Regarding rhadatip and it's scaling values, i have ran myself through simcraft to find the stats scaling values however am not sure what values to put on the 2pct9 and 2pct10 to use rhadatip accurately.
I believe this method was posted a few weeks ago, and it worked great for me:
Import your character to simcraft as usual, then without changing any of the stat values on the gear rename the items to that of the set you want to test. For example I was wearing 4pc t9 at the time with chest offset so I ran a simulation as imported, then renamed two pieces to a neutral name (ex. "legs=guldans_leggings_of_triumph," to "legs=offset_legs," ran the sim again, renamed the last two pieces, ran it again, then edited the names again to t10 pieces two at a time. In then end you have 5 simcraft runs all using exactly the same stats but with, respectively, 4pct9, 2pct9, 0pc, 2pct10, and 4pct10 values. The difference between each run should be the value of that set bonus, with no correction for the actual dps value of the gear itself.
It feels like a bit of a hack job, but even if this isn't the correct way to do it it at least provided values eerily similar to those posted in other threads around here recently.
I had a quick question about stat weights when running SimCraft.
I can't say I know exactly how they work, but I thought they were relatively, if not completely, the same for any character. However, when running SimCraft for my imported character, SP beat out Haste. Can someone explain to me why and what I should do? Do I regem SP where I had Sp/Haste gems?
Stat weights are not static. They fluctuate in many different ways (relative amounts of other stats, fight types, different specs, glyphs). The best way to calculate these is to run simcraft everytime you change gear, respec, or lose a consistent buff in your raids.
I noticed Simcraft's new gui has a nice little option to specify a number of adds for a fight. However, I suspect I'm doing something wrong using this option. I would expect to see an increase in the % uptime of Eradication if there are more adds being dps'd. Is this an invalid assumption? When I add or remove adds from the simulation, I see neither a significant change in dps nor an increase/decrease in the uptime of Eradication. Is the add function bugged, or does the simulator ignore them completely?
Edit: a little digging indicates that simcraft doesn't calculate dotting multiple targets (and some AoE abilities).
I've tested first with 2 [RoTD] then [1 RotD(heroic)]+[Flares of Heaven] then [RoTD(Heroic)]+[Nevermelting Crystal].
Iteration: 1000
Length: 500
Smooth RNG: Yes
Patchwerk fight style.
Well, i've seen simcraft result from other people in this forum and conclude that RotD(h)+ Crystal will fare highest DPS( for my current gear). but my result show otherwise. Am i doing something wrong?
The val’kyrs spawned by the Nibelung staff will now always hit with their smite attack and their damage has been increased for both normal and heroic versions of the staff.
It's hitting for ~1800-2000 now with curse of elements up on a level 80 target dummy. (No partial resist)
The val’kyrs spawned by the Nibelung staff will now always hit with their smite attack and their damage has been increased for both normal and heroic versions of the staff.
It's hitting for ~1800-2000 now with curse of elements up on a level 80 target dummy. (No partial resist)
Wow, that's nearly double its previous damage range, which was about 1050 to 1200 iirc. I'm going to try to do some testing this weekend on this item if I can find the time and will post results. Despite the fact that it’s a “known” 1% chance to proc, I believe (based on multiple reports on internet forums and my own limited experiences) that the tooltip word “harmful” is being interpreted in error. This staff may be quite a bit better for some specs than we had previously thought
Couple Examples:
- There have been reports that spells that apply secondary effects (like imp shadowbolt and imp blizzard) get 1% for the spell and another 1% for the effect applied (i.e. those debuffs can effectively double or triple the proc chance. I plan to test this extensively on target dummy by going affliction and spamming shadowbolt (once corruption is up). That should make the proc chance 4%, which would be easily distinguishable from 1% with only a moderate sample size.
- Last night on Heroic Faction Champs my job was to solo the druid. I cast only banish, fear, and spell lock the entire fight and procced two valkyr’s (it was in fact a bit annoying to have to spam the pet defensive command to keep her from breaking fear). This supports the notion that the word “harmful” on the tooltip may be more broadly defined then had previously been thought.
I did a target dummy test where I casted shadow bolt 1000 times as affliction with corruption up. I procced 26 valkyrs so 2.6%.
Given my previous post I'm not sure what to make of this except that there is/are clearly some secondary effects that can cause procs. I think its safe to say that for affliction, shadow bolt certainly has a chance greater than 1%, however.
I guess the next test is to do it without corruption and then maybe as demonology to isolate the various refreshing debuffs?
Update:
I did another test of 1000 shadow bolt casts, this time with no corruption (or anything else). I procced 33 valkyrs. 3.3%.
Based on the above two tests my current hypothesis is that shadow embrace and imp shadowbolt (and nearly certainly other debuffs) get 1% to proc each time they are applied. Refreshing corruption does not get 1%. I think this is because refreshing corruption doesn't technically overwrite the debuff like applying shadow embrace or imp shadobolt does.
Next up is to do the test as destro and perhaps demonology. I may also start looking at a way to arrive at true total proc percentage considering all spell casts and their individual chances.
2nd update:
My account was compromised and I anticipate getting everything restored will take a couple weeks so I won't be able to do the testing planned. I would be grateful if someone would do the additional tests I described.
Regarding Nibelung:
Have had a play around with this (ungemmed/unenchanted as of yet) on the target dummies a few times.
With CoE up: the proc itself seems to have around 25 seconds of actual DPS time, it runs around for 5 seconds prior to casting, and does around 24.5k dmg (not sure how it scales with buffs/debuffs/dmg, although it doesnt seem to recieve demonic pact).
I only did 3 quick tests on the dummies, for about 1 million damage each.
1st: Procced once around 450k.
2nd: Procced twice, around 400k and 800k.
3rd: Procced once around 600k
Each time I have tested (in these conditions, and throwing a few random spells between arena games) the pet has done about 940-960 dps.
So the pet itself does nice dps, thats undeniable. But its uptime appears to be pretty terrible.
Interestingly, the first time i equipped the staff and just cast a few spells on the dummy (with no real intention of checking any dps) it procced once quite early on, then again about 10 seconds after it dropped, then again while the 2nd was still up.
Maybe I was unlucky during testing (or lucky the first time). Would be interested to see some more sustained figures.
Please take these changes with caution.
A change like that for Destruction will not cover the close to 1000 DPS gap between Destruction and Affliction right now(with max ICC gear).
15% more imp damage would translate into roughly 150 DPS gain, thats ~1.3% increased damage.
I've looked for a long time and haven't found the simcraft tag for SoC, seed_of_corruption wasn't working out for me. I want to find the best AoE trash build, and wanted to test it out... It'd be a good waste of an offspec, but I want something better than just pulling my succy out for 5% more crit.
Import your character to simcraft as usual, then without changing any of the stat values on the gear rename the items to that of the set you want to test. ...
I just wanted to mention this since it might be helpful for a few people and many don't seem to know this. There is an easier way than to rename items to exclude set bonuses. SC supports a simulation setting (enter it in the simulation tab) to disable certain set bonuses. The syntax is: tier(6|7|8|9|10)_(2|4)pc_(caster|melee|tank)=0
For example to disable my current t9 2pc bonus as a warlock I'd use
tier9_2pc_caster=0
Now to calculate the value of a set bonus just run it once with and once without the setting. The dps difference is the value of the bonus for your current gear combination.
Along the lines of what Flyfunner posted Here in the affliction thread, how would one edit the simcraft file to simulate the warlock recasting corruption and rolling the boosted damage buff from 4pt10 and possibly tricks?
Last edited by Calidus : 01/13/10 at 12:35 PM.
Reason: spelling
Along the lines of what Flyfunner posted Here in the affliction thread, how would one edit the simcraft file to simulate the warlock recasting corruption and rolling the boosted damage buff from 4pt10 and possibly tricks?
Right now it is not easy to re-cast a DoT early. (ie willful clipping)
This is in the (slow) process of being changed. DoTs will no longer have an implied ready-check regarding their duration. Instead, it will have to be explicitly specified in the action list.
Remember that you will not always be able to recast. If your ring or lightweave or trinket or someone else's SP buff was there on the original cast but no longer is, then you will get the "A more powerful spell is already active" warning.
Is there a method when using the param wowhead=1234567 to pull the gear from wowhead, and the talents from wowhead as well so that we can model slight variations in spec without have to respec?
Has anyone found a way to "hardcode" Dark Pact to the action list for 56/0/15 ? I'm doing it manually each time now but that gets annoying after a while.
The way I use Simulationcraft is with a parser like like World of Logs, though there are some nice standalone ones too. Simulationcraft can tell you the cast interval of each spell and % of DPS and these are the 2 main factors you can control directly. For example Simulationcraft might tell you that it calcs Shadow Bolt to be 25% of your DPS and a cast interval of 2.5s, but when you check your log parse you find it's 20% of your DPS and a cast interval of 3s. Note, with WoL you manually need to check the cast interval by doing
fight time / number of casts
Then as a sanity check I find the highest DPS lock on WOL and compare their cast intervals and % of DPS per spell to see if the optimal 'rotation' is remotely achievable.
Basically that little block of output data on page 1, or better yet from your own running of Simulationcraft, is the data you want to analyze and you just want to compare all of that with your own parsed logs.
ie. compare this with this.
The flat line is a haste cap correct? Also does this graph factor in spellstone or not? I need to know if simcraft is factoring spellstone for the stat weights or not as demonology. 240 haste is pretty major when calculating all of this which is why I ask.
for the record I have 916 haste in this profile and 1156 haste with spellstone. Either I'm 20 haste from a cap or I'm 50 haste over already :o
edit: Also is there a way to make this graph display more than the +/- 200 ratings for a stat?
Simulationcraft should always be viewed as a guide. You should never take too seriously the number it gives you, but that doesn't mean to say you should never strive to come close to it. Comparing logs with simcraft is the best way to see where you are going wrong, taking into account the fight. On any Patchwerk fight for example, it's likely Shadow Bolt will do a higher percentage of your damage, however on fights with adds to think about, Corruption will most likely take over. Guilds often do fights different also with some warlocks asked to stay on the boss, some on the adds. Always take this into account when comparing with Simcraft and indeed any other warlock on WoL or others.
The main thing I find to be most interesting is up times. For some reason, the Nevermelting Ice Crystal doesn't show up on my logs, has anyone else find this? But things such as the life tap buff, uptime of dots etc can really tell you what you are missing. For example, I know I slack on keeping haunt and Unstable Affliction up on add switching fights, so I know this is something I need to improve on. Active time is another important one. I often find myself lower down the meter simply because my active time was not quite as good. If you remember there being no excuse for it, you can improve on it.
01/06/10, 4:31 PM
