - looks like the value is multiplicative

worthy spire
#

yup, clearly multiplicative. At least the tooltip "massive damage" makes more sense now...

#

Also, without that "bug", WM is trash.

upper wave
#

Even if it wasn't a bug, taking WM for a 6% chance at@'massive damage' seems trash anyways

#

Emphasis on 'seems'

sage timber
#

i think @pseudo grove posted something like WM prob > st and definitely for multi target, if i remember correct

torn flume
#

He said even without the bug it was probably better

cerulean scarab
#

I don't see anything pinned - are there any resources for simc tci configs?

#

I was wonder if there were just any sort of generic configs that could be reused with your own profile, etc

#

mostly because tci is pretty verbose; mostly looking for some examples

karmic drift
#

is chronoshard up there for us

upper wave
#

@sage timber fair enough. I still find it weird though. I'm taking WM into raid tonight, so I'll make my own decision on whether it matches up for me or not lol

real crag
#

Probs not the place but, has anyone followed through and switched to sub from Outlaw and are seeing an improvement? I'm seeing logs at ~860 and thinking I should be dumping into assass instead.

midnight shale
#

whats better in raid hemo bleed or master poison ?

chrome sun
#

take to simple questions please ๐Ÿ˜ƒ

midnight shale
#

gotcha

plush relic
#

this may not be the best place, but what is the best way to compare gear using simc if you are comparing multiple gear pieces in different slots? try every combination of gear pieces? because each one equipped can change the stat priority for your gear set, which can affect the next piece being better or worse than your current gear

cerulean scarab
#

I'm probably not looking in the right place, but I'd like to find information about stuff like that as well.

#

I suppose other than reading the tci source there's probably not an exhaustive list of what you can and cannot configure and how to do it

#

also don't know if this is the place to be asking this stuff, though I assume so

acoustic lake
#

essentially yes, it you wanted to find the best combination you'd have to try all iterations

stuck sapphire
#

well there is no optimal way you can sim every possible compination or sim your character and change the gear peaces due to the stat prios you got and repeat that till you like the result

acoustic lake
#

actually looking at it, perhaps the automation tool is what you're looking for @plush relic

hot cargo
#

any idea on if saytr is still the best neck enchant

plush relic
#

@acoustic lake exactly what im looking for, thanks. Now my only question is there a good way to export gear in my bags including bonus modifiers to a simc profile? I've been looking up ids on wowhead and thats kinda slow

grave copper
#

For assa, 835 horn of valor (882 vers + 2574 agi for 30 seconds / 2 min CD) or 870 unstable arcanocrystal (837 crit, haste, mastery, vers)? Which of these do I keep on with bloodthirty instinct

stuck sapphire
#

just sim them ๐Ÿ˜ƒ

grave copper
#

i honestly dont know how to use mr robot, is there anything special i need to do other than run the sim?

stuck sapphire
#

there is a guide how to sim in the faq selection, we are in the theoriecraft selection.

icy mango
plush relic
#

yeah, that guy is good

#

for those interested, here is how i automated all possible combinations of several gear slots

#

first, create a binary table of combinations for how many slots you want to replace

#

e.g. 5 slots = 00001, 00010, 00011, etc

#

there are websites that do this easily

#

then paste this table into excel, transpose the table

#

replace 0s with blanks and each line of gear with the gearslot you want to change

#

e.g. 1 is replaced with neck=,id=134492,bonus_id=3418/1532/3337 on the first line, head=,id=134440,bonus_id=3413/1808/1502/1813,gem_id=130217 on the second line, etc

#

then use excel macros to combine the table into 1 column, remove blanks, then insert blanks above your default gear profile and paste into simc

#

a bit complicated but beats typing in 64 (or however many slots you have) combinations of gear manually

#

there is probably an easier method but this worked for me

soft marlin
#

why is the bloodthirsty instinct the bis trinket for assassination?

torpid tree
#

the 60%+ uptime

pseudo grove
#

@plush relic try AutoSimC

#

And you can use SimulationCraft add-on to export ur item easily

plush relic
pseudo grove
#

Yes

#

That's my python script

plush relic
#

gotcha

pseudo grove
#

Let me push the update

#

With artifact

#

It's Old but it works

#

Will bรฉ merged at some point into CombinX

#

I wrotr i 2years ago

#

And didn't changes much since then

plush relic
#

each time i try to sim, simcraft crash even with the new build 703 04

#

thanks, one last question, you think its fine to sim pieces one by one as you get em, adjust stat weights, repeat, or better to sim sets like this? seems like doing this is a very small amount of optimization within the margin of error

#

most of the gearsets i simmed were like 100 dps within each other despite using different pieces

pseudo grove
#

I do always do gearset tbh

distant gate
#

Does Elaborate plannig snap shot rupture?

plush relic
#

Best stat percentages for outlaw

twin moth
#

Hi guys.. Just to update something weird.. I got a higher ilvl ring, and calculate weights with AMR. Even though it has a higher ilvl and worse stats for sub (critical), the contribution to the DPS on paper was lower than the older ring. However, when I checked on a raid dummy I managed outputting 20-30k more DPS with the new ring.

acoustic lake
#

I would be very surprised if a single ring was going to make such a contribution to your dps. Your tests are anecdotal, unless you're going to test a large number of iterations its hard to say that the difference between the tests are not just rng

twin moth
#

Hmm.. I tried it twice..

#

My rotation was similar

#

If not identical

#

Maybe the trinkets went off in a weird way..

#

I wonder how many tests I should perform.. What is random except the trinkets? Energy and CP?

acoustic lake
#

Simc exists for these sorts of tests ๐Ÿ˜›

twin moth
#

But then I need to trust them :p

#

I don't trust my own algorithms.. So trusting another's..

#

You know

acoustic lake
#

Well they have been looked over by a lot of people. They aren't perfect naturally and are generally better than a real player. Maybe do a 10 minute dummy test as perfectly as you can and compare your damage per ability on logs to what simc gives you for a similar length sim

#

Then you can check cast intervals as well, these sorts of things might give you a bit more trust in simc's results

twin moth
#

Skada isn't good enough?

#

What are cast intervals?

acoustic lake
#

Skada is probably fine I just find warcraftlogs a lot easier to work with

#

cast intervals are the average length of time between casting a certain ability

twin moth
#

Hmm.. I thought about installing it.. Maybe I will

#

Ahh

#

AMR has those?

acoustic lake
#

Not sure, I don't use AMR

twin moth
#

Maybe I should compare with Simcraft..

#

So you say 10 minutes, huh? I will go crazy

acoustic lake
#

thats about how long I do them if I want to check numbers are all right in simc ๐Ÿ˜›

worthy spire
#

Regarding trinket

acoustic lake
#

swap out the chrono shard I'd say

hard drum
#

anyone know if it's been calculated if using FoK on single target is worth it with the legendary cape?

#

I'm assuming it is when its at x stacks, just don't know how many stacks that roughly is

slate plume
#

Looking at the stat weight spreadsheet I have, the agi is better than the stats from the new trink.

#

and that's not considering that he has a chance for 30% mobility which is pretty sweet.

worthy spire
#

Hmm

#

I sim stronger with arcano, roughly 8k more dps. But they are just sims... Who knows the validity of those.

slate plume
#

oh. I guess my weighting is out of date or there's something in sub that isn't initially taken into acount.

#

Also, wouldn't the math on FoK with cape be something like if 50% of mut is less than FOK*11.5 then FOK would be beter wouldn't it?

#

since FOK is 35 energy rather than 55 but it only awards 1 CP.

worthy spire
#

Well, weight scales are variable, depending on current balance. But I still dont like the concept of having no base agi on trinket.

slate plume
#

oh, I derped at the math there but I just looked at my stats and it would seem worth it to pop FOK when you're at 30 stacks.

#

a little over 300k for FOK at 30 stacks compared to just over 100k for mut.

hard drum
#

yeah I'll probably just use it at 30 stacks each time

tawny mantle
#

Does anyone know the string to add to artifact in Simcraft for From the Shadows?

pine gust
#

i have it a sub build.. its op on adds burst

hard drum
#

@tawny mantle just go to wowhead artifact calculator, do the tree you want then click the link button

manic pulsar
#

@slate plume for assassination, it's mostly ideal to sync it with Vendetta

hard drum
#

the dreadlords deceit Cape?

manic pulsar
#

So, if Vendetta CD is lower than 1min => use FoK every Vendetta, if Vendetta CD is higher than 1min: use FoK when Vendetta is at 55-60s CD, then when at 29 stacks during vendetta

#

Yes

hard drum
#

cool, thanks :)

manic pulsar
#

And 29 stacks is better than 30, because you don't waste any uptime

hard drum
#

would you say it's worth switching to vendetta relics and AP (depending on the boss) with the cape?

manic pulsar
#

It is worth already for poison build

hard drum
#

I have rupture relics and waiting to switch out with 3 vendetta ones

#

hmm okay

#

from the shadows with more mastery makes the vendetta relics even better too

manic pulsar
#

It's a pretty tricky question, because I don't think the APL is perfectly optimized for every possible combinations of Master Assassin relics + boots

#

(my fault, mostly)

merry radish
#

Not sure if this is the place to ask, but at 17% haste, haste is still simming >= Crit (as Outlaw). Whereas in Sterjn's guide Haste is bottom, and also in BIS TEN Mythic gear list, the template only has 10% haste, when it could get more.

#

Anyone know why?

manic pulsar
#

trinkets, talents, relics, legendaries, all of that change stat priority

#

Best answer is: follow what simc says

pseudo grove
#

wow skasch is on fire this morning ! ๐Ÿ˜„

celest hill
#

Is weaponmaster still a mystery or is it really better than MoS? Last info i had is that we don't know and that simC values it more than it should or someting like that?

pseudo grove
#

that's somewhat true

#

don't have time to breakdown exactly the things

#

evisc and shuriken storm are well handled

#

I assume that backstab & shadowstrike aswell

acoustic lake
#

I tested dfa and couldn't find any evidence that weaponmaster could proc on the aoe or that it scaled with targets hit

pseudo grove
#

where I am unsure is nightblade & goremaw's bite

celest hill
#

aight Aethys, thanks for the quick info

pseudo grove
#

from what I remember but I would have to ask to navv to be sure

#

but that's the evisc part that is weaponmastered

#

at least, this one is for sure

#

or was at some point when it was tested ^^

#

since blizz like "ninja fix"

digital sluice
pseudo grove
#

date

#

mostly ๐Ÿ˜„

digital sluice
#

yeah, but on all specs you have higher deepsss

kind lake
#

i got a bit bored today and run a couple of extreme stat distributions on the Rogue_Outlaw_T19H SimCraft profile out of curiosity. dont know if its showing something new or interesting tbh. as someone who hates feeling energy starved, i'll try to get some more haste (obviously).
https://docs.google.com/spreadsheets/d/1iKMoSQR38PqGGb247--RfmcXON2Kr7EDjyX-B5Uf3Ko/edit?usp=sharing

Google Docs

Sheet1 Rogue_Outlaw_T19H, R_O_T19H( all mastery-> haste), R_O_T19H( all mastery-> vers), R_O_T19H( even distribution), Mastery, Mastery/ Haste, Mastery/ Crit, Mastery/ Vers, Haste, Haste/ Mastery, Haste/ Crit, Haste/ Vers, Crit, Crit/ Mastery, Crit/ Haste, Crit/ Vers, Versatility, Vers/ Mastery...

manic pulsar
#

Outlaw requires quite some haste

cerulean berry
#

So yesterday after i updated my Simc, I noticed my stat weights changed significantly from valuing vers to haste as outlaw. Has anyone noticed a similar situation. I would post my sims but I am at work right now.

manic pulsar
#

Dunno; when did you last update it before that? Do you have fatebringer relics?

zenith berry
#

just got a 875 signet of melandrus (the 10% autoattack ring); is this thing worth using on sin/sub? :/

crisp holly
#

Sim it!

zenith berry
#

my simcraft doesnt seem to recognize the extra white hit damage

manic pulsar
#

Update it

fickle wedge
#

can we get that chart pinned

#

cause simple questions chronically gets "how much haste should I have"

rain lance
#

a) no one actually reads the pins, b) I'm not sure that really answers the question about how much haste you should have because that question isn't really answerable from a general perspective

fickle wedge
#

yea but then they are like "at high ilevels, what are the stats I should be looking for and are ther numbers and blah blah"

i would love to say "here is a pin with a chart, gain what information you will and then sim yourself"

#

cause now i just say "sim yourself" and people get all frustrated

#

although, lets be honest here. most pople will jsut look at this and say "master sucks eh?"

pseudo grove
#

you can already get somewhat the answer there

#

the thing is

#

if you move from 18k to 21k to 24k

#

it doesn't change much

#

so it's valid to have a global idea

#

nothing can replace the SIY (sim it yourself)

#

because things like trinkets (haste that procs by stacks, big amount low uptime, low amount big uptime)

#

or relic (finisher cost, forgot the name)

#

can modify it

#

so there is no point to give exact value

#

also, you have to add in consideration the talents

#

and the fight length

fickle wedge
#

I agree

pseudo grove
#

when you put everything altogether, its' only valid in this given context

fickle wedge
#

and im not looking to give a static answer as much as information to go with saying "sim it"

pseudo grove
#

when you go out of that, you CANNOT know directly if it's valid or not

#

but that's the only trustable answer

#

people most of the time

#

ask for X or Y trinkets

#

here or pm me about that

#

but I cannot answer this just by my brain and looking at the screenshot

#

I need to see more infos

fickle wedge
#

but providing information liek that chart or the one you just linked better illustrates why simming is important

#

so that instead of "sim it" I can say "sim it, here is why"

#

because when i say "stat values are relative to your personal situation" no one ever understands that

pseudo grove
#

then

#

you should rather make a guide

#

to explain how things works

#

๐Ÿ˜„

fickle wedge
#

but im laaaaazy

pseudo grove
#

it's 10000X more valuable than just giving people put X haste and Y mastery

#

or whatever like that

plush relic
#

i agree simming it helps, but i was running into the issue the other night of taking it too far i think. When you get a gear drop, how do you personally determine if its an upgrade or not?

#

what i was doing was keeping all the questionable pieces and simming all combinations of them together but i think thats a bit much

#

and dps was within margin of error for the sim

rain lance
#

thats why we generate stat weights, you can use that a useful heuristic to determine if something is an upgrade

plush relic
#

right, but the issue i was having was trying to overthink it, because each piece of gear changes stat weights from what it was before

#

so i guess is it fine to just sim a gear drop as you get it, equip if higher dps, adjust stat weights, repeat

#

and how big is enough of an upgrade that its worth equipping, because several pieces of gear I have were maybe +70 dps or -70 dps and i think thats <0.5% error margin

rain lance
#

at the start of the night sim yourself get stat weights, use those stat weights for raid, repeat the next night using upgrades

#

values shouldn't shift really dramatically over a single nights worth of gear unless you get some special item in which case you can pre-sim that as well

plush relic
#

like a good trinket or relic, basically

rain lance
#

or set bonus

plush relic
#

sounds pretty good to me

rain lance
#

Also once I finish ShC that should also help quite a bit because you can just look at gear on the fly

plush relic
#

yeah that what i used to use, was very handy for stuff like that

#

simc is nice and all but the workflow is a bit painful

feral zinc
manic pulsar
#

I won't do that

#

And won't recommend doing it

#

It's error prone, because the ilvl equivalence of a relic is highly dependent on the rest of your gear

#

The spreadsheet I made is a raw approximation to give a general good idea of what people should be careful about

torn flume
#

Is there an addon that show's item id from Tilto

#

Tooltip

plush relic
#

theres a few that do it by default, but when i was trying gear on the best method i had was

torn flume
#

Do you can sim something quickly before rolling

plush relic
#

create equip set with current gear

torn flume
#

So

plush relic
#

as a base

#

equip the piece you wanna check

torn flume
#

Well

plush relic
#

and use the /simc addon

#

oh for that, not sure

torn flume
#

More for simming drops

plush relic
#

i think even addons that show id only show in your bags or if its linked

#

not with loot addons

torn flume
#

That you are deciding on wether to roll

plush relic
#

and you would have to find the bonus/wfd or whatever it had on it too

manic pulsar
#

Well, I see ids on tooltip

#

But I don't know which addon does that, maybe pawn or elvui

plush relic
#

i dont have pawn

#

elvui does that for me

pseudo grove
#

@torn flume IdTip

#

it shows every id possible

#

like on artifact trait

#

it show both powerid and spellid

#

on item, it's itemid

#

etc... etc...

#

I use it since for addon dev it's so cool ๐Ÿ˜„

manic pulsar
#

Does it show bonus ids? ๐Ÿ˜ฎ

#

drool

plush relic
#

if it allows you to copy ids/bonus ids and just paste, best addon since vanilla

torn flume
#

Thanks! @pseudo grove

pseudo grove
#

na, I don't think it show bonus id

#

but it's easy to do

#

just have to dump the item link

#

and add a line for it in the tooltip

#

I'm just lazy ๐Ÿ˜„

#

but you can suggest it to the guy who make the addon

#

on the curse page

torn flume
#

Is assass still bugged to where you can use both exsang and ap

manic pulsar
#

I would guess not; however, as Blizzard is well aware of that bug, I would not dare using it if still is.

torn flume
#

For sin the traits that apply to deadly poison, do they also apply to agonizing poison

manic pulsar
#

If you look over on wowhead artifact calculator, you can see the details of agoP vs deadlyP effects

static zenith
#

@manic pulsar did you already update your sims for "Bloodthirsty Instinct"? SimCraft changelog from yesterday stated that it wasnt simming correctly, as in proccing too often

manic pulsar
#

Hum

#

It will automatically be updated this weekend

#

That's bad news though ๐Ÿ˜ฆ

static zenith
#

this might drop the trinket to literally anywhere

#

"Reduce uptime of Bloodthirsty Instinct based on uptimes observed from logs."

plush relic
#

aka

#

6 feather fan is the best

static zenith
#

well theres always a second slot ^^ but even previously it was more or less evenly matched with chaos talisman and friends.

manic pulsar
#

Yeah

#

Main difference is, it can be more reliably dropped at 880+

static zenith
#

if i had known that a bonus roll for the arcanocrystal mightve been a decent idea today, seeing as it is on par even 5 ilvls below

#

fortunately that trinket seems to be the only one changed

plush relic
#

i rolled on it anyway despite my better judgement because of getting 0 loot all 3 world bosses, gold again.

#

but i think you can reroll if you go kill him again

stuck sapphire
#

not sure about that

#

but i dont think that this discussion belongs to theorycrafting

manic pulsar
#

It doesn't

stuck sapphire
#

btw. even if it becomes annoying did anyone test old war potion with t18 2p bonus yet?

plush relic
#

it should work i think

#

unless they disabled the set bonus

#

old war works off of damage increases

#

and thats a flat increase

stuck sapphire
#

the point is they designed it in wod, this means if its logical it doesent have to be right

plush relic
#

try it and see?

stuck sapphire
#

akaari for example is not impacted bythe bonus

#

well i will but not befor 7.1 its not worth to waste a potion for that^^

manic pulsar
#

For science!

plush relic
#

its 800g-1200g

stuck sapphire
#

oh a volenteer ๐Ÿ˜ƒ give me feetback after you tested it

manic pulsar
#

like, shoes?

plush relic
#

i vendored my 18m set

#

disabled item restoration atm

fluid nebula
#

Those of you who main Sub out there, are you finding that Stjern's opener is weird? SoD -> Shadow Blades -> Shadowstrike to 5- 6 combo points -> Nightblade -> Goremawโ€™s Bite -> [weirdness right here] Shadow Dance -> Shadowstrike to 5-6 combo points -> Eviscerate -> Repeat.

#

I find, with shadow blades, goremaws adds 4 points and I normally have one from just hitting the target

#

so im at 5 CP and it's not worth dancing again to SS

#

so I just evis then dance -> ss like normal

#

this is for the opener**

stuck sapphire
#

yes i think stjerns guide is bad, but enforcers say it should only provide a overview

fluid nebula
#

yeah, but posting an opener like that you'd think he would test it or at least think it through once...you would waste CP with that opener

#

and I think the 3x SS from stealth opener is probably better anyway

manic pulsar
#

@pseudo grove is the MoS around here

fluid nebula
#

3 SS, NB, blades, goremaw, evis, dance ->SSx2 -> finisher like normal

stuck sapphire
#

that also seems wrong

plush relic
#

i think proccing blades right away and 2xSS is better

#

but honestly openers are marginal compared to getting your rotation down

#

as long as you dont do it terribly

manic pulsar
#

^

stuck sapphire
#

you can look at the apl for what they use. else you need to do a exact calculation

rain lance
#

Do a not dumb opener (get your maintenance stuff up quick, use cds early, use better abilities first) and the delta is really small, I know someone did a long analysis of assn openers in MoP and the difference was <0.2% for basically everything

stuck sapphire
#

for me it seems best if you : 13.5 (14)sec prebull symbol 3.5(4) pre bull refresh symbol. open with 100 energy. and then i need to do calculate it again but i think it was 2(or 3)ss-nightblade-shadowdance-shd rotation->vanish->spells use stacks of shd and gorens to maintain energy and shd rotation

fluid nebula
#

when would you use Shadow blades fuu?

rain lance
#

if you want to do opener experiments you can hardcode openers in SimC and look at them, once you start talking about what seems best it isn't really theorycrafting

fluid nebula
#

hmm, I'll look into putting them into simc, that will most likely yield the best result

#

I dont know how to tweak the rotation that much though

stuck sapphire
#

well thats true: let me rephrase: its definitly best to open with 100 energy and extendet symbol but i cant exactly tell you the rotation becouse i did it long time ago and forgot about it (mainly becouse sub was bad)

fluid nebula
#

yeah, my assumption is starting with fresh symbols and 100 energy

#

but do you pop blades right off the bat? or go 3 SS before popping blades/goremaw

#

ill just have to sim it and c ๐Ÿ˜›

stuck sapphire
#

ye just adjust the apl^^

abstract void
#

where does the rotation come from? is it regenerated from the simulation tool or do you have to supply a "fixed" rotation for the simulation tools?

rain lance
#

Rotations are defined using special SimC action priority list (APL) syntax

plush relic
#

i think he means how was it developed

#

not sure

abstract void
#

but the user has to define it prior to simulation?

rain lance
#

Yes

abstract void
#

okay, I see... but that is totally man-made and is not the result of a optimization

rain lance
#

yes

#

There are some experiments with generated APLs but they haven't been particularly successful

abstract void
#

are they on github as well? I am a chemical engineer and my field of research is the optimisation of chemical processes and reactors based on a more or less self optimizing method. I am pretty sure it could be easily adapted to this simulation problem. And your simulations don't takes weeks to converge... but I don't want to invest time in something that has been done already ๐Ÿ˜ƒ

rain lance
#

you should also talk to @pseudo grove @manic pulsar @chrome sun who have been toying with this

#

I'm still somewhat skeptical because the search space for an optimal APL is almost certainly non-convex

#

and it isn't entirely clear to me what features you'd actually optimize over

manic pulsar
#

Well, let's just take a huge network of GPUs and let a deep learning algorithm toy with it

stuck sapphire
#

that wouldnt be that hard tbh

manic pulsar
#

Optimizing everything has a skyrocketting complexity

rain lance
#

Thats where the feature selection and encoding comes in (at least as far I know, this is somewhat outside my area) but a very hard task regardless

pseudo grove
#

it might comes at some point in the future

#

but atm, my APLs are done using my brain and sometimes WCL

#

^^

rain lance
#

I am curious to see what Kuar has to say on this as a domain expert

pseudo grove
#

like, back in the days I watched a lot @acoustic lake logs

#

right now, I do try to optimize through my not yet released Addon

manic pulsar
#

I did with my custom made weakauras for assassination, which works basically the same ^^

stuck sapphire
#

ye you can basicly implement the apl into the game with weakauras ^^

abstract void
#

maybe I am missing something, but the APL looks quite "simple" in terms of search space complexitiy... it will certainly not be possible to optimize the APL for every peace of gear... non-convex, non-linear (?) search problems do take time, yes, but with the right algorithm this could be done within a day of computing time on a normal computer (first rough guess)

rain lance
#

Oh yeah gear is likely very doable, I thought you were talking about the actual rotation where the search space complexity seems intractable

manic pulsar
#

The problem is the APL itself

#

And more precisely, the if conditions

abstract void
#

No, I meant finding the actual rotation for every piece of gear combination e.g. with trinket X only do 4 CP then Envenom because... stuff

manic pulsar
#

Hum

#

The number of possible "states" for the player, for single target condition, can be probably reduced to a finite number

rain lance
#

Hmm, ok now I'm very interested

#

Yeah skasch but it is a pretty big number, 100 energy states, 6 cp states, lets quantize relevant buffs at 0.1 seconds, and say 5-6 relevant buffs per player and you are getting into the billions

manic pulsar
#

Yeah

rain lance
#

of course you could likely prune a ton of those, for instance any energy value below the min energy cost is known to be invalid

manic pulsar
#

I would go by 5 energy steps and 1s buffs xD

rain lance
#

or rather uninteresting

manic pulsar
#

But yeah, that what I had in mind

ember wigeon
#

maintaining APL for more than 8 legendaries would be complicated in the long run if blizz go this way

#

he's right

fickle wedge
#

see, i dont hang out in other class forums and chats often, are all classes as nerd centric as rogue?

manic pulsar
#

Yeah, billions of states is not reasonable in any way

fickle wedge
#

cause this chat is like, developer porn

ember wigeon
rain lance
#

Thinking about encoding in more detail, since the APLs are now more tied to these variables perhaps a first pass encoding would look at those variables themselves

#

and I don't know Cerin, rogues have had a weird math bent for as long as I can remember but I don't know if that is some outlier

fluid nebula
#

@fickle wedge I dunno about you, but I've finished 3 times since they started talking about APLs

fickle wedge
#

Right? thats what im saying. Like i ahve seen theorycrafters before, but we are talking about some hardcore code right now.

fluid nebula
#

GPU farms and shit ๐Ÿ˜„

manic pulsar
#

Math is pretty much the only thing I really have been good at ^^

fluid nebula
#

which APL would I adjust to change the opener?

#

Stealth cooldown?

abstract void
#

I still have some cluster time available this year. But I would rather aim for a solution anyone could run at home

rain lance
#

Ok lets say we fix the APL order and organization and then using our knowledge we constrain the set of possible conditionals for each ability, then you could encode state reasonably

abstract void
#

I am no theorycrafting professional but using some sort of boundary conditions these states can be reduced by several orders of magnitude in complexity

rain lance
#

You mean constraining the range different values can take?

abstract void
#

Not sure we mean the same thing, but yes...

fluid nebula
#

generally adding boundary conditions to calculations like this will reduce complexity by orders of magnitude

rain lance
#

Ok lets be more concrete, consider this row:
actions+=/variable,name=ss_useable_noreroll,value=(combo_points<5+talent.deeper_stratagem.enabled-(buff.broadsides.up|buff.jolly_roger.up)-(talent.alacrity.enabled&buff.alacrity.stack<=4))

#

here we know that cp thresholds are integers 0 to 6, alacrity stacks are integers 0-20, and then we have some boolean states with buffs

#

so our state can be represented by those values and that looks a lot like an integer optimization problem

abstract void
#

yep, but you probably don't have to answer this particular question for every energy state from 0 to 100 indepently but can boil it down to energy above 50 and below 50

rain lance
#

hmm, ok

manic pulsar
#

So you would do that reduction by hand ?

abstract void
#

We do something like that in the optimization of our chemical processes... We take a million fluid elements and solve heat and mass transfer for each fluid element... and optimize it... after that we take the optimized fluid flow field and add some chemical reaction... optimize it again... then we let this flow field with the reaction interact with reactor walls and heat transfer to these walls... and optimize it again... until we can derive a optimal chemical process from it. Yes, it might not be the global minimum if we made mistakes in the first steps, but it reduces computing times for these problems from years to weeks

#

I'm not sure how to do it mathematically and code wise here, but I am imagining something similar... so the reduction where this border is is done in the first step... and so on

fluid nebula
#

someone's an engineer. I'm jealous, I graduated when there were no engineering jobs because of the recession and so im in web development now ๐Ÿ˜›

manic pulsar
#

APL is much more discrete by nature

#

But I get the general idea

#

I have to think about how to translate it into that specific problem though

abstract void
#

yeah, the APL will require a different approach and frankly I am not totally sure it can be done at all within a resonable time ๐Ÿ˜ƒ - but I would have lots of fun thinking about it and try to solve it

chrome sun
#

i was tagged

#

are we talking about optimal APL search?

ember wigeon
#

Next step is machine learning

#

Bots will find the best rotation for you

chrome sun
#

most of the interesting work is basically brute force + genetic stuff

#

i'm not the worlds biggest fan of genetic algorithms in general

#

but they work well here

#

particularly because i wasn't searching for the "best" APL in general

#

just wanted to explore the build/spend options for outlaw

#

so you have a feature vector like (num_CP, jolly_roger_buff, broadsides_buff, opportunity_buff, time_till_max_energy) and you want to learn what it should do (ghostly strike, pistol shot, saber slash, gouge)

#

i took as gospel the "outer" bits of the APL

#

and just generated a lot of options for the inner bits

#

and them simmed them, and recombined them intelligently

#

(s/intelligently/with a genetic algorithm)

#

where "them" refers to the chromosome that those mappings define

plush relic
#

If i may, indeed what you want to fin the optimal APL is machine learning. But as you mentioned the problem is too big to bรฉ brute force. This is very similar to the Go game that Google deepmind dรฉveloped. Basically you use deep learning to add some sort of "intuition" before doing ร  random tree search.

#

What's nice is that once your algorthm is trainer it is really fast to compute and everyone cab use it at home

#

But training requires huge gpu :p

ember wigeon
#

if Blizzard buff and nerf here and there you would need to retrain everything

#

would cost much cpu ๐Ÿ˜ƒ

plush relic
#

Indeed

#

But still very interesting to develop!

chrome sun
#

what's the function that your network is trying to learn? that's the hard part

#

it's the representation space

rain lance
#

ok I think this deserves its own channel

chrome sun
#

ehhh, maybe not worth fragmenting?

#

we can take it to general if you'd prefer

abstract void
#

the network is aiming for maximal dps on a patchwerk type encounter

rain lance
#

Figure we can keep this channel for less speculative TC discussion

chrome sun
#

fair.

ember wigeon
#

I don't think you can train for any kind of gear easily

#

would cost so much to sim(train) for each gear setup

#

but Idk

rain lance
plush relic
#

@pseudo grove still need some sub WM logs or is it enough now? ๐Ÿ˜ƒ

tough goblet
#

Hey guys. Do you think it's worth to take either Swordmaster or Quickdraw after getting Waterlogged Lgendary? (Between the eyes has 20% per combo point to increase the damage of pistol shot by 400%) Also I have Blunderbaus

#

Any solid math on this?

fickle wedge
#

i would doubt it

#

10% damage incerase vs 9 pistol shots per minute

#

i can find some logs to try and compare

tough goblet
#

Blunderbauss is becoming more prevelant now that we are getting later into Legion. Without the legendary (Pistol shot Legendary) I assume it's not worth it. But with the legendary, it could be different

#

Because the assumption is that 1 Between the eyes + Pistol/Blunder > Eviscerate

fickle wedge
#

i would say definitly not saberslash

#

since it only bosts the damage of your next one

#

sorry

#

swrod master

#

sword*

tough goblet
#

I knew what you mean

#

meant*

fickle wedge
#

lol

tough goblet
#

I think there might be an arguement for quick draw

fickle wedge
#

maybe

#

most of what i have said so far doesnt make sense

#

like at all

#

lol

tough goblet
#

hey

#

I found some logs

#

wanna discuss this?

fickle wedge
#

yea, im looking at some too

#

trying to math it out now

open gust
#

Is it normal that simc and amr adwise different stats? amr vers and simc haste on 863 itmlvl outrogue...

#

8k+ haste according to simc is ok, but according to amr just 2k+

oak knoll
#

amr still has bte in the rotation i wouldnt really trust it

#

unless they changed it frm last time I checked ...

cobalt torrent
#

bte does more dmg for me atm then rt but the rng is not always great

#

idk amr sim dps #'s seem more on par then wtf simc sim's me at good lawd

waxen pilot
#

Ive got a trinket question. Trying to figure out if an 835 six-feather fan is better than an 840 tirathons betryal. I dont want to equip the fan to sim it in case its not btter so I can trade it to someone else, but I dont know how to add it to a sim if it isnt equipped

acoustic lake
#

trinket1=,id=141585,bonus_id=1497/3336

#

will do an 835 six-feather fan

#

for future though, you can get the required ids from wowhead

waxen pilot
#

thank you

pseudo grove
#

@plush relic it's fine ty

mellow wagon
#

So is agonizing poison really better than ensanganate now?

manic pulsar
#

Depends

#

BtE is worth it if you have the legendary and SiW

pseudo grove
#

might be just legendary btw @manic pulsar

#

there were bugfix on this on simc

#

didn't tested since ... well .. sub 4 life

#

but since u're not busy ๐Ÿ˜›

manic pulsar
#

Yeah

#

I'll test that

maiden hatch
#

Where are the 865+ sims where sub is king? Their absense is tangible.

#

Even tho all the really good ppl here go sub ๐Ÿ˜‰

manic pulsar
#

absence?

maiden hatch
#

Someone like aggregate dps for EN mythic over the last week

#

Outlaw is absolute lowest spec ingame

#

Sub is shit tier

#

Assass is middle of pack