Thats not entirely true. You might be right about base damage of the spell calculated by spellpower, but modifiers like Soul Siphon etc. are influenceing the damage of every single tick. Though this does not relate to "Trinket Discussion", I think its important to mention this at that point.
If you're referring to the "number of Affliction effects on target" check that Soul Siphon does, it has not been confirmed here on EJ that Soul Siphon rechecks for the number each time Drain Soul ticks.
If hit isnt an issue. Would [Eye of the Broodmother] with [Illustration of the Dragon Soul] be a good choice? 325 extra spellpower up for the whole duration of a boss encounter + 87 crit. vs chance on procs every 45 secs or so. At least till a better trinket drops?
Thanks for the quick reply. Really, I would have never imagined Eye of the Broodmother being that low in terms of dps (on par with embrace and the sundial [THAT I CANT GET RID OF GRR]) I currently have Dying Curse. With my gear setup, though, I'm going to be over the hit cap with 3/3 supression if I have Dying Curse.
I guess I could drop a point or two in that but, eh..
I think the problem I have with the DPS calculations on items vs eachother is that usually hit items have more dps, but that's because it's assuming you aren't hit capped from other items.
One large mistake you're making is the assumption that procs happen instantly the moment they're off cooldown, with no delay at all. You should be accounting for the proc chance and casts per second to establish an average "time to next proc"- as opposed to merely taking the cooldown.
The mind is like a parachute. It works best when it's open.
One large mistake you're making is the assumption that procs happen instantly the moment they're off cooldown, with no delay at all. You should be accounting for the proc chance and casts per second to establish an average "time to next proc"- as opposed to merely taking the cooldown.
I'd agree here as well. With the numbers being so close, I'll take the one I KNOW will be up at all times. I can't tell you how often I got a sundial / spider proc right after a heroism goes away.
Also of note procs happen at crap times, like right after you refreshed your dots, almost never when you are reapplying doom, clipping the end of the LT buff so you have to sacrifice that ~200 spell power to utilize the 500+ on the trinket proc, and that is to say nothing of right as kologarn starts chasing you with his laser, grips you, ignis puts you in his furnance, razorscale drops fire on you, Auiraya fears you, Hodir does deep freeze, and so on and so on and so on.
In game broodmother kicks the crap out of proc trinkets.
Also of note procs happen at crap times, like right after you refreshed your dots, almost never when you are reapplying doom, clipping the end of the LT buff so you have to sacrifice that ~200 spell power to utilize the 500+ on the trinket proc, and that is to say nothing of right as kologarn starts chasing you with his laser, grips you, ignis puts you in his furnance, razorscale drops fire on you, Auiraya fears you, Hodir does deep freeze, and so on and so on and so on.
In game broodmother kicks the crap out of proc trinkets.
You also lose the bonus of the passive stats during movement every time. Movement fights end up being slightly in favor of procs (in the long run), if you consider that you are more likely to proc it during stand-still portions.
One large mistake you're making is the assumption that procs happen instantly the moment they're off cooldown, with no delay at all. You should be accounting for the proc chance and casts per second to establish an average "time to next proc"- as opposed to merely taking the cooldown.
I agree, but I am looking for some help to devise proper formulas for trinket dps. I have been using the formula found in the raiding compendium for proc trinkets only adding a multiplier for spec specific dps modifiers and thats all well and fine.
(505*1.55)*20/120 for the proc of 505 sp for 20 seconds 2 minute cd. This tells me that the proc from living flame yields 84 dps. Not very impressive, but when you consider the trinket gives a massive 4.07 to hit, its not so bad.
Am I calculating the dps for on use correctly here?
Yes you are, however while this is the correct method of calculating this- a note should always be added to on use trinkets. That is if you- for example, synchronise a 2 minute trinket with meta morphosis, and use meta on cooldown-- you'll probably see a DPS increase. /use trinkets give you control, and while you can't put a good number on that level of control, it's a definite benefit that should be noted.
The mind is like a parachute. It works best when it's open.
Flare of heavens beats illustration of the dragon soul, so if you wanted to waste a trinket slot on hit then you should go for [hit trinket] + Flare of the heavens. Other than that it seems correct
The mind is like a parachute. It works best when it's open.
Flare of heavens beats illustration of the dragon soul, so if you wanted to waste a trinket slot on hit then you should go for [hit trinket] + Flare of the heavens. Other than that it seems correct
Rather than asking in the other thread, I cut and pasted some info you posted regarding DPS values for trinkets for deep destro.
Item SP Haste Hit Crit DPS
Flare of the Heavens 142 0 0 120 338.9
Illustration of the Dragon Soul 200 0 0 0 310.0
Scale of Fates 125 72 0 0 288.8
Eye of the Broodmother 125 0 0 87 279.9
Dying Curse 128 0 71 0 269,4
Embrace of the Spider 98 84 0 0 262.8
Elemental Focus Stone 87 0 108 0 242.9
Living Flame 84 0 107 0 237.2
Where does the Sundial fit into this hierarchy? I was told it is close to Eye of the Broodmother and Embrace of the Spider (somewhere in between them supposedly.)
Where does the Sundial fit into this hierarchy? I was told it is close to Eye of the Broodmother and Embrace of the Spider (somewhere in between them supposedly.)
Item
SP
Haste
Hit
Crit
DPS
Flare of the Heavens
142
0
0
120
338.9
Illustration of the Dragon Soul
200
0
0
0
310.0
Scale of Fates
125
72
0
0
288.8
Eye of the Broodmother
125
0
0
87
279.9
Dying Curse
128
0
71
0
269,4
Embrace of the Spider
98
84
0
0
262.8
Elemental Focus Stone
87
0
108
0
242.9
Living Flame
84
0
107
0
237.2
Sundial the Exiled
98
0
0
84
235.6
This is assuming:
* Scale factors SP=1.55, Haste=1.23, Hit=1, Crit=0.99
* Trinkets with a 45s internal cooldown and 10% proc chance proc once per minute.
* Trinkets with a use effect are used on cooldown
The reason hit is valued at "1" dps, is because if you value hit too high, you'll get way too much of it and it loses it's value. If you don't value hit you'll get too little hit rating and hurt your DPS aswell. I did some gear mixing and matching and found that if you valued it at roughly 1 DPS, you'd get to roughly the amount of hit rating you need.
Note though, since I made this list the scale factors have been updated. The value of haste has increased a lot since the latest change.
The mind is like a parachute. It works best when it's open.
Nice recap warlo. Here is my 2 cent to improve it.
afair, dying curse proc is 15% not 10% so SP should be 139 and 286 dps
Moreover Elemental Focus Stone provide 520 haste not spell power so it should be 87 haste which give 215 dps if you rate hit to 1
Right, lets re-do it then with the new scale factors and see what the table looks like. Also lets use proper proc chances, still assuming the 45s internal cooldown- and assuming a cast frequency of 1.5s (You probably need roughly 600 or so haste to have this cast frequency).
Proc chances taken from wowhead spell info.
Formule: Castfrequency / Proc chance = Time to proc.
15% proc chance with a 1.5s cast frequency means 10 seconds to proc after cooldown is up.
10% proc chance with a 1.5s cast frequency means 15 seconds to proc after cooldown is up.
Benefit from proc = Proc / (Internal CD + Time to proc) * Proc Duration
That change basically puts Dying Curse above Eye of the Broodmother. Interestingly enough none of the other trinkets have anything but a 10% proc chance.
The mind is like a parachute. It works best when it's open.
It seems you multiply the spell power/haste/crit columns with their proper values, but use 1.00 for the hit column? All +hit trinkets should have a much higher average DPS in your table...
It seems you multiply the spell power/haste/crit columns with their proper values, but use 1.00 for the hit column? All +hit trinkets should have a much higher average DPS in your table...
Read post 368 for the reason.
The mind is like a parachute. It works best when it's open.
That's just silly. What you do is give the correct +hit values for all items. Then you juggle your trinket with whatever other hit gear you have to find the ultimate combination, *assuming* correctly calculated gear DPS values on all items.
If you've got some combination, for example 4 items in trinket slot, ring slot, wand slot and Legs where you can juggle around hit gear, write down the combinations you can have with your specific set of items to find the highest DPS from that table.
How is it silly to devalue hit? The itemization will cause all hit trinkets to be exceedingly inflated in value, giving an inaccurate comparison of hit vs. non-hit trinkets. Hit is itemized in a similar fashion to haste or crit, which have roughly a dps value of 1, probably slightly higher. Every spec has hit as its highest DPS value and every spec has reaching the hit-cap as an understood necessity. The actual value of hit doesn't hold an accurate value when assessing gear. Giving it an arbitrary value of '1' prevents hit gear from having an either inflated value or no value, and gives it an accurate assessment.
Your method may be better if you are compiling a complete BiS list, but for the evaluation of a single-trinket, it is better to devalue hit.
How is it silly to devalue hit? The itemization will cause all hit trinkets to be exceedingly inflated in value, giving an inaccurate comparison of hit vs. non-hit trinkets. Hit is itemized in a similar fashion to haste or crit, which have roughly a dps value of 1, probably slightly higher. Every spec has hit as its highest DPS value and every spec has reaching the hit-cap as an understood necessity. The actual value of hit doesn't hold an accurate value when assessing gear. Giving it an arbitrary value of '1' prevents hit gear from having an either inflated value or no value, and gives it an accurate assessment.
Your method may be better if you are compiling a complete BiS list, but for the evaluation of a single-trinket, it is better to devalue hit.
I don't agree. A trinket with 71 or more +hit allows me to replace other +hit gear exactly as much as the 71 hit is worth. Take an example:
The hit rating of the first combination is 68, for the second combination 71. Let's call that a draw for the sake of argument.
According to the lootrank using this thread first post numbers for Destro spec warlock, the first Shoulders give more DPS than the T8 by itself (ignore set bonus for this argument). In your table, Dying Curse is worthabout the same as Eye of the Broodmother, meaning the first combination is a better combination since the shoulders are clearly better.
However, just using the first post scaling numbers on lootrank.com gives:
435 + 281 = 716
399 + 429 = 828
The T8 plus Dying Curse wins hands down, despite the combinations having very similar +hit total. The +hit rating of Dying Curse allows you to replace the same amount of +hit on other gear, thereby increasing DPS. There is no magic "it'll take you over your hit cap", cause then you're not using your items optimally. The only accurate way of calculating gear is to give +hit the correct scaling factor compared to the other stats, so you know how much its worth to replace +hit with other stats.
This is just a simple example, but it always applies the same way. If an item has a lot of +hit, you are free to use non-hit gear in other slots, often combinations of several slots due to these trinkets' high +hit rating.