• landsoul

Haste does absolutely affect rage generation. Rage is give on a hit!! Think of it this way: You get a static amount of rage based on if you crit, or non crit, with added rage based on the damage dealt. The formula was released by blizzard for crying out loud. If you have haste, it means you are hitting more frequently than without haste, granting additional rage PER SECOND NOT per swing.

The thing is is: If you are swinging more frequently, then also you have opportunity to use HS more frequently, which requires the increased rage you get from haste at the same rate. While the rage generation is higher, the rage consumed is also higher, with a net change at or near 0.

Haste gains are definately non-linear in this sense beacuse of the increased amount of heroic strike, which negates the miss chance in the off hand, which increases the probability you can proc an effect.

 

raja989

I've been searching the elitist warrior dps forums and havn't been able to find anything on hit rating as far as what the cap is for DW fury war, I've read that the 2h cap is 96, but nothing on the DW cap, is there a fury DW cap or just speculation?

 

Axodry

[quote=Machinator;695077]Shard of Contempt proc is Item - Sunwell Dungeon Melee Trinket - Spells - World of Warcraft

Seems to be 10% chance on hit, with a 45s hidden cooldown, which is something like 30-40% uptime. Its much better than it first looks. QUOTE]

When you say hidden, do you mean that for instance it shares a cooldown with Hourglass and that only one of the two trinkets can proc every 45s or so or that SoC itself can only proc once every 45s thus meaning it would stack with ourglass...i'm thinking the latter? I just got my SoC last night and haven't been able to do any testing yet to see if they do indeed stack. I apologize if this has been covered elsewhere.
Thanks,

 

• Apate

It means that it isn't in the tooltip.

See you, auntie.
"You don't need a machine to make a rainbow. For rainbows are made of happy thoughts, and dreams, and chocolate unicorns, and gumdrops, and licorice sunsets, and fuzzy gumdrop bears, and sugar-coated chocolate gumdrop land."

Originally Posted by DeeNogger I am coming for you Apate.

 

Machinator

The question is if the speed variable in the rage formula is affected by haste, like PPM enchants. If it is then haste would have no effect at all on rage per second.

The whole thing with haste affecting HS is the feasibility of using HS on every MH swing, which people keep shooting down every time it comes up so I'd like people to back it up with something because you need a hell of a lot of rage to do that. Or why they dont just use a faster MH then, which would be more efficient than stacking haste.

"Information is ammunition."

 

• landsoul

(On Haste and Rage generation)
Well then to test it out, go to blasted lands with a shaman.
I will do it as soon as I get time and opportunity.

1) Use a grey low level weapon
2) Have the shaman tank
3) Do not dual-wield or record when flurried or proc unbridled wrath.
4) hit (not crit) the mob, record rage gained.
5) have shaman pop heroism
6) hit (not crit) the mob, record rage gained.

If all these variables are true, you should hit for a static damage range both times, one being hasted, and one not, with both being hits and not crits. Be mindful also of anger management

If all these speed-modified assumtions are correct, then:
rage gained per hit = .01365d + (f * "attackspeed") / 2

If the assumptions are not correct, then:
rage gained per hit = .01365d + (f * "Weaponspeed") / 2

 

Machinator

I will test when I get a chance, but I dont understand why you would need a low level weapon or not crit. Both still allow you to determine if haste changed anything. Also flurry and haste should be tested separately.

"Information is ammunition."

 

dr_AllCOM3

Low level weapons have a small minimum range.

I'm very interested in some results too. Luckly it's easier to test than PPM .

 

hamsterboat

so im using this spread sheet, i have roughly 1360 armor pen with divine ret, using my current gear its stating divine ret is higher dps than felfury legs and even the leather imortal night legs. i dont know why, i doubt its right. 40 str 50 haste vs 350 arp.

 

Rezarel

I dug around and realized I still have the movie I made a couple weeks ago using [Gorehowl]. Looking at the last hit in the movie, a 1423 damage crit that gave 32 rage, while flurried and with the dragonspine haste up:

If the formula uses hasted weapon speed
rage = (1423 * 7.5 / 274.7 + 7 * 3.6 / (1.25 * (1 + 325 / 1577))) / 2 = 27.8

If the formula uses base weapon speed
rage = (1423 * 7.5 / 274.7 + 7 * 3.6) / 2 = 32.0

There's a pretty big difference between the two, so I'm pretty confident the second is right.

Attached Files

Rage Gen test 2008-03-17 2.3 live.wmv (691.6 KB, 97 views)

 

Machinator

Cool, so haste does increase rage gain. Ill have to edit my sheet and see what effect that has.

Edit: My sheet did already have that, just used current speed for UW. As far as I can tell, the other sheets use base speed as well. I am still not seeing much evidence that haste is better than other primary warrior stats with any reasonable gear.

"Information is ammunition."

 

Gruntle

Not that I would know on my gear level but I do think that HS on almost every hit is possible in BT+ level gear making haste scale very good. As to why people don't equip a fast MH, well ask them. They erroneously (in my opinion) think that the tiny amount of extra WW damage from a slower weapon is higher than the dps gain of being able to use HS optimally. And even if you're not able to make HS on every single hit you will still be able to churn out more HS hits over a given time period with a fast weapon (or haste equipment).

A faster weapon will enable you to do more HS swings but it will not supply the rage that you need to do so. Haste does both and that's what I mean by "slightly non-linear".

edit: solid evidence Rezarel, thanks for posting it

 

Shha

If it indeed increases rage generation , then its even better (i wasnt sure, so i assumed the worst - that it reduces it)
Following conclusions then:

a) Haste INCREASES your heroic:white ratio. Its pretty obvious but in short. Your instants "eat" certain CONSTANT amount of rage. That number doesnt grow with haste. Now instead of making complicated calculations of how more rage generation increases number of heroic, which in turn reduces rage generation, just use simple but accurate approximation. Just pretend that heroic strikes also generate rage, but their cost is not 12 rage but 12+average rage from MH swing. Now since the TOTAL rage gained is amplified by x% depending on haste, but the rage for heroic = base rage+x%- constant instant rage, we can see that the rage for heroic went up more then x%.

b) Therefore Haste not only increases your heroic+white damage by the percentage (assuming like most warriors you are at 0 haste) but also has secondary effect of increasing heroic% , further amplifying damage. So X% haste increases white+heroic by MORE then x% for 0 haste warriors.

c) Even taking a more careful approach of 65% white + heroic, lets see for 88 haste rating vs 88 crit rating. Its 5.5 haste% vs 4% crit. Haste increases the 65% of your damage by MORE then 5.5 but lets assume the 5.5. 65%x5.5% = 3,58% increase in damage. It seems to be low then 4% from crit, but since warriors have a lot of crit already lets check it for 45% raid buffed crit. white damage at 35% not affected by impale - 1.49/1.45 = 2.76 increase. 65% damage affected by impale - 1+(0.49x1.2)/1+(0.45x1.2)= 1.59/1.54 3.25% increase.

35%x2.76 + 65% x3.25 = 3.07% damage increase. Haste is ahead.



Logical explanation. First - haste IS worse stat then crit. But the more important rule is balancing out your stats. Its like the old prebc dilemma of crit vs ap. And the answer was always the same - you cant just boost one with totally neglecting other. The more crit you have , the more valuable ap becomes in comparison to crit. Other way the same.

Now in BC we have in a way 4 stats "multipling" our damage.
AP, crit, haste, ArP.

First 3 have constant diminishing returns - the more you have of them the less % dps increase will you see by adding another constant amount of it. Last actually doesnt, in fact it has increasing returns, however its limited by 2 factors : ArP cap on some bosses, and lower returns on high armor ones (Brutallus comes to mind).

Therefore when looking for stat value we can see following rules:

a) At some point ArP will get ahead of other stats since it actually increases in value rather then decreases. The more crit ap haste AND ArP you have the more valuable ArP becomes. In a way it simply means : "the higher the ilvl of your gear the better ArP becomes). For haste i believe the point is actually at 0 haste, meaning that ArP will ALWAYS be better then haste (within reason - look above for caps).

b) For rest of stats we need to find balance. The more AP we have, the better crit and haste become. The more crit we have the better AP and haste etc. THATS the answer to machinator dilemma. Haste on its own is worse then AP/Crit. Its just that warriors with BT gear pushed AP/Crit so high, that diminishing returns on those put them below haste. Remember you look at value of haste for someone at ZERO haste and FORTY FIVE crit. Thats way "offbalance". Im sure if you put 45% haste into your spreadsheet, the haste value would go way below crit.

 

Graul

I still think Landsoul's sheet was a little more "real world" for DPS figures, but the above fits to explain the strangeness I was seeing in Grim's sheet with haste. The more Sunwell gear you added (crit, AP, exp, haste) the more valuable haste was with SEP. It's value would increase with every item you added that had some haste on it, but it would do so similar to how ArP scales in value starting with 0. It's not like that in Landsoul's, but then it doesn't seem like anyone has really come up with the best way to model haste yet.

 

Jevade

I am getting different results depending on my spreadsheet and gear in terms of which will give me more DPS when i compare [Shard of Contempt] Vs [Berserker's Call] My expertise is currently 5/5 DW axes. What about you guys? Dr. Al said bosses might have a higher dodge % than originally expected- believed to be 6.5%- both spreadsheets account for that?

 

Machinator

Well this is why I said reasonable gear level, haste would become best once all your other stats are stacked, but that point seems to be beyond the maximum people can get. Haste does increase rage generation and white damage, but the other stats all do too and do not diminish enough to make haste better with current gear.

To test the value of Haste at different stat levels I totaled up the Item points from the gear I was using and am doing things more in sandbox mode with manual stats. There are ~2000 item points, I am leaving the weapons, and the trinkets.

Edit: Damn it didnt meant to hit post yet. Still getting numbers at various gear combos.

Edit 2: I have been working on figuring out the exact dodge %. It is definately more then 5.6%, beyond that I dont know.

"Information is ammunition."

 

Serpentine

hello,
Its my first post here so I just hope I dont gona ask a qustion that been asked million times before.
My guild started MH / BT not long ago and with that came the new weapons and gear .
I soon got hold over [Claw of Molten Fury] as MH and [Swiftsteel Bludgeon] as OH
now, when I run my gear throw the spread sheet , it seems that Dragonstrike yields more dps.
could it be?
and if so should I just stick with Dragonstrike and give up on the new shiny epics?

 

Kaoz

This would throw every single spreadsheet out there out of wack. They all use the hasted weapon speed for rage generation. Considering flurry, DST and the presence of haste on lots of gear, it's going to have a pretty big impact.

It was so simple to test, really can't believe we'd miss something this huge. Next up is the impact of haste on PPM abilities like enchants and windfury. This would also have a HUGE impact.

 

Amyannirving

I've been thinking about the whole haste issue ever since I started using Grim's spreadsheets (which is from late SSC/TK to now where we're farming illidan and starting sunwell). Things like swiftsteel shoulders, seemingly a butt of jokes from all dps warriors, are modeled here to be better than Blood-Stained, and I still haven't upgraded my mainhand yet because the spreadsheet models it to be better than even the S3 mainhands. The only theory I have really is that it WF is a flat 20% chance to gain an extra mainhand attack and thus would benefit from having more white attacks, and that the number of heroic strikes you can use between BT/WW cooldowns can potentially be increased if you have enough haste. Or it's possible that the spreadsheet is miscalculating something in the simulation and I've been a fool for consulting it all this time. Then again though, the numbers it's given me have always been pretty on the ball when it comes to what I see in tank-and-spank fights.

 

• landsoul

Well on the contrary, my sheet never used attack speed for rage generation or PPM abilities. Windfury is not a PPM ability.

Yes increased attack speed would have an impact, but keep in mind the difference is not large and results do not exponentially grow out of control. In my sheet tests I have not seen the numbers build up to an unreasonable amount. Average uptime of any given proc cannot exceed one, and in reality will never reach one. The effect of haste increasing returns of procs is so small that it does not push haste up in worth much to have a HUGE impact. You can see all of this in my spreadsheet.

It amazes me how tests are publised showing small inconsistencies of random data, displayed from mods, and then having intuitous players try to come up with a reason for the variance in this data, like fixing the idea of: "all formulae that uses speed as a factor uses current attack speed instead of listed weapon speed."

While I agree this is why this forum group exists, there really needs to be more accurate and convincing testing. WWS is a step in the right direction, but still inaccurate due to the arguably uncontrolled environment that the WWS pulls its data from, and the failure of people to realize this fact and still making conjectures from it.

6.5% dodge? show us hard evidence in a controlled environment.

Haste affecting PPM? That may be very hard to prove via evidence. A good test neerds to be designed.

 

Machinator

The sheets all seemed to use the base speed anyway, so this does not change anything, just confirmation on something that was assumed. Haste on PPM abilities was tested, and windfury is a flat % of MH hits.

The real question is why Grim's sheet shows haste better than others, and what is the most accurate model.

Also that weapon mastery + 15 expertise still gets dodges means the dodge rate is not 5.6% or we have a mechanic wrong.

"Information is ammunition."

 

• landsoul

[Dragonstrike] really is a keeper and there are only sidegrades until Season 3 weapon. There's no real need to replace it, especially if you are human. Swiftsteel Bludgeon is fine also and it has decent stats, yet a slower weapon will give you a little more WW damage.

 

• landsoul