Edit: delete please, useless answer.

You are right, in fact due to a coding error, the Raise dead cooldown is shorter than ghoul uptime. Unfortunately, I can't correct this and make it available right now, I'm making some major change in DRW management.

I'm getting similarly weird results with an Unholy setup. After multiple 24hr runs, some seem to give rational results, others give similar results to those Illandras posted (e.g. with -600 "Strenght").

I have a possible solution.

In the character.xml file, if you under weapon dps uses a decimal number, for example 222.9 it will give huge dps numbers, and weird EP results. However if you round it up to 223, it will display normal results.

Doesn't seem to help here; rounding my weapon DPS figure to 0 decimal places still gives me unreasonable results (such as -310 "strenght").

Is there a problem with desecration?

I just ran a couple simulations with the 12-00-59 build and then edited it just removing desecration (setting it to 0) and didnt get any noticable difference at all.


Otherwise thanks for the nice work.

It's possible that haste levels are too high. While Haste got a 30% melee boost, it's still the same for spell haste. If he increased spell haste by 30%, since 0/20/51 tends to fill all of it's GCDs, that would effectively increase the damage of all the spells. If you fill all GCDs, that would represent a 30% increase in the damage potential of Haste. So if you had 350 Haste Rating, you would get a 13% boost in damage rather than 10%.

A lot of the things in 0/20/51 got buffed. The buff to Armor Penetration, melee haste, 10% increased shadow damage, Gargoyle, Plague Strike, and Blood Strike are all buffs. Icy Touch, however, got the Nerf Bat of Doom. It lost 15% from Improved Icy Touch, 10% due to no Glacial Rot, 20% from Black Ice, and 5% from Impurity. It gained 5% from Razorice. That meant the Icy Touchs that were needed to generate runic power would scale poorly compared to strikes.

So if the AP levels were lower, it would look good. As you get more stats, it should get further behind.

I know all of this, but where it looked like a good chunk of damage was coming from was DC. It said it had an average of about 4k with a 30% crit which I thought was pretty good damage. When I tested the DnD spec also, it showed DnD for only 1% of Damage and similar DC stats but I think the DnD percentage is way too low...

Here are the last addition/correction to the simulator, thanks to your feedback.

0.7.4 Release notes
* Much better Dancing Rune Weapon simulation.
* Untalented ghoul uptime correction.
* Desecration is now resetting between simulations.
* Pestilence Glyph and spell implemented. The simulation will always use Pestilence to refresh disease if glyphed on “Priority” mode.
* Add a warning message when EP calculation time is inferior to 500 hours


I also been surprise by Death Coil damage, but keep in mind that on 3.1 patch, it win 15% more damage due to Dark Death glyph and 10% from Black Ice.
To clarify on haste, only the melee haste is affected by the 30% bonus.
Death and decay is not managed efficiently in priority mode. The simulator will try to cast it only if the 3 runes are available and I have not made it smart enough to wait for the missing runes and not use the available one. I suggest you to tune a rotation based on your needs.

I've been simming a few dual wield builds, most recently the 10/10/51 build in unholy presence. I've been getting some wierd results when swapping the main hand weapons. Silent Crusader - 156.6 DPS, 2.5 Speed is simming much lower than hailstorm 156.7 DPS, 1.5 speed. That wouldn't bother me over much but the bulk of the lost DPS is coming from main hand damage, to the tune of over 30 mil (20% down to 12%) in a 24 hour simulation. I'm hard pressed to explain why the main hand would lose such a large portion of damage with similar DPS weapons of differing speeds.

I'm trying to make a Rotation that includes GF but I haven't been able to figure out what command you allow for GF to have in rotations. I've tried:

<GF retry="1"></GF>
<GF retry="0"></GF>
<GhoulFrenzy retry="1"></GhoulFrenzy>
<GhoulFrenzy retry="0"></GhoulFrenzy>

But they all break the rotation and nothing else in the rotation proceeds. But this has given me an idea for the Dps we do by just auto-attacking. It's doing about 2.2k with my stats. It's actually a combo of Necrosis, White Swings, BCB with no diseases, Ghoul, and a tiny bit of RI damage. I just thought that was very good.

Edit. Delete this post. Answered my own question.

I have corrected 2 things, thanks to your feedback.

0.7.5 Release notes
* Correction on BCB.
* Correction Main Hand damage that should not be normalized.


Kyruski,
Ghoul Frenzy is ready in the code, but I'm still not sure how to implement it. There are 2 options I can't decide.

• Priority/Rotation. You will have to include it on the config file.
• Automatic. The simulator use it every time it's up with blood tap.

Kal Choedan,
Very strange result for a long simulator run. I'm looking deeper into it.

I would say priority/rotation. However, add a prompt for using GF alone whenever it's duration has expired and another for using GF only when you can blood tap for the rune to use it. If that's workable for you. Only because I'm sure some of us would like to see how it works if we always keep it up but it would be nice to compare to only using it 30 sec/minute when we can tap for the rune.

<GhoulFrenzy> and <TapFrenzy> or something like that.

The 00-10-61.xml appears to be missing 7 talent points.

I created an issue and attached a diff with the proposed changes to the spec template.

Issue 2 - dksimulator - 00-10-61.xml is missing 7 talent points - Google Code

I'm having a lot of trouble getting the EP calc to work as well, I ran 4 different 5000 hour tests and all of them seemed to give completely random numbers, half of which were negative. What does the EP base actually do? i've never changed it from the default 50, could that effect something?

I did following to get EP numbers for myself:
- lowered hit and exp to 0 in character template (as I want value of both before cap)
- changed base EP to 150 (as 150 exp rating will still be before cap)
- ran ten 24 hour tests, dropped lowest and highest EP value of each stat and got avg. from 8 left
- for hit "after melee before spell cap" I just took what percent of damage would be affected (DC, UB, Gargoyle, HB, IT) and multiplied it with pre-cap value (example: 2.0 * 10% = 0.2)
- I rounded up values some, there is no real difference between for example 1.4 and 1.4146546
- I ignored weapon dps, values were seriously random and you will want higher dps weapon anyway so why bother?

Obviously 10 x 24h sample is not super accurate but you can increase both number and duration of simulations for more accurate results. Just remember to drop 5-10% lowest and highest numbers of each stats to lower RNG.

I am not saying this will get your perfectly perfect EP values but it should be good enough to point out potential upgrades and numbers I got looked correct.

0.7.6 Release notes
* Butchery not resetting well between simulations.
* Real time DPS available in the bottom of the application.
* The report is updated more often on EP calculation.
* The glyphs now have to be configured inside the template configuration file. The default templates are updated.
* The 00-10-61 template is corrected.
* The Weapon DPS accept decimal on the character configuration.

I hope the strange EP value was from Butchery not resetting between simulation. From my test, the DPS is more stable between runs. But I wait your feedback.

Fugazor, your way to obtain EP might be a good walkaround, but honnestly I want that the EP calculator of the simulator as a "push and wait".

Twistedmoon, I hope your super high dps come from a digit in the Weapon DPS (corrected now), because you should'nt have such DPS.

Ghoul Frenzy will be available in the next release.

I noticed a much slower simulation time with 0.7.6, also the simulator seemed to give exceptions and had to be restarted. If it happens again I will pm you the error log I recieved.
The huge dps values seem to be gone, and the new way to edit glyphs is great.

Would it be possible to add the tier bonuses in the same way as glyphs? or at least let us edit them inside the application so you don't have to close it down after a change?

I tried to look for the few lastest updates, but I can't find if the current version takes rime procs in to count from obliterate?

I'm testing a rotation like this with Howl glyph adding frost fever (only use howl when refreshing disease)

HB-OB-BS-BS-FS-FS
OB-OB-FS-OB-FS-FS

so, if one of the 4 OB proc freezing fog, the first HB would be free of charge, therefore giving you an extra OB in it's place.

Clearly you can't do it a solid rotation, so how would this go in the priority?
(With howl glyph, it would use howl to apply frost fever?)

Frost fever > obliterate > blood strike > frost strike , this would fail in the last part of the rotation, since you'd have one death rune ready before second parts first frost strike. Would prolly do HB->OB->BS->BS->FS->FS -> OB->OB->BS->BS->FS->FS->FS

and

Frost fever > obliterate > frost strike > blood strike , this would fail at the start by doing HB->OB->FS->BS->BS->FS and would delay the second parts ending even more.

I can't find a way around it, so I need your help

0.7.7 Release notes
* Simulator now load previous setup on start up.
* Add an option to display real time DPS.
* Ghoul frenzy implementation. All templates are updated. If talented it will use automatically it with Blood Tap, unless added to priorities/rotations.
* Set bonus are now configurable via the character config file.

Illandras,
I have to leave all the credits to the idea of integrate glyph into talent spec config file to EvilNuff.
I will look at your crash problem.


Energizr,

Yes Rime proc are generated by Obliterate.
Try this rotation, the 2nd Obliterate will be skipped if runes are not available, but you have to be hit/dodge capped as is will not retry it.
However, you should try a priority like this one. It will use every Rime proc and it look like it give more DPS with the simulator.

What is this command, <GhoulFrenzy>?