Latas

EDIT: Erased my previous comment but added another.


Is there a way I could simulate the 3% crit from focus magic being added to everyone in my list of sims I'm running? It would make things much easier to see the benefit of the 3% crit on everyone all at once and just compare that one batch sim to the previous batch sim where nobody has it.

One more thing, is the dranei hit aura toggled on with the optimal raid setting? If it is toggled on is there a way in the batch of character sims to turn it off for some and leave it on for others?

 

Starfox

Heroic presence is off by default, unless you have race=draenei or a party set up with at least one draenei.
Don't really know what you wanna do with focus magic, you have to set it up for it to work. If you don't supply a target in the action the sim assumes 100% uptime on the 3% crit feedback buff

Hello.
Light the fuse.
For all my homies.
Do not run, we are your friends.

 

Latas

I'm not concerned about the feedback buff, what I'm concerned with is the 3% crit the original buff gives. I haven't found the stat scaling to be entirely accurate as it doesn't seem to take raid buffs into account. Essentially what I'm trying to show is how much each person in my batch would benefit from having the 3% crit from focus magic as accurately as possible. So I was looking to run one batch of everyone without it, and then another with everyone getting +3% crit.

 

dedmonwakeen

No easy way to do this. After each player description, you have to add enchant_crit_rating=138 to effectively give everybody 3% crit

http://code.google.com/p/simulationcraft

 

Althor

It's not perfect (and dedmon may decide it's not needed), but I've just committed default_enchant_<stat> options.

e.g. Rather than adding enchant_crit_rating=138 to each profile, you could instead have a single line with: default_enchant_crit_rating=138 and it'll then apply that (in addition to any enchant_crit_rating the player has) to each player.

One caveat is that it only supports things like crit_rating rather than spell_crit% and melee_crit%. So you may want to take that into consideration if you've got melee in your profile and you were just wanting to model what happens when you boost crit_rating by 138 for casters...

The complete list of default_enchant stats (they're the exact same as the player specific ones, but with "default_" in front) are:

default_enchant_strength
default_enchant_agility
default_enchant_stamina
default_enchant_intellect
default_enchant_spirit
default_enchant_spell_power
default_enchant_mp5
default_enchant_attack_power
default_enchant_expertise_rating
default_enchant_armor_penetration_rating
default_enchant_armor
default_enchant_haste_rating
default_enchant_hit_rating
default_enchant_crit_rating
default_enchant_health
default_enchant_mana
default_enchant_rage
default_enchant_energy
default_enchant_focus
default_enchant_runic

By default they're all set to 0 both in the binary and in Globals_T<whatever>.simcraft

 

Althor

Yes. All of the items were accidently exposed on the EU wowarmory and they've since been mirrored by mmo-champion.

I used the full list (including the ilvl 272 cloaks) when creating the Shadow Priest BiS sets.

Of course it's always possible I guess that not *every* item was found, but the loot list on mmo-c certainly *looks* complete.

 

dedmonwakeen

Version r3168 available for download.
SampleOutput wiki pages updated.

http://code.google.com/p/simulationcraft

 

dedmonwakeen

SimulationCraft r3248 available for download
* Paladin module officially supported (and then there was one.....)
* Support tanking stats including diminished returns
* Support boss auto-attack to properly model some Warrior and Paladin talents
* Support for passive and active racial bonuses
* Various trinket refinements
* Improved detection of tier set bonuses
* Complete aura/buff/debuff re-architecture for improved handling and reporting

http://code.google.com/p/simulationcraft

 

Maje

I've been looking at the mage simulation and the empowered fire mana gains where quite a bit lower then real life situations, mage model uses:
However empowered fire was changed to 2% base mana, just before 3.2 was released, so the above line should probably be * 0.02.

 

dedmonwakeen

Thanks for the catch. This is fixed in r3334.

http://code.google.com/p/simulationcraft

 

Mandalay

Could you please explain further?

Anything over the soft cap should result in zero reduction in DPS for the loss of a single point +hit rating. Only when the +hit reduction drops below the soft cap should it have any effect on DPS, but instead I am getting the following results when using this script

cd /d "%~d1%~p1" simcraft.exe armory=us,Dath'Remar,XXX max_time=600 optimal_raid=1 smooth_rng=1 threads=2 iterations=20000 calculate_scale_factors=1 smooth_scale_factors=0 html=XXX.html

DPS: 7691.5 Error=1.8 Range=504 DPR=24.8 RPS=310.6/277.7 (mana)
Origin: The World of Warcraft Armory
Core Stats: strength=319|68(10) agility=327|75(10) stamina=1554|1093(932) intellect=1230|1014(857) spirit=804|607(436) health=22500|18236 mana=22571|19233
Spell Stats: power=3271|2273(2273) hit=14.68%|14.68%(385) crit=27.87%|20.51%(446) penetration=0|0(0) haste=20.78%|11.68%(383) mp5=0|0
Attack Stats power=1185|58(0) hit=11.74%|11.74%(385) crit=23.84%|13.82%(446) expertise=0.00|0.00(0) penetration=0.00%|0.00%(0) haste=38.04%|11.68%(383)
Defense Stats: armor=2618|2114(1964) blockv=160|34(0) defense=0|0(0) miss=0.00%|0.00% dodge=0.00%|0.00%(0) parry=0.00%|0.00%(0) block=0.00%|0.00%(0) crit=13.90%|13.90%
Glyphs: life_tap/unending_breath/souls/conflagrate/drain_soul/incinerate
Actions:
chaos_bolt Count= 39.0|15.5sec DPE= 11931|10% DPET= 8310 DPR= 49.3 pDPS= 775 Miss=0.0% Hit=7531 Crit=15738|18565|53.6%
conflagrate Count= 55.0|10.9sec DPE= 12855|15% DPET= 9860 DPR= 23.2 pDPS=1179 Miss=0.0% Hit=7462 Crit=15589|18651|66.4%
curse_of_doom Count= 9.0|61.0sec DPE= 16715| 3% DPET= 12549 DPR= 28.9 pDPS= 251 Miss=0.0% TickCount=9 Tick=16715
immolate Count= 38.1|15.9sec DPE= 12646|10% DPET= 11097 DPR= 21.5 pDPS= 803 Miss=0.0% Hit=2480 Crit= 5184| 6188|48.4% TickCount=188 Tick=1799
incinerate Count=244.8| 2.4sec DPE= 9873|52% DPET= 6105 DPR= 20.4 pDPS=4029 Miss=0.0% Hit=6094 Crit=12731|15515|56.9%
imp (DPS=655.3)
fire_bolt Count=317.2| 1.9sec DPE= 1238| 9% DPET= 655 DPR= 6.9 pDPS= 654 Miss=0.0% Hit=1121 Crit= 1682| 1854|20.7%
Constant Buffs: arcane_brilliance/blessing_of_kings/blessing_of_might/blessing_of_wisdom/divine_spirit/fel_armor/fortitude
Constant Buffs: mark_of_the_wild/replenishment
Dynamic Buffs:
backdraft : start=55.0 refresh=0.0 interval= 10.9|10.9 uptime=49% benefit=46%
bloodlust : start=1.0 refresh=0.0 interval= 0.0|0.0 uptime= 7%
dying_curse : start=11.3 refresh=0.0 interval= 54.4|54.4 uptime=19% trigger=15%
empowered_imp : start=64.2 refresh=1.6 interval= 9.2|9.0 uptime=11% benefit=17%
illustration_of_the_dragon_soul : start=1.0 refresh=384.9 interval= 0.0|1.6 uptime=100%
life_tap_glyph : start=13.2 refresh=2.8 interval= 42.0|38.6 uptime=99% benefit=98%
lightweave_embroidery : start=9.9 refresh=0.0 interval= 63.7|63.7 uptime=25% trigger=35%
pyroclasm : start=36.5 refresh=0.0 interval= 16.3|16.3 uptime=60% benefit=48%
Procs:
283.5 | 2.61sec : judgement_of_wisdom
16.1 | 38.56sec : life_tap
Gains:
13061.4 : blessing_of_wisdom
22570.5 : initial_mana
21862.8 : judgement_of_wisdom
58741.8 : life_tap (overflow=6.3%)
26997.8 : replenishment
45900.8 : soul_leech
Scale Factors:
Weights : Sta=0.00 Int=0.34 Spi=1.02 SP=1.53 Hit=3.05 Crit=0.81 Haste=1.10
Wowhead : Items - World of Warcraft

Auras and De-Buffs:
Constant: abominations_might/battle_shout/bleeding/blood_frenzy/crypt_fever/curse_of_elements/earth_and_moon
Constant: elemental_oath/expose_armor/faerie_fire/ferocious_inspiration/flametongue_totem/heart_of_the_crusader
Constant: hunters_mark/improved_faerie_fire/improved_moonkin/improved_scorch/improved_shadow_bolt/infected_wounds
Constant: insect_swarm/judgement_of_wisdom/judgements_of_the_just/leader_of_the_pack/mana_spring_totem/mangle
Constant: master_poisoner/misery/moonkin/poisoned/rampage/sanctified_retribution/savage_combat/strength_of_earth
Constant: sunder_armor/swift_retribution/thunder_clap/totem_of_wrath/totem_of_wrath_debuff/trauma/trueshot
Constant: unleashed_rage/windfury_totem/winters_chill/wrath_of_air
Dynamic:

Waiting:
All players active 100% of the time.

Baseline Performance:
TotalEvents = 66476797
MaxEventQueue = 23
TargetHealth = 4609215
SimSeconds = 11999914
CpuSeconds = 24.757
SpeedUp = 484708

RNG global Actual/Expected Roll=1.000101 Range=1.000000 Gauss=1.000000

Scale Factors:
XXX Sta=0.00 Int=0.34 Spi=1.02 SP=1.53 Hit=3.05 Crit=0.81 Haste=1.10 Lag=0.00

Hit also seems to be reporting incorrectly when following the link to either Lootrank or Wowhead, with a massive bias toward hit gear despite the 0% miss rate reported with the scale factors.

Could anyone please offer some insight?

 

dedmonwakeen

Scale Factor = ( DPS( Stat + Delta ) - DPS( Stat ) ) / Delta

Hit:
Stat=385
Delta=-150

SF = ( DPS( Hit=235 ) - DPS( Hit=385 ) ) / -150

The scale factors for Hit and Expertise are derived by REMOVING Hit/Expertise and seeing how much DPS is LOST.

We do this because for many classes the cap is "hard" (not soft).

Given the frequency with which I have to answer this question, I'm starting to regret this decision......

http://code.google.com/p/simulationcraft

 

• Binkenstein

Welcome to my world
Probably best to add it to a FAQ or something and hope people read it.

 

dedmonwakeen

SimulationCraft r3393 available for download.

See RecentUpdates for list of key changes.

For those folks who roll their own executable, I encourage you to start playing with the Qt wrapper. It is not yet complete, but I need help making sure it will work on all platforms.

http://code.google.com/p/simulationcraft

 

StormyParis

Thanks and congrats for a very useful et impressive tool. I'm in awe of your work.

It badly needs a way to take mobility into account, though. Most fights are very mobile, or at least have very mobile phases. Event if we don't take those phases's impact on raid buffs/debuffs into account, they have a huge impact on individual DPS, varying with your spec.

I remember when my main toon was a hunter, I did more damage on Ignis et Kolo as BM, because while my toon was wasting time running around AOEs, or I was messing up the rotation looking out for them, Wolf was single- mindedly chewing its prey.

Mobility must be a bitch to model, between the "can't stop for more than 4s" phases, the "move for 3s every 15s" and the other variations. But it would really take simulcraft to another level :-)

 

Starfox

There are some raid events you can trigger, movement, stun, vulnerability, but that feature is pretty undocumented I think

Hello.
Light the fuse.
For all my homies.
Do not run, we are your friends.

 

dedmonwakeen

I was contacted by a friendly-but-firm employee of simcraft.com, a racing/flight simulation hardware/software company:

Per his request I am renaming our downloads and our execs to start out "simulationcraft" instead of "simcraft".

He also asked me to publicly request that people stop saying "simcraft" when they mean "simulationcraft" because we are corporately diluting his name brand. So....

Please stop saying/typing/humming/signing "simcraft" when you mean "simulationcraft".

I'm not sure what good that will do...... but he asked nicely and I try to accommodate nice people.

http://code.google.com/p/simulationcraft

 

• Zakalwe

I'd love for the executable to be sim.exe or something... surely that would be permissible?

 

dedmonwakeen

Good idea. How about sim.exe and simgui.exe?

http://code.google.com/p/simulationcraft

 

Erdluf

I see a lot of google hits for sim.exe and wowsim.exe. simwow looks less "busy."

 

Lhivera

So, I'm trying to get action lists built for Frost's 3.3 changes. I'm finding the results odd, thought I'd see if anyone had any suggestions for what I'm doing wrong.

Here's the action list (the items I've added are in bold):

actions=flask,type=frost_wyrm
actions+=/food,type=tender_shoveltusk_steak
actions+=/water_elemental
actions+=/arcane_brilliance
actions+=/focus_magic
actions+=/counterspell
actions+=/mirror_image
actions+=/icy_veins
actions+=/mana_gem
actions+=/speed_potion
actions+=/frost_bolt,frozen_core=1
actions+=/ice_lance,frozen=1,haste<=0.40
actions+=/frost_bolt,frozen=1
actions+=/cold_snap
actions+=/frost_bolt
actions+=/evocation
actions+=/ice_lance,moving=1,frozen=1
actions+=/fire_blast,moving=1
actions+=/ice_lance,moving=1

Water Elemental is moved up top since you should have it out before the fight starts now.

The Frostbolt and Ice Lance additions are intended to simulate this behavior:

- If the Frozen Core buff is up, you cast a Frostbolt, no matter what.

- If FoF is up and haste has not exceeded 40%, you cast an Ice Lance. This should trigger Ice Lance on the second charge, because the sim won't (I think) interrupt a Frostbolt cast that's already begun when the proc happens. (I'm ignoring Ghost Charges for this particular action list.)

I would expect that this should cast an Ice Lance as often as FoF procs, as long as no active haste buffs are up. Instead, what I'm getting is about 4.9 Ice Lances in about 23.8 FoF procs, which seems really off. Removing the haste constraint from the Ice Lance cast increases the casts to around 10.

Output is here: Simulationcraft Results

Anyone able to spot what I'm doing wrong here?

At Veridian Dynamics, we can even make radishes so spicy, people can't eat them. But we're not, because people can't eat them.

 

Althor

Due to trademark reasons etc. etc., in the current SVN build we've changed the short-name of the project from simcraft to simc.

Thus, all the default profiles have been renamed from <whatever>.simcraft to <whatever>.simc

simcraft.exe (and simcraftqt.exe etc) have been renamed simc.exe (simcqt.exe etc).

When discussing the project, ideally we'd prefer it it was referred to either as SimulationCraft or simc rather than simcraft to avoid confusion with the non-WoW related product of that name.

 

foopeemoa

For dual-wielding classes, how does it handle syncing of weapon autoattacks? If both weapons are identical speed, does it offset them and leave them that way, or is there a way to force the sim to sync the autoattacks of both weapons if they are identical speed?

I ask because weapon syncing happens often on mobile fights, and can be very annoying for Warrior rage generation. It would be interesting to be able to model this happening.

 

dedmonwakeen

Actually, the weapons will be sync'd by default...... I need to add an option to go the other way....

http://code.google.com/p/simulationcraft

 

dedmonwakeen

New version available for download:

simc-322-0-win32.zip

It includes a alpha-level GUI front-end: simcqt.exe

I imagine there will be trials and tribulations at first, so please open Issues on our site.

Note that the exec name has changed from simcraft.exe to simc.exe

http://code.google.com/p/simulationcraft