For Demo locks... Imp Swarm Glyph

[Zargul]

I'm pretty sure your example is the absurd one, as you're equation doesn't even track with basic multiplication, much less algebra.

Would you mind explaining to me how it is supposed to work then, rather then spouting off a number that is impossible to get to and proving it with math that doesn't make sense?

It is my understanding that with current gear it is impossible to get anywhere near 100% haste with Haste rating, and considering Tempus Fugit and Bloodlust are both Multiplicative, they wouldn't get you to 100% either. I've gotten my Imp Swarm Cooldown as low as 40 seconds during Bloodlust, and around 1 minute elsewise and I don't play a troll, so please enlighten me, what math do you go by?

It's common practice in video games to reduce the time necessary to do something by dividing the base by the bonus + 100% instead of multiplying the base by 100% - the bonus. This creates an asymptote or value which limits how low or high you can go.

My math seemed pretty simple to me:

modified cooldown = base cooldown / (100% + haste%)

also can be stated as:

modified cooldown = base cooldown * (100% / (100% + haste%)) - I missed some parentheses here, maybe that confused people

This means that getting 100% haste doubles how often you can use an ability, which is intuitive to me.

Getting 50% haste means you can use an ability 50% more often, but in actually you only reduce the cooldown by 33.3%. This is because period is an inverse of frequency. Using something 3/2 times as often means the period between uses is 2/3 that of normal. For example: Say Dark Soul has a 3 minute cooldown. You can use Dark Soul 2 times in 6 minutes. Now reduce Dark Soul's cooldown by 33.3%, so it's on a 2 minute cooldown (3 * 66.7%). You can use Dark Soul 3 times in 6 minutes now. 3 Dark Souls / 2 Dark Souls = 1.5, or a gain of 0.5 (50%) more Dark Souls.

As far as Imp Swarm cooldown goes, I've tested the values myself and it checks out to follow my above formula. Here's an easy way to check:

1. Look at your haste value on character sheet
2. (using a calculator) divide 2 (base cd of Imp Swarm) by 1 + your haste as a decimal (.18 for 18% haste)
3. Compare the value with the cooldown of your Imp Swarm, they should be the same

Edit: examples:

Haste%                Imp Swarm CD
--------------------------------------------------------
0                         2.0   min
10                       ~1.81  min       (NOT 1.8 but roughly the same)
25                        1.6   min
50                       ~1.333 min
100                       1.0   min
100000                   ~0.02  min

At low haste values, it's easy enough to confuse the two methods of computing cooldown because they give similar numbers.
However, at higher haste values, it becomes clear as to why you should compute cooldown by dividing by 1 + haste, instead of multiplying by 1 - haste.

[Orloth]

I see, and I see where you were unclear.

Your example did not specify which values were variables and which were constants, and failed to take into account that the '%' sign means 'parts out of 100' or if you prefer 'divide by 100' which made the whole thing unclear. This makes 100% = 1.00 .

The actual Haste equation should then be: Modified Cooldown = Base Cooldown / (haste% + 1), the goal then should be to get enough haste to get the bottom to be equal to '2' in order to get half the Cooldown. That happens at 100% haste.

Thanks for the lesson.

[Zargul]

Yeah sorry, 100% means "1" to me, but as haste is expressed in a %, I chose to use a lot of "100%" instead of "1".

[rijn dael]

Imp Swarm CD in seconds = (120 * (1/ (1 + (UnitSpellHaste("player")/100)) ) )

Done

[Bloodfire]

Less imps.
They're likely to nerf TO THA GROUND UVLS in some matter.
I wonder what's gonna be next.

[JV Chequer]

When I reached almost 300% haste with affliction, spells should activate like 3times if haste works that way :P

Less imps.
They're likely to nerf TO THA GROUND UVLS in some matter.
I wonder what's gonna be next.

Not a huge nerf, probably Imp Swarm still be a dps increase, but not mandatory, just a choice between burst and stable dps.

[Bloodfire]

When I reached almost 300% haste with affliction, spells should activate like 3times if haste works that way :P

True.
But GCD won't allow you to do so

[JV Chequer]

True.
But GCD won't allow you to do so

The GCD would even notice me passing by.

[Orloth]

Less imps.
They're likely to nerf TO THA GROUND UVLS in some matter.
I wonder what's gonna be next.

Actually, it looks like they are nerfing ... *ahem* 'redesigning' the Glyph of Everlasting Affliction, so we can't get 2:25 of 100% crit dooms from a single UVLS proc. I haven't seen anything new on that, past the news that it was on the chopping block, but I wouldn't be surprised if it's in the next couple of builds.

[Bloodfire]

Actually, it looks like they are nerfing ... *ahem* 'redesigning' the Glyph of Everlasting Affliction, so we can't get 2:25 of 100% crit dooms from a single UVLS proc. I haven't seen anything new on that, past the news that it was on the chopping block, but I wouldn't be surprised if it's in the next couple of builds.

Well, that's gonna affect our dps hence UVLS is superb for demolocks mostly (Spriests and Owls benefit from it lower).
I wonder whether we'll replace it by T16 Trinkets or leave it till the end of MoP.

[Biomega]

The last time people complained about a previous tier trinket being BiS, the blue response was "is that necessarily a bad thing?". I wouldn't mind if UVLS remained BiS throughout the next tier, in much the same way Shard of Woe remained BiS throughout FL (and even some of DS).

[Zinnin]

This isn't really a surprising change, although it might mean you have to be slightly smarter to get a gain out of the glyph. Right now it is,almost impossible to use it so poorly that it is a dps loss.

[Werst]

If its intended to nerf demo's imps its an odd way to go about it. If anything it extends the shelf life of UVLS since spec based imp generation just took a hit. I would think they are looking to do the opposite but removing (theoretically, its still there as of latest build) the EA glyph already does that. As Zinn said, the glyph right now is hard to use badly even if you only ever think about it during darksoul. I'd almost rather see demo get the 5 imp glyph on a 2 min CD baseline and let the glyph'd version simply have its CD reduce by haste. It might feel less mandatory then if that is what they are actually aiming for.

The glyph being mandatory is more something I could see blizz picking at. Right now demo is all about the 2 min money shot with darksoul, engineering, lifeblood, orc racial (if applicable, if not likely troll every 3rd DS), and whatever else we can hope to pile on that....hopefully a RPPM proc if we are lucky. Toning the glyph down a bit would enable a slightly more steady playstyle which could suit a more casual player. By steady I mean fixed professions, wrathguard vs svc felguard, non on use racials, etc. You could try sync'ing the imp swarm glyph with the 11K int proc from the amp trinket but I doubt it would compare to darksoul. You could get both for the first and maybe part of the second but beyond that you'll need AV's 2nd charge to line them up.

In the few mins the PTR was up since the new patch I didn't get to test it extensively. Hopefully they also cleaned up the bugs in terms of demo's T16 set bonuses.

[Deleted]

Hmm, I see, I seem to be a little behind in some of the latest changes then Still, not happy lol. Anyone got a link to the latest version of the amp trinket?

EDIT: Oh, I see, yeah this is kinda interesting then. Prismatic Prison of Pride

That trinket you linked says it procs off of heals, the dps one is from immerseus, though I see you already corrected yourself in the other thread. Thought I'd clear it up for anyone just reading this thread though.

[Deleted]

If UVLS was to be left unchanged, it will be BiS for T16, and T17, and T18, and T19, need I go on? It's too strong, If it isn't to be nerfed, we'd be using it at lvl 95 and lvl 100 unless they change Demo's imp mechanic entirely.

[Deleted]

If UVLS was to be left unchanged, it will be BiS for T16, and T17, and T18, and T19, need I go on? It's too strong, If it isn't to be nerfed, we'd be using it at lvl 95 and lvl 100 unless they change Demo's imp mechanic entirely.

UVLS has never been too strong, what UVLS did was expose demo's short comings, as it needs a very steady supply of wild imps to be competative, something that demonic calling couldnt supply alone, i will agree that UVLS increased it to a point where it prolly shouldnt be and changing that by nerfing UVLS wont really help demo at all, where a change to its imp mechanic will, im certainly envisioning doom being changed to something along the line of a say 50% chance on tick to summon wild imp, it would completely remove the necesity of UVL, ofc then blizz just need to make the t16 trinkets worth getting, which they arent atm.

[varren]

Doesn't the RPPM change kind of kill off UVLS anyways? From what I understand it will make it so UVLS will proc anywhere from 1 sec into the fight to 90 sec into the fight. Sounds way too RNG to me for an already pretty RNG playstyle on live. Not to mention EA glyph removal gets rid of the RNG protection.

[Deleted]

Doesn't the RPPM change kind of kill off UVLS anyways? From what I understand it will make it so UVLS will proc anywhere from 1 sec into the fight to 90 sec into the fight. Sounds way too RNG to me for an already pretty RNG playstyle on live. Not to mention EA glyph removal gets rid of the RNG protection.

that depends entirely on which version you are using and even with the EA glyph you werent 100% certain to get a proc while you had your crit doom up anyways but yeah, with the recent changes to UVLS it makes it even more difficult to use for demo, but it will still be a viable trinket.

[Werst]

With the removal of haste scaling RPPM I don't think even the HTF 2/2 UVLS will even average a proc to keep up crit doom all the time. As we know the odds of having "average" procs is not going to happen. Even trying to use UVLS that cuts our imp generation very significantly. Thus it makes little sense to hit the glyph when it wasn't the source of demo's imps compared to doom ticks.

I think a far better solution is to have the live imp glyph (5x on 2 min CD) be baseline but let the glyph reduce that based on haste. That would make it far more optional without beating down demo's playstyle (esp non UVLS playstyle) so much.

[Biomega]

Haste scaling for RPPM added a statistically almost insignificant amount of uptime, due to the small sample sizes of real life raiding.

But isn't UVLS around 0.5x RPPM even after the patch? Doesn't that mean that on average, it'll proc around once a minute? Even with the loss of EA, shouldn't that still give a decent chance at having 100% crit Dooms ticking? My math may be entirely off there, I'm not very good at these things...

Also, from what I understand the on-pull change was done to reduce randomness, wasn't it? I know of few guilds that actually waited out the time every pull to get the "guaranteed" procs, so really it just benefits faster pulls, doesn't it?