Unholy Dps | 3.3.5, King of Anything - Page 67

Stealthah · 10/22/09, 11:36 PM

Im just wondering if there is a trinket analysis anywhere? Or what the forumla is for trinket analysis. Is it (time up / cool down) X ep values? Or how does that work?

Clearly dual death verdicts are BIS, or would greatness be greater than the lower DV? Was just wondering if there was a list or an equation for trinkets! Thanks

At first eval for DW UH I would think it would go: H-DV , Greatness , DV, Comets Trail, Wrathstone?

Larisroth · 10/23/09, 12:02 AM

Interesting. The relative damage break down of the skills is probably well worth looking at as well.

Part of the high value of hit is probably due to large dps loss if you miss a pestilence. With a 21 second duration on diseases, and a 10 second rune cycle, you're refreshing within 1 gcd of the diseases expiring, So I'm pretty sure you have to reapply diseases if you miss. At the melee hit cap you'll only miss on about 1% of spells, below that, it'll get steadily worse. I'm not sure what delta the sim uses for hit rating but it's quite likely that it could result in a 3-5% chance of missing a pestilence. Depending on how it actually handles it, missing a death rune scourge strike could also have the same effect as it can push the refresh timer out. Without double Death's choice procs the disease damage could be quite a lot less. Without the low chance 2T9 bonus, you're probably going to max diseases in the sim within 20 seconds, but a there's not a lot of chance you'll actually manage to do that later on.

In practice, hit rating is pretty easy to come by, so that's not likely to be a major issue, although running with the spell hit debuff could be annoying (One nice advantage of dual-wield, is that hit is still useful past the cap). From experience, the movement requirements and target switching of modern fights can however make it quite hard to manage refreshing the diseases though. At least the new bone shield will free up a blood tap.

Because haste rating affects proc chances, there's a fair bit of randomness in the EP values for it.

ArPen looks like the third best stat, and is reasonably easy to come by, its EP should also increase as you stack it. I'd like to see what its value at a pretty easy to achive static value would be.

Kaejin · 10/23/09, 12:03 AM

@Stealthah: What makes you think Greatness would be better than DV? Even though DMC:G's static 90 str is slightly more than DV's static AP, the proc has it beat by 150 str, and they have the same proc rate and ICD.

Consider · 10/23/09, 12:03 AM

None is really needed. So long as you know the stat weights, it's no harder to figure out the value of different trinkets then it is to figure out the value of different pieces of gear. Time up/cooldown + 5 seconds is how I would do it, as rarely would they proc immediately.

Or you can always use the sim. Kahorie's sim can/will calculate the value of trinket procs if you want it to. Then simply add it to the value of the passive aspect of the trinket, and you're golden.

Heroic DV > DV > Greatness (after which you have the two Algalon trinkets - Comet's Trail and Dark Matter).

In other news:

Originally Posted by Ghostcrawler

the shadow portion benefit from vicious strikes and subversion.

It should.

15% more crit to the shadow portion is another moderate buff (assuming, of course, that Subversion + VS is the cause behind its lower crit rate. It could potentially be the simple fact that it's running off spell crit, but I find that doubtful - that wouldn't explain the size of gap, and testing without Subversion and VS seem to make the crit rates of the two much more comparable, if not identical). It also makes GoD that much stronger (as every buff to SS does).

Edit:

Originally Posted by Larisroth

Without double Death's choice procs the disease damage could be quite a lot less.

Although true, it wouldn't significantly change things. DC doesn't give that much more AP than Mirror of Truth, a trinket anyone should have access to. Banner of Victory, Wrathstone, Pyrite Infuser, so on... all trinket procs give very similar AP values. What makes DC so great is the fact that it's strength, not ap.... but that's irrelevant when considering GoD and nothing else.

The missing Pest thing is something I didn't quite think of. Hit more than doubling in value still seems a bit excessive, but I suppose it doesn't really matter - if it's higher than strength, which it will be regardless, we'll want to cap it, whether it's worth 3.1 or 4.0 or 7.0 or 1000.0. Then, once capped, it doesn't matter.

I could never get the value of ArP to top the value of crit (unreachable amounts aside, of course).

Larisroth · 10/23/09, 12:50 AM

Originally Posted by Consider

In other news:

15% more crit to the shadow portion is another moderate buff (assuming, of course, that Subversion + VS is the cause behind its lower crit rate. It could potentially be the simple fact that it's running off spell crit, but I find that doubtful - that wouldn't explain the size of gap, and testing without Subversion and VS seem to make the crit rates of the two much more comparable, if not identical). It also makes GoD that much stronger (as every buff to SS does).

Edit:

Although true, it wouldn't significantly change things. DC doesn't give that much more AP than Mirror of Truth, a trinket anyone should have access to. Banner of Victory, Wrathstone, Pyrite Infuser, so on... all trinket procs give very similar AP values. What makes DC so great is the fact that it's strength, not ap.... but that's irrelevant when considering GoD and nothing else.

I could never get the value of ArP to top the value of crit (unreachable amounts aside, of course).

Actually all that was really relevant was that it was double proc trinkets and while they were pretty much guaranteed to be synced and up at the start of the fight, there's actually only a 30% or chance or so for each trinket to be up at a random time if you're forced to refresh diseases. The sim tries to maximise the AP of the diseases, so if both drop off it might end up reapplying the diseases 4-5 times before it actually gets optimal ones, You'd have to check the logs or damage output from running a low hit sim to confirm.

With SS double dipping from talents and crit percentage to the large extent it does crit rating (and agility) will super scale like ArPen. I don't think it'll ever get to the point where they'd be more valuable than strength, but it might be possible that say an updated Mjolnir Runestone could compete with the DC trinkets.

The big downside to stacking ArPen and over stacking crit is that SS is a single target spell. What happens in true AE situations? (reaping has no benefit in these cases either)

Consider · 10/23/09, 1:22 AM

Over the course of a 100 H sim, it only did PS + IT 4 times each. It's gotten much more intelligent about that sort of thing.

But, the thing is, even without the increase in disease damage via rolling, GoD is superior to the alternatives, simply because 1 SS > 1 PS + 1 IT + 1 BS (especially in any fight where you can make use of the additional GCD, or any fight with multiple mobs, of course). Increased disease damage is icing on the cake. So even if your trinkets don't sync, even if your trinket procs aren't that great, any of that - you still come out ahead. It's just a matter of how far ahead.

Or such is what my math and simming has shown me.

As to crit/arp, it will be interesting. A lot of it depends on ICC itemization, which is still a big question mark.

ShinKosh · 10/23/09, 4:40 AM

Now that the mechanics of the ability have settled down and the numbers are flowing, has anyone tested the new Scourge Strike on a resilience heavy target? Both light and heavy on the armour range?

It would be interesting to see how these 15k Scourge Strikes from PvE transition over to PvP.

Vyria · 10/23/09, 4:48 AM

This is what I'm getting for my sim

AttackPower	1
Strength	2.89
Agility	1.73
CritRating	2.3
HasteRating	1.08
ArmorPenetrationRating	2.03
ExpertiseRating	2.24
BeforeMeleeHitCap<8%	5.86
SpellHitRating	7.12
WeaponDPS	7.57
WeaponSpeed	486.49

The spell hit seemed pretty ridiculous till I realized it was probably due to Pest.

It's kind of bothering me that there are several dozen applications of IT and PS in my sim but I assume that rotation timing is doing something weird so they're rarely falling off.

Consider · 10/23/09, 4:58 AM

Spell hit rating can't be higher than pre melee hit cap rating. It just isn't logically possible. Anything which increases the value of spell hit rating (such as pest) will also increase the value of pre melee hit cap rating, as the latter does what spell hit rating does (and more).

That aside, your numbers are all within 10% of mine (hit aside, but that's already mentioned, and haste aside, but I know mine can't be right - although yours still seems somewhat low as well. Most likely something the sim is doing/calculating incorrectly).

@ ShinKosh, This really isn't the best place to discuss PvP specific info, as that does have its appropriate forum and subsequent thread: [Deathknight] PvP Thread .

That said, it's simple enough to take your average SS based in your PvP gear, multiply it by whatever you consider the average resilience value, and then multiply by different armor values. It doesn't double dip into resilience (and obviously not into armor), so there's nothing complex about the math.

ShinKosh · 10/23/09, 5:22 AM

I understand what you’re saying Consider and I wasn't trying to derail the thread into some god-awful PvP fueled E-rage like a lot of PvP topic end up like, merely hoping for a few numbers for the newer Strike and its mechanics. A few days ago there was 2 pages of 'concern' about the 3.3 SS when faced with a PvP geared target and in light of faulty game mechanics at the time the topic was dropped.

I'll post up on the PvP thread and see if anyone over there has some PTR numbers, it's just they are still talking about heathens ways like Frost and Blood.

If by any chance anyone happens to accidently swing a two hander into their PvP geared mate, it would be interesting to see how Unholy will fair for 3.3 PvP.

Vyria · 10/23/09, 5:27 AM

Originally Posted by Consider

Spell hit rating can't be higher than pre melee hit cap rating. It just isn't logically possible. Anything which increases the value of spell hit rating (such as pest) will also increase the value of pre melee hit cap rating, as the latter does what spell hit rating does (and more).

I thought it was wrong at first, then just assumed it was calculating the value of melee hit and spell hit independently of each other. This is my first time getting deep in the nuts and bolts of this stuff so I've got no clue here.

It makes sense to me that haste drops in value because you're not casting IT every rotation and the use of fewer single rune abilities gives you more wiggle room.

eventually · 10/23/09, 5:48 AM

The sim numbers for scourge strike are definitely incorrect, my average scourge strikes on festergut were no where near 15k.

Hyber · 10/23/09, 5:59 AM

Strength weights: Proc-able vs static

I'm curious what the weights are for proc-able strength buffs that are a bit less common than Fallen Crusader and Sigil of Virulence. Thinks like Death's Choice and Greatness.

The weight for strength from these sources should be less valuable than static strength, because the procs aren't available for pre-casts like Ghoul and AotD, and may not be up for Gargoyle. My question is "how much less valuable?" Yes, it's largely academic, since there aren't many alternatives in those slots end-game and it's probably easy enough to swap and check, but I'm curious about the math.

I imagine the weight is actually a calculation of several factors, including proc rate, up-time, and rotation, and then applying that to whatever formula we base Unholy spec weights on. But the math there is a little beyond me.

seraphthrone · 10/23/09, 11:14 AM

Some problems on the new sigil not fully stacked when applying diseases can be simply remedied by using the old sigil first to apply diseases then swap in new sigil for other damage.

As long as the diseases don't fall off, they will maintain the strength even with the sigil not equipped. More interestingly, once the new sigil is fully stacked as well as all procs are up, one can simply reapply the diseases to receive an even higher damage on them.

And all these costs 1 GCD.

Kithus · 10/23/09, 12:11 PM

Originally Posted by eventually

The sim numbers for scourge strike are definitely incorrect, my average scourge strikes on festergut were no where near 15k.

How do your stats compare to the stats Consider and I are using? The stats we plugged in relate to a BiS set of gear from ToC Hardmodes, including iLevel 258 weapons and tribute chest cloaks. Its quite possible the sim doesn't have the new SS quite right. However, they are not "definately incorrect" because one person who may have wildly different stats from those we are using does not see his numbers match up. Please provide more the anecdotal evidence to back up your point.