 |
11/12/08, 3:29 AM
|
#196
|
|
Von Kaiser
|
The current accepted 50/0/21 rotation is
PS->IT->HS->HS->OB
PS->IT->HS->HS->HS->HS
This gives you, per minute
Plague Strikes: 6
Icy Touches: 6
Obliterate: 3
Heart Strike: 18
The very nature of the rotation suggests that 2 Heart Strikes > 1 Obliterate.
As such I present this rotation, a 30 second rotation:
OB->PS->IT->HS->HS
HS->HS->DS->HS->HS
PS->IT->HS->HS->HS->HS
Which gives you, per minute
Plague Strikes: 4
Icy Touches: 4
Obliterate: 2
Heart Strike: 20
Death Strike: 2
This trades 2 PS->ITs for 2 Death Strikes, and 1 Obliterate for 2 Heart Strikes.
It does, however, also require epidemic, and the initial OB is undiseased, but only the very first time, each other time through the rotation it is diseased.
Upon looking further, if you waited ~1.5s after the initial oblit before doing the PS, i believe its possible to get through the HS's before the 2nd PSIT application within 15 seconds, requiring only 1/2 epidemic, but I suppose this would require ~perfect latency and 2 is simply much safer.
Anyway, Just food for thought.
|
|
|
|
|
11/12/08, 4:12 AM
|
#197
|
|
Don Flamenco
|
Originally Posted by Oxylos
<snip>
This trades 2 PS->ITs for 2 Death Strikes, and 1 Obliterate for 2 Heart Strikes.
It does, however, also require epidemic, and the initial OB is undiseased, but only the very first time, each other time through the rotation it is diseased.
Upon looking further, if you waited ~1.5s after the initial oblit before doing the PS, i believe its possible to get through the HS's before the 2nd PSIT application within 15 seconds, requiring only 1/2 epidemic, but I suppose this would require ~perfect latency and 2 is simply much safer.
Anyway, Just food for thought.
|
That doesn't look like a lot of a DPS increase to me, trading two PS&IT for Death Strikes doesn't look like a DPS gain, since I presume that the assertion of IT&PS > DS as 50/0/21 is true. How big is the difference between two Heart Strikes and an Obliterate in your model?
At any rate, although this is a subjective matter, it looks like an increase in rotation management, which frankly should be as minimal as possible. The requirement of epidemic also decreases the overall damage since that talent is outperformed by other talents, as has been stated previously in this thread. It would be nice if you could provide the numbers behind this rotation.
|
|
|
|
|
11/12/08, 8:48 AM
|
#198
|
|
Glass Joe
Human Death Knight
Zangarmarsh
|
Items
Since this is a Compendium for DPS, it might be useful to include links for various specs with stat weights on Wowhead. It was posted for Unholy in the 160+ page thread, but it might be useful for this thread to include the unholy one and expand it for the Blood and Frost specs.
Thanks to Methods and Zurm for the stat weights and the majority of the work done for these comparisons.
I've done the Blood (except for AGI and Armor weights, so I just used the weight they had on the Unholy weight): Wowhead - Blood spec stat weight
And here is the one posted on the other thread for Unholy: Wowhead - Unholy spec stat weight
Here is the Frost spec weight, again I had to use the AGI and armor weight value from Unholy as I don't know those weights. Maybe Methods could provide those and I would be willing to update these. Anyways, here is the link for the stat weight: Wowhead - Frost spec stat weight
As always, if anything is wrong, please correct them.
Last edited by Hunglo : 11/12/08 at 8:58 AM.
|
|
|
|
|
11/12/08, 10:36 AM
|
#199
|
|
Glass Joe
|
Concerning DW tri-spec weapon selection:
Are we looking into a Slow / Fast or Slow / Slow weapon selection.
Runeforging selection: Obviously if we end up running 2 slow weapons the order doesnt matter, but if Slow / Fast ends up being better would it be more benificial to run fallen Crusader or Cinderglacier on the fast OH?
|
|
|
|
|
11/12/08, 10:38 AM
|
#200
|
|
Glass Joe
|
Weighted Stats
I would really like to see the weight numbers for Agility for each spec, because it should change as crit changes. However as was pointed out eariler, agility doesn't apply for all crits only melee crits so the question is how do we capture that number?
Armor should be the same because it is a straight conversion from Bladed Armor, 5 / 180 = .02778, techincally it is repeating 7s at the end but you get the point.
The other things that I would love to put a number to is the value of meta gems. We all agree that Chaotic Skyflare Diamond is the best meta gem, but I am currently using Swift Skyflare Diamond as my placeholder because I need a value for the meta gem slot.
Another hole is that I am missing set bonuses for the tier sets. Anyone have numbers for these?
|
|
|
|
|
11/12/08, 11:00 AM
|
#201
|
|
The Ultimate in /facepalm Technology
|
Originally Posted by Infectus
Concerning DW tri-spec weapon selection:
Are we looking into a Slow / Fast or Slow / Slow weapon selection.
Runeforging selection: Obviously if we end up running 2 slow weapons the order doesnt matter, but if Slow / Fast ends up being better would it be more benificial to run fallen Crusader or Cinderglacier on the fast OH?
|
Slow MH, Fast OH. I should probably have Illundai mention that. The reasoning is that you want a fast offhand to generate a maximum number of procs, but a slow, hard hitting main hand for more powerful strikes and BCB procs.
|
Back, semi-casual, and proud of it.
|
|
|
11/12/08, 11:32 AM
|
#202
|
|
Glass Joe
|
Originally Posted by Hunglo
|
Are these weightings for the 50/0/21 spec, the 51/13/7 spec, or are they valid for both?
|
|
|
|
|
11/12/08, 12:54 PM
|
#203
|
|
Don Flamenco
|
Originally Posted by Infectus
Concerning DW tri-spec weapon selection:
Are we looking into a Slow / Fast or Slow / Slow weapon selection.
Runeforging selection: Obviously if we end up running 2 slow weapons the order doesnt matter, but if Slow / Fast ends up being better would it be more benificial to run fallen Crusader or Cinderglacier on the fast OH?
|
I seem to recall that BCB doesn't proc the mainhand from offhand hits anymore - confirm/deny?
If that's the case, fast-fast is definitely better. Of it still does, Slow/Fast would be better, ala BC combat-sword rogues.
edit:
|
Originally Posted by zurm
Slow MH, Fast OH. I should probably have Illundai mention that. The reasoning is that you want a fast offhand to generate a maximum number of procs, but a slow, hard hitting main hand for more powerful strikes and BCB procs.
|
Oh, our strikes aren't normalized? I had been spreadsheeting with the assumption they were.
|
|
|
|
|
11/12/08, 1:03 PM
|
#204
|
|
Von Kaiser
Human Death Knight
Bloodhoof
|
Originally Posted by Janraea
Oh, our strikes aren't normalized? I had been spreadsheeting with the assumption they were.
|
I'd like confirmation on this as well.
|
|
|
|
|
11/12/08, 1:05 PM
|
#205
|
|
The Ultimate in /facepalm Technology
|
Originally Posted by Janraea
Oh, our strikes aren't normalized? I had been spreadsheeting with the assumption they were.
|
BCB isn't (and possibly rune strike/frost strike?), although it might be true that the increased procrate of fast/fast beats out increased BCB damage. In terms of strikes being normalized, I believe they are. I mis-spoke when I said increased strike damage.
Originally Posted by Janraea
I seem to recall that BCB doesn't proc the mainhand from offhand hits anymore - confirm/deny?.
|
Deny, as of last beta build if memory serves.
|
Back, semi-casual, and proud of it.
|
|
|
11/12/08, 1:08 PM
|
#206
|
|
Piston Honda
Human Death Knight
Khadgar
|
Why do people often not understand that attacks are normalized so that they benefit from attack power equally (i.e. treated as though they're all the same speed for AP calculation purposes) but having a slower weapon (all else being equal, specifically DPS) means you have a higher damage per hit (from the higher base damage) and thus you want slow weapons for instant attacks. So even for normalized instant attacks it makes a lot of sense to always the slowest weapon.
|
|
|
|
|
11/12/08, 1:26 PM
|
#207
|
|
Don Flamenco
|
Originally Posted by Lanlaorn
Why do people often not understand that attacks are normalized so that they benefit from attack power equally (i.e. treated as though they're all the same speed for AP calculation purposes) but having a slower weapon (all else being equal, specifically DPS) means you have a higher damage per hit (from the higher base damage) and thus you want slow weapons for instant attacks. So even for normalized instant attacks it makes a lot of sense to always the slowest weapon.
|
If it were purely a question of strike damage, that would be true. But one of the major damage sources for a DW build is the Killing Machine proc, which goes off significantly more often with fast-fast than with slow-fast. As it stands, I'm confident that slow-fast is more damage for a BCB build and fast-fast is better for a build without it (if a DW build can be worthwhile without BCB).
|
|
|
|
|
11/12/08, 3:43 PM
|
#208
|
|
Von Kaiser
Dwarf Death Knight
Alleria
|
Originally Posted by Zurm
I mis-spoke when I said increased strike damage.
|
Actually, you didn't. Slow weapons *will* have increased strike damage, but not simply because of weapon speed. It's because slower weapons have higher average damage.
I know I'm picking at nits, but just thought I'd clarify....
|
|
|
|
|
11/12/08, 3:58 PM
|
#209
|
|
Don Flamenco
|
Originally Posted by Crax
Actually, you didn't. Slow weapons *will* have increased strike damage, but not simply because of weapon speed. It's because slower weapons have higher average damage.
I know I'm picking at nits, but just thought I'd clarify....
|
Put more clearly, only the portion of damage added by AP gets normalized - the damage of the weapon itself is not. People always use words like 'average damage', 'base damage', 'top-end damage' - that confused me for a long time (I would say that 'average damage' is the average of your min and max weapon damage after AP).
|
|
|
|
|
11/12/08, 6:09 PM
|
#210
|
|
Glass Joe
|
Excuse my lack of knowledge, I do not understand why crit rating has such a low value for death knights?
I understand that crit is a powerhouse for us compared to enhancement shamans for example, but for 1 crit rating to be almost equal to 1 AP is dramatically low.
|
|
|
|
|
|