The Retribution Concordance - 4.3, Voice of the DPS - Page 8

Naididae · 12/21/10, 7:36 PM

Originally Posted by chippydip

Nice catch! Its actually quite a bit better by either set of stats weights. 50 AP / 2 / 1.05 / 1.05 = ~22.7 Str. 65 crit * 0.2 = 13 effective Str. So, the AP looks to be at least 75% better than the crit (which requires Maelstrom Crystals). This also shows that the Leatherworking bonus is ~27.3 Str better than other professions.

Similarly, 32 AP / 2 / 1.05 / 1.05 = ~14.5 Str to boots looks better than 50 rating * 0.2 = 10 effective Str.

This makes me wonder how Berserking stacks up with Avalanche/Hurricane until Maelstrom Crystals are more readily available for Landslide. (Currently I've just been using Avalanche, but I'd like to know if Hurricane or Berserking would be a better choice for now.)

You also gotta add 10% to the AP right? Because of blessing of might, which, well, is always up when you are in group anyway. So I guess the old AP enchants are the way to go.
Also, just to add that they DO work on ilvl higher than 300.

ingen · 12/22/10, 12:55 AM

Originally Posted by chippydip

The PvP gear is beastly right now. By my calculations, 4pc blue PvP gear is BiS pre-raid. You get a total of +250 Str from the set bonuses which more than makes up for the lack of a second stat on the pieces. +5% CS damage and -1 second Judge CD are just icing at that point.

This is pretty interesting, for my gear set, replacing two pieces of ilvl 346 with the Bloodthirsty equivalents is a DPS upgrade due to the sheer bonus in Strength, even using Redcape's values. Assuming you aren't losing any valuable stats like Hit from the swap-out, this is definitely worth considering pre-raid.

On a related note, has anyone else noticed that the 2pc T10 bonus is actually more valuable than any ilvl 346 gear? Even when using only 251/264 pieces it seems to be the case, 5% increase to all damage is quite a lot.

chippydip · 12/22/10, 1:23 AM

Originally Posted by Naididae

You also gotta add 10% to the AP right? Because of blessing of might, which, well, is always up when you are in group anyway. So I guess the old AP enchants are the way to go.

Actually no. The point of those calculations is to figure out how much Strength would have given that amount of AP. Since the +10% AP buff effects both, you don't need to worry about it. Plate specialization and Kings/MotW only effect Strength, though, which is why the 1.05 factors are in there.

Exemplar · 12/22/10, 3:06 AM

AP values from Redcape's spreadsheet and my spreadsheet added to OP. Wrath +AP enchants added to the table, as they do indeed beat the pants off other options in bracer and boots slots.

It appears PvP gloves one on one beat normal non-epic gloves (CS bonus helps). The other three pieces individually are not standout, but 2 piece (Str bonus) beats two non-epic pieces. 4 piece does the same (a second Str bonus). The reduced CD on Judgement is nice, but not the greatest gain.

I'm working on improvements to my spreadsheet's item comparison routine and don't expect to release an update until complete. With holidays that's likely next week sometime.

Tiara · 12/22/10, 3:15 AM

Originally Posted by chippydip

Actually no. The point of those calculations is to figure out how much Strength would have given that amount of AP. Since the +10% AP buff effects both, you don't need to worry about it. Plate specialization and Kings/MotW only effect Strength, though, which is why the 1.05 factors are in there.

Maybe i don't understand, but if you use the 65 haste enchant on your wrists don't gain anything extra. While if we enchant it with 50 AP we get an addition 5 ap from our Blessing of Might.
So the comparison should be how much strength 55 ap gives us compared to what 65 haste would give us (using the stat weightings).

So that would be 55 * 0.45 compared to 65 * 0.11 instead of 50 * 0.45 compared to 65 * 0.11.

But maybe i missed your point.

Since we're on it, and somebody asked this before, doesn't berserking blow hurricane out of the water? Hurricane gives you a 12 sec haste increase, has a 45 sec cooldown and a 12,5% chance to proc (if im not mistaken). Using some napkin math that would take us about a minute to proc one, giving you an uptime of ~20%. So hurricane would give you a static bonus of 450 * 0.2 = 90 haste. Compare that to the ~169 static AP bonus (if my memory serves me right) a berserking enchant would give you. It's not even close, even if you disregard the sloppy calculations.

edit.
i forgot about the decreased chance of procs on lvl 80 enchants

Simqn · 12/22/10, 3:30 AM

Level 80 enchants have massivly reduced proc chances at level 85 so their use becomes (intentionally) underoptimal as a viable option. Im sure someone here remembers the exact calculation for how reduced proc chances becomes above the intended item level but i dont remember it off hand.

Hogun · 12/22/10, 12:45 PM

This makes the 32 AP enchant to boots from Wrath a topic of debate then. From Exemplar's stat weights it is better than 50 hit. However under Redcape's numbers it comes out as inferior.

Exemplar · 12/22/10, 1:59 PM

Originally Posted by Hogun

This makes the 32 AP enchant to boots from Wrath a topic of debate then. From Exemplar's stat weights it is better than 50 hit. However under Redcape's numbers it comes out as inferior.

Don't debate it. Get the 32 AP. Then reforge 50 Crit/Haste/Mastery (which are worse than AP or Hit) into Hit if you need it. This will uniformly be a win as your 32 AP beat 50 Crit/Haste/Mastery. Via Redcape's #s you have AP to boots (32 * .45) + Reforge Crit to Hit gain (50 * (.42 - .20)) = 25.4 vs Hit to boots (50 * .42) = 21. The difference increases if it's Haste or Mastery you reforge.

Berserking without a proc reduction from level models below Avalanche in damage. I would use Avalanche until Maelstrom are readily available for Landslide.

T.K. · 12/22/10, 2:44 PM

Originally Posted by Exemplar

General rule of gear appears to be:
1 socket provides a socket bonus of 10 rating/Str
2 sockets provides a socket bonus of 20 rating/Str
3 sockets provides a socket bonus of 30 rating/Str

At best I'm finding 1 hit rating is worth .4 Str. Other ratings values are even lower. This impacts socketing.

One socket:
Swapping a single Rare Quality gem from 40 Str to 20 Str/20 rating for +10 Str would net you 30 Str and 20 rating. 20 rating is less increase than the lost 10 Str. This is a loss.

Two sockets:
Swapping a single Rare Quality gem from 40 Str to 20 Str/20 rating for +20 Str would net you 40 Str and 20 rating. You gain 20 rating. This is a gain.
Swapping two Rare Quality gem from 40 Str to 20 Str/20 rating would net you 60 Str and 40 rating. 40 rating is less increase than the lost 20 Str. This is a loss.

Three sockets:
Swapping a single Rare Quality gem from 40 Str to 20 Str/20 rating for +30 Str would net you 50 Str and 20 rating. You gain 10 Str and 20 rating. This is a gain.
Swapping two Rare Quality gem from 40 Str to 20 Str/20 rating would net you 70 Str and 40 rating. 40 rating is about equivalent to the lost 10 Str. This may be a small loss, break even, or small gain.
Swapping three Rare Quality gem from 40 Str to 20 Str/20 rating would net you 90 Str and 60 rating. 60 rating is less increase than the lost 30 Str. This is a loss.

Any socket bonus other than Str would uniformly be a loss. Even 30 rating for 20 Str is less DPS.

Helms tend to break the standard bonus rules. As noted, it's possible to see 30 Str with a single socket requirement. This would follow the 3-socket single gem replacement rule.

I've been test driving [Thundering Skyflare Diamond] as a Meta since we've had a formal patch with no correction to the Crit meta requirements. I'm getting procs from a minute to 2 minutes apart. It may be sharing cooldowns with other procs or its proc rate could be reduced at 80. I'm going to start modeling it at 6 seconds of 480 haste every 90 seconds. This comes to an average of 32 haste, which is far from phenomenal.

Since it requires a Green gem to activate you would lose 40 Str and gain 40 rating + 32 haste (effectively 72 rating). - this would only be viable in a single gem for 30 Str situation. 30 Str 72 rating > 40 Str, while 20 Str 72 rating < 40 Str. In my gear this means inactive meta for me.

Note: Based on my test I did find you must have a Meta in your hat to activate the head socket bonus, even if the Meta is inactive due to failed requirements.

While I agree and respect Exemplar's work all this time along, I have to rumbly say that I think this model might be incomplete.

In a gem by gem level I agree with his conclusion in wich gems are gains and loses and how we should be gemming slots, but the part I think it's missing it's the comparison between the overall gain from all STR gemming against the benefit provided by the Meta 3% extra crit damage and thus, the number of gem slots in your gear.

To better exemplify what I mean, I'll use absolute and round values to try to better provide my point:

Let's say that a STR/rating gem nets us 1500 damage over the course of a fight;
Let's say that a pure STR gem nets us 2000 damage over the course of a fight, and as such, better then a STR/rating gem;

With that, as Exemplar stated, if the socket bonus is 500 or more added damage, we have a win (that's the case of a STR socket bonus with only 1 gem, for example).

Let's now assume that the benefit from the crit rating and 3% extra crit damage from the meta nets us a 10000 damage over the course of the fight;

Although pure STR gems beat STR/rating gems if the socket doesn't add the difference, we gotta balance the overal gain against what we'd loose not using the meta, in my example 10000 extra damage over the fight, wich in this anedoctal example would be a break even point of 5 gems.

I couldn't figure by your post if this was accounted or not and if it was already considered I apologize in advance.

Calefax · 12/22/10, 2:54 PM

Originally Posted by Tiara

Since we're on it, and somebody asked this before, doesn't berserking blow hurricane out of the water? Hurricane gives you a 12 sec haste increase, has a 45 sec cooldown and a 12,5% chance to proc (if im not mistaken). Using some napkin math that would take us about a minute to proc one, giving you an uptime of ~20%. So hurricane would give you a static bonus of 450 * 0.2 = 90 haste. Compare that to the ~169 static AP bonus (if my memory serves me right) a berserking enchant would give you. It's not even close, even if you disregard the sloppy calculations.

edit.
i forgot about the decreased chance of procs on lvl 80 enchants

Your assumptions about hurricane are quite incorrect. There are actually 2 separate procs, one from melee and one from spells that can also STACK, and there does not appear to be an ICD of any kind as far as I can tell. I have been running hurricane for a few weeks now just because I was curious to know exactly how it worked. So far I've seen up times as low as 30% and as high as 66%, which mostly has to do with how much movement is going on in the fight. I have yet to see up times lower than 45% on fights where you don't need to stop dpsing ever, and surprisingly the 66% uptime that I saw was on a Valiona & Theralion kill (28 hurricane procs in 5 mins 32 seconds) which has a decent chunk of movement involved.

It's difficult (at least for me) to math out exactly what the theoretical dps gain is from hurricane, but I'd be willing to bet that with reduced proc rates of lv 80 enchants that it beats out berserking. Avalanche is very likely still a lot better though, and of course once landslide is readily available this discussion is completely moot.

Tiara · 12/22/10, 3:27 PM

Originally Posted by Calefax

Your assumptions about hurricane are quite incorrect. There are actually 2 separate procs, one from melee and one from spells that can also STACK, and there does not appear to be an ICD of any kind as far as I can tell. I have been running hurricane for a few weeks now just because I was curious to know exactly how it worked. So far I've seen up times as low as 30% and as high as 66%, which mostly has to do with how much movement is going on in the fight. I have yet to see up times lower than 45% on fights where you don't need to stop dpsing ever, and surprisingly the 66% uptime that I saw was on a Valiona & Theralion kill (28 hurricane procs in 5 mins 32 seconds) which has a decent chunk of movement involved.

It's difficult (at least for me) to math out exactly what the theoretical dps gain is from hurricane, but I'd be willing to bet that with reduced proc rates of lv 80 enchants that it beats out berserking. Avalanche is very likely still a lot better though, and of course once landslide is readily available this discussion is completely moot.

Yeah i was a bit quick to jump the gun on that, that'll teach me to base my calculations on information found in wowhead comments

Mist · 12/22/10, 7:00 PM

Is there a DPS priority list for trinkets yet? I got offered a Hurricane deck cheaply and I'm wondering if it's worth it.

PS: 1 Point in Acts of Sacrifice should be considered almost mandatory for the spec, not merely honorable mention. It even gets rid of most snares from aura/area spell temporarily, like the Desecration effect on Commander Springvale in SFK. This allows you to then save your HoFreedoms for when someone else needs snare removal rather than using it on yourself.

♦ Rodimus Prime · 12/22/10, 7:15 PM

Originally Posted by Mist

Is there a DPS priority list for trinkets yet? I got offered a Hurricane deck cheaply and I'm wondering if it's worth it.

AFAIK, it is, especially with current weightings.

arthureld · 12/23/10, 1:18 AM

Speaking of trinkets, I got my hands of Fury of the Angerforge tonight. As an FYI for people considering it, as it stands right now, it's a pretty annoying trinket on any fight with movement or target switches. The buff you can gain from a melee attack lasts 15s, but there's an ICD of 10s, so if you get unlucky in that 5s window, you lose all your stacks. The ramp up time is pretty long already, so I've gone 6m fights without being able to use the trinket at all. I'm hoping we can get them to either eliminate the ICD or lengthen the buff, but for now, it can be terribly frustrating.

Exemplar · 12/23/10, 1:38 AM

Originally Posted by T.K.

Although pure STR gems beat STR/rating gems if the socket doesn't add the difference, we gotta balance the overall gain against what we'd loose not using the meta, in my example 10000 extra damage over the fight, which in this anecdotal example would be a break even point of 5 gems.

I couldn't figure by your post if this was accounted or not and if it was already considered I apologize in advance.

Except a Purple Gem counts as Red and Blue. To activate the +Crit % meta you would need to use 1 straight blue and the rest purple to qualify. Going for other socket bonuses would just make it worse (if you use an orange, you now need a straight blue to step ahead, otherwise you must use a Yellow). The more gem slots you have, the more possible Str you lose.

In Wrath BIS gear it was less DPS improvement meeting the more blue than red requirement for the crit bonus than it was to use absolutely no meta gem at all (much less the destructive). If you're not sure about your gear, use one of the spreadsheets and try gemming for each meta and see which produces more DPS in your gear.

Originally Posted by Kawke

I would think not. Until we can get landslide id say go with Hurricane if you have a good weapon to put it on. If your talking about a herioc blue that maybe it isnt to bad. For me I have Zin Rokh so I put Hurricane on it in the mean time. It procs really well and I don't believe it has a ICD.

I haven't crunched further numbers since we've learned more about the dual-proc of Hurricane, but I believe it's still less gain than Avalanche. Until we have hard math to prove Hurricane superior I am going to continue to recommend Avalanche until we have the Maelstrom to make Landslide possible.