takel

Guys, I appreciate the drive and intent to evaluate the numeric value of some of these talents, but I don't believe that arguing over the validity or methodology of a modelling attempt is in anyway beneficial to this thread in regards to WotLK hunter talents and abilities.

While I dislike seeming to overstep lines and pretend to be an armchair moderator, I don't think it's all that courteous to people who want to dip in and find out more about the current talents and ability mechanics to end up being caught in the crossfire of an argument over napkin maths.


Regardless, on the question of LnL for a hunter, it would really come down to how hard Arcane shot would be hitting for... In any case, we need to know if Chimera has the potential to proc the effect, then we have to figure out the raid-wide implications if we do choose to drop some potential talent sacrifices to pick up LnL. I must admit, I really like the idea of LnL as a fun little burst tool and I really hope Blizzard adds more of these kind of event->action response mechanics, but Arcane Shot's poorer scaling as well as the talent sacrifices required have me leaning away from it all.
We could split out the numbers for all we like, but until we do see some 'Dungeon 4' set gear so we know what kind of stats we'll be packing, there's no real way to determine how our DPS will line up with the MM's shot selections. For Survival it's fairly self-evident due to TNT and Explosive's synergy that the talent is a DPS boost plus the HP link up, but for MM the only use of the talent would be to boost DPS and perhaps save a bit of mana.

 

Tyranna

back on topic good news for you wolf lovers.

MMO-Champion BlueTracker - [Feedback]Wolf Skill: Furious Howl
I'm not sure why Furious Howl had such a low range. My suspicion is it just hadn't been touched in a long time.

It's 40 now.

 

Eurytos

I based my stats based on comparible numbers for what has been released. Blacksmiths have an item level 200 2h mace that is 186.5 DPS. I just took all the scaling from that and come up with an estimate for stats.

 

Eurytos

Hell yeah... I was already going to be using a wolf... now I'm DEFINITELY going to be using a wolf. Oh how I wish I tamed a Ghost Wolf when I had the chance.

 

Zurai

You specifically asked us to correct you if the math was wrong. We did, and you took umbrage to it.

As for the comparative math, using your numbers and just correcting the rounding, we get 21.685 DPS and 42 MP5. Those are a 7.8% difference and a 5% difference, respectively. Simply by correcting one small rounding error, we got a result that was almost 10% different from your result. The actual magnitude of the numbers isn't big, true, but you give every indication that you'd use this same style of "math" in all cases, and when you're calculating expected DPS over an entire fight, a 7.8% difference can mean 200-300 actual DPS being lost in translation.

The amusing/frustrating thing about this is that you apparently don't even realize that, by ignoring the average case, you're making your entire conclusion vastly less reliable as you decrease your sample size, which is what you've given as the strength of your reasoning. As sample size decreases, the results will deviate more and more from the average, and you've said many times that you object to results deviating from the average by very much.

 

Trogdor

And this goes back to my post about the "waterfall". Blizzard doesn't need to theorycraft because they have the numbers, and we simply try to simulate it. Blizzard knows what's going over the waterfall; but I want to know what's flowing downstream. I don't care that my math says 44.0 and super-precision says 45.87. It's in the ballpark of 45. If it's off a percent or two in actual performance, whatever.. because that's how random number generation works. That's all I need. Everyone else needs to get over trying to force super-decimals down my throat, honestly. I just want to know the nutshell.. or if the peanut is going to taste good, not the chemical makeup of the peanuts and a dissertation about health benefits with expansive colorful graphs.

But A+ for maintaining hardcore 3rd grade probability, with no flexibility to unfocus a bit and see the general picture.



Well that's because I don't care about a 1.685 DPS difference.. or a 2 MP5 difference. You're looking at 7.8% and 5%, and yeah that seems a lot. But I'm looking at the end. Is a 1.685 DPS difference such a big deal when it's compared to 1500+ DPS output? Is a 2 MP5 difference so big when it's compared to 150+ MP5? To me it doesn't, and I don't choose to stress over such little details because there are more important things to me. I'm also not telling people they shouldn't stress about it, because if they want to.. sure do it.

 

• Intermission

Blizzard have the exact same numbers that we do. They dont have some magic machine that knows what is going to happen, they simply derive the expected results just like we do.

If you enjoy changing a number for the hell of it, with no gain other than to be inaccurate, then go ahead, but it's probably not the best idea to publish it in a thread.

I don't know why I'm clogging up this thread with an argument with someone that believes that 5 heads = more chance of tails. Apologies to everyone else, I get trolled easily. I'll be quiet now.

 

takel

Whoo hoo! Time to pull out my poor neglected doggy out of the stables. Perhaps this may be a sign for things to come? (perhaps all party auras/buffs are going to be 40 yrds?)

 

Eurytos

Most are... with talents.

 

Malan

Then why are you still defending math that a 6th grader could tell you was wrong?

Shitting up every single thread on EJ since '06

 

Trogdor

The probability of having 6, consecutive coin flips ending up as heads is 1/64. The probability of a single flip ending up as tails is 1/2. There's a greater chance for a flip to end up as tails than to have 6 consecutive heads. On a flip per flip basis, it's all 50%. I mentioned the 5 crits in a row because the probability is low, and that it goes both ways. People have been saying "it's all about the averages" yet the averages don't dictate what happens in a short window. The short window is the practical window because it's the combat window. Of course everything averages out in the long run when you add everything up, but I'm not doing that work.

You mentioned probability. The probability of a 6th consecutive head is less than a single flip of tails. I didn't go try to attack you and put words in your mouth or discredit you, even if I might not have completely understood what you meant (probability versus odds). You might want to employ some tact and respect.

When everyone is trying to argue with you and tell you you have to do things "their way", AND when others purposely misconstrue your statements instead of trying to clarify the idea.. it's pretty hard to get the idea across. Maybe.. just maybe, if a few certain of you accepted what I've posted for what it's worth, instead of trying to drill in your own bias (backed by ego), you might have an easier time understanding.

 

Catalept

TBH, I can't see a downside to standardizing all fixed-range buffs to 40 yards and leaving it at that. Micromanaging my range according to various conditions sounds un-fun


Edit: Oh God... it's the 0.999~ thread all over >.<

 

Tyranna

I thought totems were changed to 30 yard base and the talent to increase their range was removed.

 

takel

Neither can I. All healing spells were changed to 40 yards.
Some abilities which had dual purposes were changed to 30yrd offensive, 40yrd 'defensive' so I'm wondering why is there still a stratification between buff ranges from most other beneficial effect in the game. Heck, totem ranges were bitched at way back in the AQ20 days and with the emphasis on more dynamic encounters, I can't imagine the rationale behind keeping totem ranges to 30yrds.

 

Eurytos

Somomabitch... I didn't see the part where they took the talent away...

Well... Sniper Training is downright horrible then.

 

• Intermission

Sorry, I really cant help myself.

Then what does dictate this? "Random number generator"? The thing that says: you have 50% crit, so I'll flip a coin and heads = crit, tails = not. Which by definition, creates a predicted average of 50%.
It's the exact same level of work. How on earth can it be more work when it's the same calculation, with a "multiply by 100" on the end?

--

Regarding totem range:

I think the totem range is a limitation that they intend to have. Totems lost the party limitation, which is great. So now we can worry about ranges and positioning rather than what person to bring in what group. A far more interesting thing to work around, and becomes an additional thing to think about when optimizing strategies.

 

Eurytos

I agree... it's actually less work. It took a lot more effort on your part to build each of those scenarios then it did for me to make average formula's and plug in the numbers. Of course, QuickMath definitely helps.

 

Trogdor

Take this link:
WotLK Talents & Abilities Discussion

Apply greater precision to my math (when I said in the post to "correct my math", I meant stuff that pertained to fat-fingering a calculator, not dictating where I was supposed to round). Use the same base stuff... not adding buffs or ArP or whatever. Just keep it close and use whatever best average you wish. Show the results. Then we can go see how big a difference the "perfectly accurate" calculations are from a "rounded-off eyeball".

You're saying: This is right, this is what it really is, this is how it's supposed to be, this is the real math, you're wrong. I'm asking you to show me differences, and from that you can try to convince me to care about the decimals and how big of an impact they make.

I threw up my math and everyone got to criticize it. Lets see you do the same for your point of view. Otherwise it's just all big talk.

 

Catalept

I don't know... I can certainly see situations where an enhancement shammy drops a windfury down before rushing in, but when they refresh it, instead of being nicely between the ranged and melee DPS, it now appears at the feet of the melee, forcing ranged to DPS to shuffle in (if possible). Also, as mentioned before, there's currently no feedback to us about whether or not the KS bonus from Sniper Training is in effect...

Perhaps ST is just a way of rewarding hunters for managing to get that much distance from someone in PvP, and it's just not a PvE talent at all. TBH, given the number of position-dependant encounters in TBC, I certainly don't look forward to busting out the trig when we're all zipping around during Felmyst v2.0.

 

Eurytos

I like it for PvP, even without the range bonus, simply because <20% is the most important time to do high damage. But yeah... it's junk for PvE.

 

Berfert

He was talking about the probability of a coin flip resulting in heads once you had already flipped that coin and had 5 heads in a row. Once the first 5 are flipped, the outcome of those flips has no relevance on what the result of the 6th flip will be. See this article for details.

Your original statement about tails made it sound like you didn't understand this. This, your more recent statement, implies otherwise.

Now, for more fun with probability, see: the Monty Hall problem

 

• Intermission

For the 3rd time, this has nothing to do with your maths, of which I did not read. It has nothing to do with LnL. I just wanted to correct someone that was looking at expected damage (ie, theorycrafting as a whole) in a completely warped way, making sure that nobody else actually thought that was the right way to look at it.

You could have said "this is a rough calculation, I did some hefty rounding", instead of "proc rate or accuracy isn't important, as I'm only concerned about a small time period, not a long one", and nobody would of even questioned it.


Yeah, I can definitely see situations where it will be a pain. However those times are really quite few a far between, and when they occur, it's not a huge impact for the vast majority of raid members. I'll throw out a few scenarios:

You have 3 or more shaman (almost all raids do at the moment, but hard to say in WotLK obviously). One drops WF at/within melee range (enhance, or resto if none). One drops WF at at ranged raid area. One drops WoA.

Now lets be more realistic. We have 3 shaman, but the fight is similar to Felmyst, where there are 3 main groups of people, quite some distance from each other. Now we need to think how we can get the most totem coverage. The melee can still get a WF. Now we can let one ranged group get WoA, and one get WF. Or both can get WoA and the hunters can move a bit closer to the boss if possible.

There is going to be a hell of a lot of variables on every fight and every raid make up, which I think is great. On the vast majority of fights though, I think "move closer" will do the trick if you are shaman starved, or if the WoA users are split 50-50 in different areas.


As for Sniper Training, I'm not really impressed, and wouldnt rate it above most other talents. It will most likely come down to a progression-based choice. You know what fights you are working on, you know what fights you can get range on, and you know how your guild handles the ranged in regards to WF totem positioning or not. So you can base your spec on that.

Putting this into a current in-game example:
In Sunwell, there is 1 fight where you can make use of Sniper Training, KJ.
Kalec/Sothravvarvvavrar = too close (almost all the time, anyway, dependant on strat)
Brut = too close
Felmyst = too close (during ground at least, so 90% of the time)
Twins = too close
Muru = too close
KJ = KJ's hitbox is huge, I never remember seeing myself out of range to hit him from anywhere in the room. Hugging the wall would probably do it though.

So a SV hunter in Sunwell would not spec into it. However, compare it to Kara:
Attumen = fine
Moroes = fine
Maiden = fine, just stand near door and be in LoS of a dispeller
Opera = fine for Wolf and RnJ, Oz probably not all mobs
Nightbane = fine
Curator = fine
Aran = too close
Netherspite = massive hitbox, so probably healing range issues, and cant get in beam at same time
Illhoof = fine
Prince = fine

So a Kara hunter would value those points.

Thats just a lot of words for "until we know what we're fighting, its hard to say its better than X".


Edit: that Monty Hall problem is really cool. The "aids to understanding" part really helped.

 

Trogdor

So you're going to jump in the middle of a topic without going all the way back to the beginning? Impressive. I did say it was "paper math", greater than napkin math but less than graphpaper math. And until you come up with numbers using your preferred accuracy with my model, you can't prove my math is so warped as you claim. And I did say that I was looking at the small combat window frames. I also did say that I wasn't concerned with extended, summed up data. If not at the beginning, in the resulting posts, trying to explain my position.

I should have clarified [better] that greater accuracy didn't matter to me. I didn't intend to make it sound that it didn't matter overall.

 

• Intermission

Yet another example of you not getting it, and why pointing out inaccuracies of pointless maths is irrelevant (no matter how small or large), when you cant even understand why they exist. These two frames have the same expected dps. THE SAME.

 

Prinsesa

I didn't see a complete response to this in the 4 pages since it was asked, so:

Judgement of Wisdom returns mana equal to (9% of the Paladin's AP + 9% of the Paladin's SP). It is now a 100% guaranteed chance to proc, but has an internal cooldown of 4 seconds.

If we take a Sunwell-geared Paladin as an example, he's going to have about 2.4k AP unbuffed, which works out to 351 MP5.

If we slap on complete raid buffs, that jumps to about 4.1k AP, or 600 MP5.

"We do want Sanctuary to be the tanking seal"

- Ghostcrawler