Dots and you: The Affliction Warlock Thread - Page 33

duhwhat · 03/23/09, 2:45 PM

Originally Posted by tuberqlosis

I was just messing around on the PTR talents for ourselves and I came up with 53/0/15 spec with 3 points left. Note that I have 16.24% hit (yes, that could be made more exact, I already know that), so I have only placed 1 point in Suppression. I also prefer to be self-hitcapped so I'm not reliant on anyone. Anyways, even with 2 more points in Suppression, I have 1-3 points left in my spec, with very little in the ways of improvement. Assuming I don't want to buff my pet through either improved imp or demonic power, i think my best options are spell pushback reductions (fel concentration for affl and intensity for destro). With the "wonderful" simplification coming to affl, I wonder if intensity would be a better option than fel concentration, as it appears we'll be seeing more sbolt spamming. With my 3 points I could even max intensity and put 1 into fel concentration, achieving 54/0/17. Any thoughts?

I think not relying on a hit debuff when considering a build is a mistake if we're talking about an optimal raid (which is what is going to be assembled for progression). That said, hit is only going to become more of a commodity on gear as tiers progress, so the consideration will become where to put the points from Suppression. What are your reasons for not including Demonic Power? Previously it's been shown that the DPS gain from the mana conservation aspect of Suppression and Cataclysm are negligible, so I agree with your reasoning for going with pushback talents. I don't think Improved Drain Soul is necessary and would favor more points in Fel Concentration. As you spec out of Suppression and into Fel Concentration, and assuming Demonic Power is taken, I feel a single point in Intensity is the best bet, rather than taking a point in Cataclysm as is shown in the default Affliction build. If the BiS setup for Ulduar gear happens to include 11/12% hit, pushback protection is probably going to be superior in practice to the dps gain from mana conservation.

Talent Calculator - World of Warcraft

tuberqlosis · 03/23/09, 7:23 PM

Chopping your post to parts I will be commenting on...

Originally Posted by duhwhat

What are your reasons for not including Demonic Power?

I don't think Improved Drain Soul is necessary and would favor more points in Fel Concentration.

As you spec out of Suppression and into Fel Concentration, and assuming Demonic Power is taken, I feel a single point in Intensity is the best bet, rather than taking a point in Cataclysm as is shown in the default Affliction build.

Talent Calculator - World of Warcraft

I am not sure whether push back reduction or a slightly faster imp fireball is better, and I personally would rather improve myself over my pet as affl(considering DP would be the ONLY pet buff I went for, unlike demo or destro). If DP is more of a DPS increase, then that is where the points would go.

The reason I take Imp DS is that I am threat capped as affl now. My gear is a mix 200/213, but I have even had significant threat problems in 25s. The most notable is Maly, where I ride the tank REAL hard. The argument could be made as sbolt spam increases that Destructive Reach, the destro equivalent, returns more overall threat reduction, but I feel locks generate enough threat to mandate the reduction.

I do agree with you that cataclysm, especially with the hit removal, is damn near worthless (even when looking at destro I hate its 2nd tier talents).

Going with 53/0/15, I have 3 points left, but I've already maxed out affl pushback. With these last 3 points I'm looking at some combo of Imp improvement via Demonic Power and/or Improved Imp, threat reduction via Imp DS and/or Destructive Reach, or destro pushback via Intensity and 1 point in the previous fields. However, I've heard the succi's DPS beats the imp's, so unless I hear otherwise I'm probably going to avoid Imp Imp. With these thoughts, my considerations and questions are as follows:

1. With the decreased CD on DG, he will obviously see more action. As his presence increases, is it worth buffing the decreasing presence of my "normal" pet?
2. With the increase in sbolt spam, should I prioritize threat/pushback on destro spells over their affliction counterparts?
3. With my limited 3 points, will the imp surpass the succi in dps (assuming as affl I'm not devoting a glyph slot to a pet and that the succi does in fact currently surpass the imp)?

• Heeno · 03/23/09, 7:42 PM

1. Remember now that Ulduar is here you will be wiping, and your doomguard is only good for one attempt. The average guild does one attempt every five minutes, so your doomguard would be up one out of every six attempts.

2. Even still with all of the changes to affliction, the affliction portion of the rotation still does more damage than the destruction portion, meaning you should prioritize threat reduction for affliction. However, affliction is composed of of DoTs and high DPCT spells, which do not benefit as much from spell pushback like a filler, shadow bolt does.

3. The best way you could deal with those three points is to move two points out of fel concentration into suppression, use a point to finish off suppression, pick up demonic power, and drop 3% hit from your gear for more spellpower.

cjwprostar · 03/24/09, 12:47 PM

Sorry if this isn't the right place to post this, but since Pandemic now makes 2 of our dots crit in 3.1 and I believe we will be casting more sbolts due to the removal of siphon life and immolate from our rotaion, would it be worth it to use a [Grand Firestone] instead of a [Grand Spellstone]? I've done some limited testing on the PTR but with mixed results. If anyone has done the math that would be great.

gherkin · 03/24/09, 1:16 PM

Originally Posted by cjwprostar

Sorry if this isn't the right place to post this, but since Pandemic now makes 2 of our dots crit in 3.1 and I believe we will be casting more sbolts due to the removal of siphon life and immolate from our rotaion, would it be worth it to use a [Grand Firestone] instead of a [Grand Spellstone]? I've done some limited testing on the PTR but with mixed results. If anyone has done the math that would be great.

Shadowbolt will need to comprise more than 50% of your total DPS to justify the firestone. Run a rotation test - my Sbolt ends up being ~35% of my damage done, with Corruption/Agony/UA/Haunt/DS making up the other 65%.

1% of 65% > 1% of 35%, regardless of combat ratings.

Shayo · 03/24/09, 1:50 PM

Follow up: after upgrading to the newest version of simulationcraft and running the tests properly (no demonic pact warlock in one raid and not in the other), I'm consistently showing spell stone to be a 21-26 dps gain over fire stone.

DPS Ranking:
26786 100.0% Raid
6871 25.6% Warlock_T8_53_00_18_SP_DG
6845 25.5% Warlock_T8_53_00_18_FS_DG
6562 24.5% Warlock_T8_53_00_18
6541 24.4% Warlock_T8_53_00_18_fs

cjwprostar · 03/24/09, 2:51 PM

Originally Posted by gherkin

Shadowbolt will need to comprise more than 50% of your total DPS to justify the firestone. Run a rotation test - my Sbolt ends up being ~35% of my damage done, with Corruption/Agony/UA/Haunt/DS making up the other 65%.

1% of 65% > 1% of 35%, regardless of combat ratings.

Well that's one reason I couldn't figure it out. My sbolt makes up 41.7% of my damage and I would think Haunt counts as a direct damage spell when it comes to the firestone and that was 5.4% of my damage, making my total direct damage ~47.1%.

Originally Posted by Shayo

10k Iterations w/firestone: (weird % numbers here were because I simply included the other warlock builds)
6955 13.0% Warlock_T8_53_00_18_DG
6644 12.4% Warlock_T8_53_00_18

10k Iterations w/spellstone:
DPS Ranking:
6869 51.2% Warlock_T8_53_00_18_DG
6558 48.8% Warlock_T8_53_00_18

So it looks like the spellstone wins out but only by ~85dps. Thanks for the help

Galanna · 03/24/09, 3:49 PM

Originally Posted by Shayo

10k Iterations w/firestone: (weird % numbers here were because I simply included the other warlock builds)
6955 13.0% Warlock_T8_53_00_18_DG
6644 12.4% Warlock_T8_53_00_18

10k Iterations w/spellstone:
DPS Ranking:
6869 51.2% Warlock_T8_53_00_18_DG
6558 48.8% Warlock_T8_53_00_18

Exactly, with those tests the spellstone is losing by ~85 dps, but they were not done with the same raid, maybe causing synergy that increased firestone DPS, synergy that spellstone DPS did not received.

Kabale · 03/25/09, 12:26 PM

Originally Posted by Galanna

Exactly, with those tests the spellstone is losing by ~85 dps, but they were not done with the same raid, maybe causing synergy that increased firestone DPS, synergy that spellstone DPS did not received.

:|

SimulationCraft models a perfect raid setting in which all possible buffs are present. The results from Shayo's testing show that under ideal conditions, the firestone will yield 85~ more DPS than the spellstone for the specs tested.

Shayo · 03/25/09, 12:29 PM

Originally Posted by Kabale

:|

SimulationCraft models a perfect raid setting in which all possible buffs are present. The results from Shayo's testing show that under ideal conditions, the firestone will yield 85~ more DPS than the spellstone for the specs tested.

And again I want to point out the follow up I did with a more recent version of the sim shows the spellstone with 20 dps higher. I'll edit the original and take that portion out since it's just confusing people.

• Zakalwe · 03/25/09, 12:41 PM

Originally Posted by Kabale

:|

SimulationCraft models a perfect raid setting in which all possible buffs are present. The results from Shayo's testing show that under ideal conditions, the firestone will yield 85~ more DPS than the spellstone for the specs tested.

The problem is that the "perfect raid" simulated by setting optimal_raid=1 assumes a Totem of Wrath but no Demonic Pact. One of his tests included a deep demo warlock actor with DP while the other did not. DP + ToW is currently quite a bit better than just ToW, due to trinkets and other procs putting the DP buff at over 280 for significant portions of the fight.

Bottom line, Spellstone is (obviously) the best choice for affliction.

shauno · 03/30/09, 7:37 PM

Moving away from PTR discussion right now.
I've been working on an application to calculate a full overall set of gear to conform to 289 hit rating while having the highest DPS score based on the following stat weights:

        
        float weight_Int = 0.16f;
        float weight_Spi = 0.34f;
        float weight_Hste = 0.41f;
        float weight_Crit = 0.25f;
        float weight_SP = 0.66f;

Obviously hit rating no longer needs to have a weight as every gear combination will be tested and if the hit rating is below 289 it is not suitable, if the hit rating is above 289 it will find the set with the highest DPS score based on the above weights.

The enchants stay the same as the highest DPS score for each enchant is obvious except in the case of Boots and Gloves (Icewalker & Precision) it will test with these enchants too. The app will also gem gear and test if the socket bonus is worth it and apply the appropriate coloured gem (whichever has the highest DPS score for that socket colour), if not it will just load up the sockets with Runed Scarlet Rubies.

I haven't added any weights on T7 2/4pc bonuses, nor trinkets at this time, right now it assumes Dying Curse/Illustration of the Dragon Soul as static trinkets. I also haven't added the 2 hit gems I want to yet (+16 hit and +8 hit/haste) as another possibility, much like the hit rating enchants, since you never know.. (highly unlikely) but that'll come soon.

Right now it is just grabbing for each slot (other than trinkets) a list of items from Wowhead based on the criteria of having iLvl 213-226, having no melee stats and at least either > 0 DPS score or > 0 hit rating. Based on this we get about 1.4 trillion different gear combinations, the algorithm is slightly optimized so we only test about 1/3 of that. On my work E4400 2.0ghz Dual core CPU this took about 24-30 hours and only needed to test 550 billion combinations.

Anyway, that's the brief explanation, I hope to add weights in for T7 bonuses, the hit rating gems and weights for trinkets soon, without further adieu, here is the latest gear combination it gave me... thoughts/suggestions for all this?

chardev.org v6 ~ a World of Warcraft character planner

Once Ulduar comes out I will easily be able to reconfigure the stat weights, any other weights, any new gems and socket bonuses and to get the new lists of gear.

angaroth · 03/30/09, 8:10 PM

Interesting. You might want to add 3 tweaks -

1) Another run where you set the hit cap at 368 and see what the gear cost is, compared to 3 talent points that might be worth more, depending on your spec.

2) Allow for the Ring of the Malevolent. A 200 ring, sure, but lots of hit.

3) Force 4pcT7 in each slot. (i.e. a run for 4pc t7 except head, then except for shoulder, then except for ...) It is certainly worth something at the moment, and even more once the Glyph of Life Tap becomes available.

• Heeno · 03/30/09, 8:16 PM

It appears your simulation is fairly inaccurate unfortunately. [Wyrmrest Necklace of Power], [Pennant Cloak], [Unsullied Cuffs], and [Arcanic Tramplers] are all by far BiS, and are not included. Some options chosen do not make much sense, such as the neck with mp5 on it. Also for 3.1 4-piece valorous is optimal due to the spirit gained through the bonus increasing the spellpower gained via life tap glyph.

shauno · 03/30/09, 9:14 PM

Originally Posted by Heeno

It appears your simulation is fairly inaccurate unfortunately. [Wyrmrest Necklace of Power], [Pennant Cloak], [Unsullied Cuffs], and [Arcanic Tramplers] are all by far BiS, and are not included. Some options chosen do not make much sense, such as the neck with mp5 on it. Also for 3.1 4-piece valorous is optimal due to the spirit gained through the bonus increasing the spellpower gained via life tap glyph.

I will go over the process again to double check everything but I encourage you to make a full set out of BiS items and see if you can beat that set in terms of DPS score and getting close to that hit rating figure. The only place where I can see a shortfall in my system is that it hasn't attempted to use any gems with hit ratings which may penalize items with sockets but no hit rating, maybe.

That's a good idea in terms of adding variations on the T7 set bonuses, I will tweak it to spit out the highest DPS score set while keeping any variation of the 2pc/4pc bonus, just for reference, I'd like to do something similar with trinkets too. Will also add the Signet of the Malevonant (using it myself even already) to the list of Rings.

Thanks for the feedback, I will definitely go over it again tonight and double check it's totally accurate. I'll put a trace log in for those specific BiS items mentioned and when it hits that set, I'll find out what score it gave it and why it didn't pass.

Drasil · 03/30/09, 9:34 PM

Originally Posted by shauno

Obviously hit rating no longer needs to have a weight as every gear combination will be tested and if the hit rating is below 289 it is not suitable

There's nothing wrong with having a gear set under the hit cap. You need a weight for hit rating in this case.

The easiest way I've found to deal with hit rating in my own calculations, is weighting it the same as spellpower (modified by their stat costs). You're going to gem for hit if you're under the cap, or spellpower if you're over. So hit and spellpower are equal, if the following are true:

- You can reach the hit cap by gemming
- If you don't gem for hit at all, you are not over the hit cap.

shauno · 03/31/09, 12:22 AM

Originally Posted by Drasil

There's nothing wrong with having a gear set under the hit cap. You need a weight for hit rating in this case.

The easiest way I've found to deal with hit rating in my own calculations, is weighting it the same as spellpower (modified by their stat costs). You're going to gem for hit if you're under the cap, or spellpower if you're over. So hit and spellpower are equal, if the following are true:

- You can reach the hit cap by gemming
- If you don't gem for hit at all, you are not over the hit cap.

Cheers for the insight, definitely something worth considering.
I found a small bug in the code and have a new better set that beat the old one by 4 points.
Going to add hit rating gems very shortly and that'll almost be the final set until I take the other aforementioned considerations into.. consideration.

chardev.org v6 ~ a World of Warcraft character planner

All that ended up being replaced was the necklace.. good, good.

angaroth · 03/31/09, 12:52 AM

Originally Posted by shauno

Cheers for the insight, definitely something worth considering.
I found a small bug in the code and have a new better set that beat the old one by 4 points.
Going to add hit rating gems very shortly and that'll almost be the final set until I take the other aforementioned considerations into.. consideration.

chardev.org v6 ~ a World of Warcraft character planner

All that ended up being replaced was the necklace.. good, good.

The math must be wrong though, or you are suffering from a local maximum effect. The Unsullied Cuffs as compared to Punctilious Bindings are definitely better.

Your gem math seems wrong. You don't have a hood with a meta. You are putting a purified twilight opal + socket of 5sp ahead of a runed scarlet ruby. I'd check that again with a fine-tooth debugger if I were you.

shauno · 03/31/09, 2:54 AM

Originally Posted by angaroth

The math must be wrong though, or you are suffering from a local maximum effect. The Unsullied Cuffs as compared to Punctilious Bindings are definitely better.

Your gem math seems wrong. You don't have a hood with a meta. You are putting a purified twilight opal + socket of 5sp ahead of a runed scarlet ruby. I'd check that again with a fine-tooth debugger if I were you.

And this is why I posted this here, you are exactly right

My gem math was wrong. I had left out the points recalculation after adding gems, I missed this because it was adding the stats to the Item class but didn't think to check the points were any different. Christ I'm an idiot. I was going to add an event handler to recalculate the points automatically if any stats on an item changed (ie. adding a gem or enchantment) but didn't so that's why it was wrong.

The necklace issue was calculating the +5 SP bonus as +8 SP .. that's what I get for late night coding >_< baah, a few number of bugs in the gem section of code... should be all fixed now, just running another recalculate.. lets see what we get this time.

chardev.org v6 ~ a World of Warcraft character planner

Now THAT looks a lot better huh

Thanks *alot* for your input

I'm running it now for 368 hit rating and soon will add in 2pc/4pc bonuses T7

368: chardev.org v6 ~ a World of Warcraft character planner

I'm currently tweaking it to have a wider item choice and be faster with more options.. see how it goes.

bhaltayr · 03/31/09, 6:00 AM

Originally Posted by shauno

And this is why I posted this here, you are exactly right

My gem math was wrong. I had left out the points recalculation after adding gems, I missed this because it was adding the stats to the Item class but didn't think to check the points were any different. Christ I'm an idiot. I was going to add an event handler to recalculate the points automatically if any stats on an item changed (ie. adding a gem or enchantment) but didn't so that's why it was wrong.

The necklace issue was calculating the +5 SP bonus as +8 SP .. that's what I get for late night coding >_< baah, a few number of bugs in the gem section of code... should be all fixed now, just running another recalculate.. lets see what we get this time.

chardev.org v6 ~ a World of Warcraft character planner

Now THAT looks a lot better huh

Thanks *alot* for your input

I'm running it now for 368 hit rating and soon will add in 2pc/4pc bonuses T7

368: chardev.org v6 ~ a World of Warcraft character planner

I'm currently tweaking it to have a wider item choice and be faster with more options.. see how it goes.

I built an Excel spreadsheet using the Solver addin that is designed at the exact same thing you are trying to achieve. However I'm getting drastically different results than what your application is spitting out. I do however have some suggestions for your app if you intent to release it to the public (and I hope you do).

1. Include the option of setting your own weights to gear. Different specs have different weights for each stat.
2. Include the option of forcing gear in a certain slot, forcing set bonuses etc.
3. Include the option of setting the hit cap.

I have been using the weights (scaling factors) from the 3.1 Simulationcraft thread for my spreadsheet, here are my results for an Affliction Build:

Best in Slot Gear: chardev.org v6 ~ a World of Warcraft character planner
4pcT7: chardev.org v6 ~ a World of Warcraft character planner

In my mind you want the 4pcT7 set to prepare for Ulduar in 3.1 as the new Glyph of Lifetap and the set bonus synergize extremely well. So much so that it's a DPS increase to Lifetap every 20 seconds to keep the buff up as shown by the Simulationcraft Thread results.

Illijilli · 03/31/09, 7:31 AM

Originally Posted by bhaltayr

In my mind you want the 4pcT7 set to prepare for Ulduar in 3.1 as the new Glyph of Lifetap and the set bonus synergize extremely well. So much so that it's a DPS increase to Lifetap every 20 seconds to keep the buff up as shown by the Simulationcraft Thread results.

Correct me if I'm wrong, but the set bonus is only a 10 second buff, the sim shows that tapping every 20 seconds is worth it simply because the glyph is amazing, not only because of the 4pcT7 bonus synergy with it. Even without 4pcT7 I'm pretty sure you want to LT every 20 seconds.

Lasthit · 03/31/09, 7:33 AM

I don't want to be blasphemous here in this thread, but I've been thinking over and over...

0-40-31 is going to beat 53-0-18 by 200 dps, except every half hour, when we have a doomguard, right? This means, either the Ulduar bosses will be encountered every 30+ minutes only (no whipes) or no affli, however good, will beat a similar good hybrid. Not taken trash/dmg-build-up there, etc. into consideration, which never has been affli-friendly...

It's just, that I need some assistance in deciding how to specc... an however much I like affliction, I'd rather not send myself off to mediocre damage-dealer...

So, is there some hope? Or is being affliction a pure decision from the heart?

baka69 · 03/31/09, 8:00 AM

Originally Posted by Lasthit

0-40-31 is going to beat 53-0-18 by 200 dps, except every half hour, when we have a doomguard, right? This means, either the Ulduar bosses will be encountered every 30+ minutes only (no whipes) or no affli, however good, will beat a similar good hybrid. Not taken trash/dmg-build-up there, etc. into consideration, which never has been affli-friendly...
So, is there some hope? Or is being affliction a pure decision from the heart?

I partially agree with you.
OK 40/31 beats, for now, affli spec, but 200 dps is not really big. Besides, the more the fight need movements, the more affli spec will deal damages (and that's the case of most Ulduar fights)

I also think that with end T8 stuff gear, affliction will scale better, because of spell power, but i've got no certitude right now

And BTW, what do we care about trashes ???

Lasthit · 03/31/09, 8:52 AM

Originally Posted by baka69

I partially agree with you.
Besides, the more the fight need movements, the more affli spec will deal damages (and that's the case of most Ulduar fights)

/snip/

And BTW, what do we care about trashes ???

Funnily enough, the complex build-up of affliction (now) and its many synergies seem to destroy the movement-friendliness, as has been dicussed. I personally don't get the impression that affliction is a specc that excels at moving encounters as it used to have, and especially spam/pause-encounters (maexxna) wreck my tries of keeping the dots up.
BUT: that seems to change with 3.1 - at least one thing

However, I don't see significant advantages of affliction in movement encounters with 3.1.

Trash: Well, you know how it is... people lock at the dmg-meter and see overall first... Maybe it's a personal thing, but I'm not that happy to explain over and over again: "But with the bosses, we're REALLY good..." - especially since the classes who hack thorugh trashes do NOT essentiylly suffer at boss damage...

bhaltayr · 03/31/09, 9:57 AM

Originally Posted by Illijilli

Correct me if I'm wrong, but the set bonus is only a 10 second buff, the sim shows that tapping every 20 seconds is worth it simply because the glyph is amazing, not only because of the 4pcT7 bonus synergy with it. Even without 4pcT7 I'm pretty sure you want to LT every 20 seconds.

This is true, even without the set bonus you'd want to tap every 20 seconds. However, my testing indicates that the 4pcT7 bonus rivals almost all best in slot Ulduar gear -- it's that good with the Life Tap glyph. You'll probably want to keep the 4pc bonus until you can accumulate 2pcT8 and other upgrades from Ulduar.