It's not the default release, but you should see it on the list. If you want the direct link you can find it above.

You mean the 91 other pages? I doubt I'll find a link even with the search. I've been to the site and the downloads is only 1.7.

WTB Pruning. =/

Current History of downloads have been from:

EnhSim: Downloads
Downloads - enhsim - Google Code

Think you could link me to a download?

Page 91. Five posts prior your question :-)
EnhSim - Release: v1.8.0

* is blind *

Edit: Tyvm btw :P

The mac binary works on my PowerPC. Nice job

From the BiS discussion:

Is someone able to shed some light on this?

Doesn't this go against testing earlier in the thread? I'm sure that it was confirmed by in game testing that MW4_LB didn't reset your swing timer.

Edit: See page 53 EnhSim, DPS simulator

The post explains that in-game testing demonstrates that the swing timer is not reset, however it is unclear on whether or not the sim models this correctly. Is anyone able to confirm?

The DPS-report in the BiS thread shows a delta of 3% crit difference between SS/WF vs MH.

This tells me that EnhSim uses 3% crit-depression for MH and no crit-depression for specials?

SimulationCraft uses "3% + delta_level*0.6%" for crit-depression when delta_level>=3.

It also applies the crit-depression to specials in addition to MH/OH.

I'm ready to make the change to mirror EnhSim, I just need some confirmation that my understanding is correct.

It's not a bug, that is how it should be modeled. Casting with any stacks of MW active changes the nature of casting any spell so that it doesn't reset the swing timer. If a cast time intersects a swing then that swing doesn't occur but the swing timer is not reset, provided there is at least 1 stack of MW active.

Hmm, ok so with 0 stacks is the only case LB and CL reset the swing timer? I can revert that part then and make sure it correctly clips.

And, not that it's super relevant but does LvB work the same way? I know MW doesn't affect it, but does it clip rather than reset when MW is up?

Ok it looks like the current implementation for casting with 1-4 MW stacks is that it will delay the swing timers by the cast time. I've double checked this with a combat log run, instrumented to show the cast time and MH swing time left when casting starts. It produced this sequence:

0h 0m 35.08s MH crit 3051.52
0h 0m 35.08s Maelstrom Weapon stacks (1)
0h 0m 35.27s OH hit 713.127
0h 0m 35.27s OH Flametongue hit 429.343
0h 0m 35.55s Magma totem hit 828.765
0h 0m 35.81s Spirit Wolf crit 1201.92
0h 0m 35.81s Spirit Wolf crit 1201.92
0h 0m 36.2s Global cooldown off
0h 0m 36.2s Casting started (126, 7) <- 1.26s to cast, 0.07s originally remaining on MH swing
0h 0m 36.77s Spirit Wolf hit 600.962
0h 0m 36.77s Spirit Wolf hit 600.962
0h 0m 37.25s Global cooldown off
0h 0m 37.46s Lightning Bolt hit 4231.89 <- LB cast 1.26s later as expected
0h 0m 37.51s Earth shock crit 6034.34
0h 0m 37.51s Elemental Focus procs
0h 0m 37.53s MH crit 3051.52 <- MH swing 0.07s seconds after

So it looks like there is a bug in this. I also noticed that chain lightning seems to be given a 1.5s base cast time, which should be 2.0s instead.

If there are no objections, I'll go ahead and fix these. In particular the MH swing would have reset at 36.27 without firing (similarly OH) and the CL base cast time will be 2 seconds.

Good catches, really glad you are on top of it. I think the old model when Tukez was designing EnhSim was that it paused the swing timer during a cast. That was how the system was initially set up (at least that was how it appeared to work from testing on the closed beta). When they pulled Lava Burst from Maelstrom Weapon the mechanic appeared to have changed to it's current state, although it wasn't discovered till later.

Any idea on when you can have that ready for testing? You've got me very curious to see how the sim results will change once you get it working properly.

I think I have got it working, but of course if folks can run some checks that's great. This release includes a combatlog build which I've already used to verify that the clipping should work.

EnhSim - Release: v1.8.1

Using the config currently in the BiS thread I'm getting 7401.05 which is actually a small increase over 1.8.0. It seems that out of the potential swings, 2.9% are getting clipped, but I guess the fact that the rest don't get delayed makes up for it (and the CL cast time).

I'll wait for some confirmation here before updating the BiS thread. Thanks Sylvand.

[Enhancement] 3.1 changes - new rules in place

In theory, a randomly timed cast should give the same dps with clipped or paused MW behavior in the sim.

However, clipping can be more streaky (lots of clipped swings in a row, or lots of well timed casts in a row).

Now that Enhsim models clipped swings, there are some interesting experiments that could be done by modifying the simulator.

• What is the dps difference if you time the cast so it doesn't clip the swing?
• What is the dps difference if you only use MW4_LB and MW3_LB when WF is on cooldown?

I don't attempt to time my casts because random casts are better than poorly timed casts.

When I've tried to time my casts against a training dummy I mistime about as often as I get it right. Perhaps someone with more skill/latency could do better.

LD50_abar is a mod that claims to handle dual wielding, however it looks to have problems modeling flurry (something causes its display to get out of sync when the combatlog says that it's still in sync). People would probably want to fix this mod before relying on it.
LD50_abar - Addons - Curse

Yeah I agree about the expected difference, both should effectively remove swings in equal proportion to the cast time. I might end up playing with clip avoidance just for a theoretical data point.

It has been pointed out that Comet's Trail incorrectly uses the tooltip's 612 haste proc instead of the 726 it actually gives (Comet's Trail - Spell - World of Warcraft). I've checked in the fix for this so it will be in the next release (or if you want to build it yourself).

The combatlog genorated by enhsim 1.8.1 seems to be ignoring some GCDs

0h 0m 0s Haste Potion used
0h 0m 0s Spirit Wolves casted
0h 0m 0s MH SS hit 734.985
0h 0m 0s Fury of the Five Flights stacks (1)
0h 0m 0s Totem of Dueling procs
0h 0m 0s OH SS hit 342.386
0h 0m 0s OH Flametongue hit 299.678
0h 0m 0s Fury of the Five Flights stacks (2)
0h 0m 0s Bloodlust gained

Stormstrike,Bloodlust and Spirit Wolves are casted the same time, It could it be assuming another shaman in raid is casting blood lust

I played around with EnhSim 1.8.1 and found a bug. When simming the T9 4-piece bonus, the simulator looks to be adding 25% damage to lightning bolt, chain lightning, and lava burst (as well as to shocks). Looking at my sim results, all of those abilities had a 25% increase in damage which led to an enormous dps increase.

I haven't done any programming in years and I am not familiar with the EnhSim codebase, but I was bored and took a quick look at the source code on codeplex and I think I found the problem. All of the shocks, lightning bolt, chain lightning, and lava burst appear to use "shock_damage_multiplier" as a component in calculating damage (in skills.cpp). That is because "shock_damage_multiplier" is the variable used to apply damage from the Concussion talent (which does affect all those abilities) -- the variable name is misleading since it affects more than just shocks. The 4-piece bonus is coded as: "shock_damage_multiplier_ *= 1.25f" (in enhsimconfig.cpp) which inadvertently affects all those skills.

I'll take a look at this tonight. I'm really not sure if it's good to assume someone else casts it; maybe it should be configurable?

You're probably right. I'll take a look at this tonight as well.

edit:
The T9 set bonus should now only include shocks, but it still currently stacks multiplicatively with concussion where applicable (I can't check myself if this is right). Wolves will also trigger the GCD, though I've left bloodlust as-is for now. If we had the source for the GUI (I don't think we do) I would feel better about adding some more config around bloodlust (maybe having it in the priority rotation) but for now I'll just let it assume someone else casts it immediately after the first skill (generally wolves).

I also found a bug with the change I made for clipping swings, though you'd only hit it when casting without maelstrom (i.e. the swing timer resets). I hit it while testing the numbers on lava burst when it would get cast without maelstrom.

I've gone ahead and put this along with the comet's trail fix into v1.8.2:

EnhSim - Release: v1.8.2

I got some new gear today that meant taking gear off with Expertise Rating on so I fell below cap. I had been at 140 Expertise Rating and I dropped to 104. I thought I could just slap two 16 Expertise Rating gems into my gear and still be on/over cap. I tested it on EhnSim with 3/3 UR and 136 Expertise Rating and it said I was over cap. When I gemmed it that way in game it had me at 25 Expertise. I messed with my gems to where I could get to 139 Rating and was still at 25 Expertise.

I know that Expertise Rating does give a point by point benefit where we once didn't think it did so maybe EhnSim was right. However the game telling me I was below cap made me want to make sure so I figured someone on EJ would know the exact rating I need to hit the Expertise cap (even if the game says I am not).

Of course I looked around a bit before posting here but I was unable to find a definite answer. Can someone please tell me how much Expertise Rating I need exactly to hit Expertise cap (even if the game says I'm at 25 Expertise and not 26)?

Thanks.