[Warrior] Levelling a new Warrior

Ret, Holy, PVP, etc

Moderators: Fridmarr, Worldie, Aergis, Baelor

Re: [Warrior] Levelling a new Warrior

Postby fafhrd » Thu Jul 09, 2009 11:48 am

Spectrum wrote:To me, rage management isn't fun. Managing cooldowns and procs is more fun.


You manage cooldowns and procs plently as fury, deathwish, imp zerker rage, bloodsurge all need to be managed with a modicum of intelligence (i.e. no more than arms needs).

If you mean you like watching your MS/OP cooldowns and your rend tick timer, and filling every single GCD, well I do that all night on both specs of my hunter, don't particularly care to on a warrior too - debating how much I can HS to keep rage steady and bloodsurge procs coming enough to fill the non-WW/BT gcds and when it's worth using a gcd on a imp zerkerrage is more entertaining for me personally, and the screenful of numbers from playing a DW class is icing (that part was better with dual 1hs though, pity that will probably never be viable again >.>).
ImageImage
1/1 Lore pre-nerf.
User avatar
fafhrd
 
Posts: 5432
Joined: Fri Aug 31, 2007 2:31 pm

Re: [Warrior] Levelling a new Warrior

Postby Jedah » Thu Jul 09, 2009 12:03 pm

fafhrd wrote:You manage cooldowns and procs plently as fury, deathwish, imp zerker rage, bloodsurge all need to be managed with a modicum of intelligence (i.e. no more than arms needs).


Don't forget bloodrage!


Spectrum wrote:For me, Arms is way more fun, so I'm sticking with it.


For me, Arms is just like playing a flashy version of Whack-A-Mole, except I can't swing my padded mallet while moving. This translates to not fun, but hey... to each his own.
Jedah
 
Posts: 313
Joined: Fri Mar 28, 2008 8:10 am

Re: [Warrior] Levelling a new Warrior

Postby fafhrd » Thu Jul 09, 2009 2:44 pm

Jedah wrote:
fafhrd wrote:You manage cooldowns and procs plently as fury, deathwish, imp zerker rage, bloodsurge all need to be managed with a modicum of intelligence (i.e. no more than arms needs).


Don't forget bloodrage!


was leaving out things arms can use too :D
ImageImage
1/1 Lore pre-nerf.
User avatar
fafhrd
 
Posts: 5432
Joined: Fri Aug 31, 2007 2:31 pm

Re: [Warrior] Levelling a new Warrior

Postby Jedah » Fri Jul 10, 2009 8:15 am

fafhrd wrote:was leaving out things arms can use too :D


But when does arms ever need to use bloodrage other than in conjunction with charge on the pull? (excluding Vezax)
Jedah
 
Posts: 313
Joined: Fri Mar 28, 2008 8:10 am

Re: [Warrior] Levelling a new Warrior

Postby Meliora » Fri Jul 10, 2009 9:31 pm

Guild has asked me to switch mains from protadin to prot warrior and have had mild success. I was curious if anyone knew a nice resource site like Maintankadin for warriors?
Image
User avatar
Meliora
 
Posts: 150
Joined: Mon Apr 14, 2008 1:09 pm
Location: San Francisco, CA

Re: [Warrior] Levelling a new Warrior

Postby fafhrd » Fri Jul 10, 2009 9:54 pm

probably tankspot, except it's the (2nd) worst place on the internet apparently.
ImageImage
1/1 Lore pre-nerf.
User avatar
fafhrd
 
Posts: 5432
Joined: Fri Aug 31, 2007 2:31 pm

Re: [Warrior] Levelling a new Warrior

Postby Bundy » Fri Jul 10, 2009 10:51 pm

I hate to be blunt but the amount of fail in this thread is amazing considering the usual quality of this site.

Here's some simple facts to try to address some of the incorrect points made thus far:



Where to begin.....

Play styles differ, so there for, specs/glyphs will differ. Just because you use something doesnt make it the geatest thing ever. Saying one spec with one set of glyphs is best for all aspects of the game is an incorrect point.

Armor Penatration is the same as Armor itself in a way. Its more uselful the more Armor the mob has. Stacking armor against a mob that hits fast and not so hard is not a good plan. And stacking Armor Pen againt a mob that doesnt have alot of armor to begin with is also not a good plan. Its all percentage based, 40% of 4000 is 1600 armor reduction, where 40% of 30,000 is 12,000. The more armor the mob has, the more effective your one stat is gonna be. AP always works the same though.

Cookie cutter specs, if you wanna use one good for you. But I like to think for myself and try things out for myelf.

When the big change happened to Arms, I went arms and at that time SD used all your rage. I was still leveling at the time so I didnt like it because it used all my rage. It seemed like more for pvp to me at the time, I mean if you did that mid fight in pve, your dps would come to a sudden stop because you would be rage starved. Thats how I thought of it.

Anyways, my DPS is just fine right now so im not worried about it at all. Its my OS and im in 2-3 epics and the rest blues and im doing like 3k dps on single targets. When I come across a warrior in similar gear doing more than just a few dps more than me, I will take a look at his spec/glyph choice and consider possible changes. Nothing is set in stone for me.

The point of talents and glyphs is to make your character unique. If everyone did the exact same thing, well thats just boring as hell and im not paying money to be bored.
Bundy
 
Posts: 171
Joined: Sun May 31, 2009 2:41 am

Re: [Warrior] Levelling a new Warrior

Postby Epimer » Sat Jul 11, 2009 11:33 am

Bundy wrote:Play styles differ, so there for, specs/glyphs will differ. Just because you use something doesnt make it the geatest thing ever. Saying one spec with one set of glyphs is best for all aspects of the game is an incorrect point...

Cookie cutter specs, if you wanna use one good for you. But I like to think for myself and try things out for myelf...

The point of talents and glyphs is to make your character unique. If everyone did the exact same thing, well thats just boring as hell and im not paying money to be bored.


Part of what I enjoy about healing on my resto shaman is that it's almost all situational. The spells you use, how you glyph - they really all depend on the situation. There is, however, a theoretically "best" way to achieve maximum throughput, and a theoretically "best" way to achieve maximum longevity - it's just that you wouldn't normally gear to one extreme or the other for most fights.

In PvE, however, DPS has one concern (setting aside raid utility) - and that's doing the highest possible DPS. Like it or not, WoW is basically one giant spreadsheet, and there are mathematically optimum ways to do things. In this sense, there absolutely is a "best" talent spec, and a "best" selection of glyphs, which will do the most damage possible when combined with an optimum rotation. This optimisation is a big part of the game for people who like to theorycraft, and if you're not one of those people, you can do one of two things: take on what these people find out and use it to improve your toon and playtime, or stick your head in the sand and be a unique snowflake who's doing less damage than they're capable of.
User avatar
Epimer
 
Posts: 1496
Joined: Wed Jun 18, 2008 6:54 am

Re: [Warrior] Levelling a new Warrior

Postby Bundy » Sat Jul 11, 2009 10:17 pm

In PvE, however, DPS has one concern (setting aside raid utility) - and that's doing the highest possible DPS. Like it or not, WoW is basically one giant spreadsheet, and there are mathematically optimum ways to do things. In this sense, there absolutely is a "best" talent spec, and a "best" selection of glyphs, which will do the most damage possible when combined with an optimum rotation. This optimisation is a big part of the game for people who like to theorycraft, and if you're not one of those people, you can do one of two things: take on what these people find out and use it to improve your toon and playtime, or stick your head in the sand and be a unique snowflake who's doing less damage than they're capable of


How exactly can you prove this though? Even if you tested every possible spec, theres still gear to take into account. Different gear for different specs. The combinations are endless. And theres always the players skill and style. And of course the situation itself. Theres too many variables to make a scientific statement like that.

I mean for gods sake, every hit is a roll of the dice, literaly. Are you gonna sit here and tell me that there is a best way of rolling dice? What if I am just that lucky, that every hit I do is a crit? WoW is not a science because there is chance involved.

And even if it were a fact that one exact single spec is "the best", how much fun is that? That seems alot more like work to me. "Use this formula over and over without thought" "Use this exact spec or be called Fail"
That kind of attitude is whats sucking all the fun out the game. And having fun is the entire point of playing any game.
There are more important things in the game than being top dps all the time. And I think we can all agree theres nothing more annoying than that dps'r that after every pull wants to see the charts. All that matters in the end is that you kill the boss, get your lootz, and have fun doing it.
Bundy
 
Posts: 171
Joined: Sun May 31, 2009 2:41 am

Re: [Warrior] Levelling a new Warrior

Postby Bundy » Sun Jul 12, 2009 2:51 am

Anyways back on topic.
Bundy
 
Posts: 171
Joined: Sun May 31, 2009 2:41 am

Re: [Warrior] Levelling a new Warrior

Postby Epimer » Sun Jul 12, 2009 3:15 am

Bundy wrote:How exactly can you prove this though? Even if you tested every possible spec, theres still gear to take into account. Different gear for different specs. The combinations are endless. And theres always the players skill and style. And of course the situation itself. Theres too many variables to make a scientific statement like that.


...I don't even know where to begin with that. Tools like Rawr and DPS spreadsheets exist because WoW is predictable. The mechanics of spells and abilities are outlined in their tooltips, and where there are ambiguities in the wording, testing by players (analysing WWS reports, spending time at a training dummy whacking away) clarifies how those abilities work. This testing provides a model of how each ability works and how each stat affects that ability, and it therefore becomes possible to do things like provide relative stat weightings, or fix a numerical DPS value to each glyph (or profession, or talent point). These spreadsheets and other tools take into account things like your gear, and your stats, which is why they work! For an excellent example of this type of testing and analysis, see Theck's MATLAB thread in this very forum.

I mean for gods sake, every hit is a roll of the dice, literaly. Are you gonna sit here and tell me that there is a best way of rolling dice? What if I am just that lucky, that every hit I do is a crit? WoW is not a science because there is chance involved.


Every hit is a roll of a dice, but the outcomes available are determined by your set of gear, abilities, talents and glyphs. There's chance involved with each individual roll, yes, but roll a six-sided dice an infinite number of times and you're going to get an exactly equal number of 1s, 2s, 3s, 4s, 5s and 6s. The testing and analysis is done by working off averages and probability (actually some sims roll the dice for each calculation required - I know EnhSim works this way), which is how you would carry out any statistical analysis.

And even if it were a fact that one exact single spec is "the best", how much fun is that? That seems alot more like work to me. "Use this formula over and over without thought" "Use this exact spec or be called Fail"
That kind of attitude is whats sucking all the fun out the game. And having fun is the entire point of playing any game.


See my previous reply; I'm not suggesting you crack out the calculator yourself. There are people who do enjoy examining the game in this kind of detail, and that's fun to them. I'm not really one of them, by the way - my maths skills are far too lacking - but I do appreciate that I can benefit from their input and use what they find to improve my toon.

There are more important things in the game than being top dps all the time.


Yes, but in any kind of discussion around "which of x and y does more dps?", that's kind of the point :P
User avatar
Epimer
 
Posts: 1496
Joined: Wed Jun 18, 2008 6:54 am

Re: [Warrior] Levelling a new Warrior

Postby Elsie » Sun Jul 12, 2009 4:08 am

Armor Penatration is the same as Armor itself in a way. Its more uselful the more Armor the mob has. Stacking armor against a mob that hits fast and not so hard is not a good plan. And stacking Armor Pen againt a mob that doesnt have alot of armor to begin with is also not a good plan. Its all percentage based, 40% of 4000 is 1600 armor reduction, where 40% of 30,000 is 12,000. The more armor the mob has, the more effective your one stat is gonna be. AP always works the same though.

Umm.. a mob's attack speed has nothing to do with its armor, in general. In fact, there are typically only 3 different types of mobs, but generally in raids there are only 1-2 different boss armor amounts that matter (and they typically do not change from raid to raid).

At any rate, the boss's armor has little effect on the damage boost you gain from armor penetration. Armor pen does not work the way you think it does. Armor penetration reduces "Up to x%" of an opponent's armor and they quite mean this. While I do not wish to delve into the gritty math, suffice to say that the benefit of Armor Penetration is reduced as the opponent's base armor is reduced to where the percentage damage gain is equivalent if you're fighting something in 'cloth' armor or 'plate' armor. But, at any rate, there are only 1-2 armor values for all bosses in raids (possibly a third, but I don't recall ever seeing a boss with lower-than-normal armor). There is an armor value used for all bosses, and a higher one for more "warrior-like" bosses. The big historical example of the latter was Brutallus with ~8800 armor in TBC where-as every other boss generally had ~6600 armor (from memory). Of course, RoS phase 1 had 0 armor, but that's a rather special case.

How exactly can you prove this though? Even if you tested every possible spec, theres still gear to take into account. Different gear for different specs. The combinations are endless. And theres always the players skill and style. And of course the situation itself. Theres too many variables to make a scientific statement like that.

Actually, it's fairly easy to do this. Simply put, values on gear are not required for the formulas to work. Skills scale at a certain rate with certain stats, and stats scale with each other for determining DPS. Thus, you need only know how your DPS scales with each stat to create a formula. After this, you're essentially left with a formula with A, B, C, D, and so on damage stats such as strength, ArP, etc. You have constants such as the scalars skills use (i.e. X damage +y% AP), enemy armor (always a constant given the enemy), and so on. Obviously, more strength makes crit better, and vice versa, so don't assume this is simply addition because there's quite a lot of multiplication going on. This may sound complicated, but I assure you it's rather simple to piece together.

The combinations for gear and spec are, quite simply, finite. It's also true that if an ability gives you x% damage, and another y%, that the first ability is better in all cases so long as x > y. Thus, spec is usually a relatively easy affair. Gear is simply plugging in values obtainable. Note that values on gear, however, do not effect the benefit from either. If you strictly gain more damage from one stat than another then no value on gear will change this. An easy example of this is Attack Power vs. Strength. Simply put, Strength is always better than Attack Power in raids because you gain more AP from Strength (due to stance or kings) than from Attack Power given a 1:2 relative cost-per-point ratio. Also, even though some stats scale with each other (i.e. More Crit = Str does more) the ratio can be such that there is no obtainable amount of, say, Stat A or Stat B that would make Stat B better.

So now you can see it's a simple matter of using the formulas, plugging in values, and popping in the parameters of the fight. This last part is key, as you define the number of mobs, time in execute range, available raid buffs, bloodlust used or not, etc etc etc. Then you are given a theoretical number for the average DPS one can expect for a given amount of time. Obviously this is an average. This should be expected though, as you're dealing with probabilities. For example, a 30% chance to crit does not mean you will crit 30% of the time in a given interval. However, it does mean that as time goes on you will average 30% of your swings as crits. Thus, some variance between real results and predicted results are not at all unexpected given probabilities and latency. This does not mean that the average obtained is without merit. Anyone can RNG high or low in a given span now and then. It can conclusively prove, for the parameters, that one option will clearly do better on average than the other.

Now, it's kind of you to bring playstyle into this. There is nothing that can control playstyle. If someone is bad at what will perform better on average, then they might do better with something that is not as good on average but that they can still perform well at. For example, if Spec A is overall better than spec B by 10%, and a player is 20% worse than optimal at Spec A but optimal at Spec B, then they will do better with Spec B. Obviously though, this is absolutely rediculous to set the standard for how to give advice. If someone asks what is better, they are best told what is better. One is flat out better than the other and is provably so. Otherwise we wouldn't have spreadsheets. If their skill is not up to par at this, then they can decide for themselves how to best approach a second option and, when they ask how to improve at the other option, they're welcome to use any existing advice and data to their advantage (and obviously such data was done in the case of warriors).




Now, as for leveling a warrior. There's no optimal for the entire time as your available gear and specs change often. In general, I'd think Fury is easier to level with because there's less to manage. However, Arms is probably easier to gear for - especially since you only need 1 good two-handed weapon the whole time. I find it can be a bit boring though since I don't like waiting around for my swing timer.

As a practical approach, if something comes from a higher level quest or instance it's better. This is because the item level is better and, quite frankly, item level tends to mean nearly everything. Obviously, certain items break this rule. This is especially true if you have a little pocket change. The Burning Crusade (and WotLK) has a lot of items that have inflated amounts of stamina so generally you want to look at DPS stats where bigger numbers = better. You also want to stick to plate for leveling for obvious reasons. Finally, some items from TBC are rediculously good simply because they have sockets. You can get +12 (or +14) gems incredibly cheap - and if not, simply buy a couple uncut WotLK green quality gems and tip a Jewelcrafter a couple gold to make them for you. Off the top of my head, I recall using Necklace of the Deep on my DK for a long time because I put two +14 STR gems in.

That all said, I would keep stamina heavy items around for quests that required killing single mobs such as small number group quests. The higher stamina basically gives you a higher time interval in which to kill the guy. For example, 10 seconds of 500 dps is better than 8 seconds of 600 dps. For typical gather-type quests though, you're simply better off killing stuff as fast as possible. Alternatively, you can also swap in the "stamina set" for health regen if you have the talent.
User avatar
Elsie
 
Posts: 3819
Joined: Sat Jan 12, 2008 11:12 pm

Re: [Warrior] Levelling a new Warrior

Postby Bundy » Sun Jul 12, 2009 10:39 pm

Every hit is a roll of a dice, but the outcomes available are determined by your set of gear, abilities, talents and glyphs. There's chance involved with each individual roll, yes, but roll a six-sided dice an infinite number of times and you're going to get an exactly equal number of 1s, 2s, 3s, 4s, 5s and 6s. The testing and analysis is done by working off averages and probability


The outcomes available are almost always the same, gear, glyphs, talents only increases the probability of any given outcome.

And could we please not use imaginary numbers here when trying to prove something as fact. Time should always be a factor in any WoW equation. A roll of the dice is a roll of the dice. You said it yourself, just because you have a 30% chance to crit doesnt mean your gonna crit 3/10 times. Ive had fights where I crit damn near every hit, and others where I couldnt get a crit to save my life. Obviously this effected my dps greatly, and a different spec would not of helped one bit.

Now if I had one of those fights where I crit damn near every hit, my dps is gonna be higher than yours even if my spec is 10% less optimal than yours. So if I out dps you in one fight, whos to say I cant do it again and again. I know that would be nothing but luck, but thats the point.

Your probably thinking that if I was the best spec to begin with I would have done even better anyways.
Well the less optimal spec happens to increase my chance to crit where as the optimal does not.
What hits harder.... a HS crit, or a imp/glyphed non crit SD execute? Im not sure but id say there pretty close to even on damage done. Except one uses a hell of alot more rage than the other.

SD itself relies on chance. What is it 9% chance to proc? There is no garuntee SD will even proc during a fight, its a chance.

Advice is advice, take it with a grain of salt. You dont have to be an expert at math and have done extensive testing to give advice. Its up to the person taking the advice to sort through the good and the bad.
Bundy
 
Posts: 171
Joined: Sun May 31, 2009 2:41 am

Re: [Warrior] Levelling a new Warrior

Postby fafhrd » Sun Jul 12, 2009 11:46 pm

:S

Are you familiar at all with how statistics and mathematical modelling are used Bundy? The type of thing used to compare talents and abilities in WoW are in principle exactly the same as those used to decide what metal to build a space shuttle out of, or predict whether it's going to be sunny or rainy tommorow. By your reasoning the past 200 years of industrial and economic and social development are based on "using imaginary numbers when trying to prove something as fact" - yes, those imaginary numbers are what decisions are based on all the time.

Not trying to convince you play your warrior differently, just arguing with you about things you seem to have wrong.

Now if I had one of those fights where I crit damn near every hit, my dps is gonna be higher than yours even if my spec is 10% less optimal than yours. So if I out dps you in one fight, whos to say I cant do it again and again. I know that would be nothing but luck, but thats the point.


Seriously? The answer is you're not likely to do it again again unless the math is on your side - the math that elsie and everyone else has been bringing up. No one can say for sure what will happen on any given attempt, but people can damn well say with a great deal of accuracy what is likely to happen. If you make an accurate statistical model, it will predict what on average will happen very accurately. If you want to outdps someone again and again, your best bet is to do what the math predicts is better on average, not do something dumb and hope you get lucky again and again (because you can't).

Your probably thinking that if I was the best spec to begin with I would have done even better anyways.
Well the less optimal spec happens to increase my chance to crit where as the optimal does not.
What hits harder.... a HS crit, or a imp/glyphed non crit SD execute? Im not sure but id say there pretty close to even on damage done. Except one uses a hell of alot more rage than the other.


You really should read the tooltips on the abilities you're talking about at least. You know those huge HS crits? Critting for 6k+? Well guess what, if you hadn't heroic stuck, you'd have crit for 5k anyway, not spent 2-12 rage on it, and generated 30-40 rage more from the white hit landing. You know why? Because heroic strike isn't an extra attack, it's a replacement of your next mainhand attack (which is your only rage generating attack as arms). All the heroic strike is doing damage-wise is adding 495 damage to your regular melee swing, and preventing it from glancing.

Sudden death on the other hand, is an actual extra attack. When you see it critting for 6-9k, it's an actual extra 6-9k damage, not just an extra 1k damage. And it absolutely does NOT take "a hell of alot more rage", because like I said in my first post about that, the rage cost of a heroic strike isn't the paltry 12 rage for pressing the HS button, it's that 12 plus the 30-40 rage you would have generated from the white attack you're replacing with a heroic strike.

You don't need any math beyond kindergarten to figure whether heroic strike or sudden death execute is a better source of damage as arms.
ImageImage
1/1 Lore pre-nerf.
User avatar
fafhrd
 
Posts: 5432
Joined: Fri Aug 31, 2007 2:31 pm

Re: [Warrior] Levelling a new Warrior

Postby Bundy » Mon Jul 13, 2009 8:42 am

Seriously? The answer is you're not likely to do it again again unless the math is on your side - the math that elsie and everyone else has been bringing up. No one can say for sure what will happen on any given attempt, but people can damn well say with a great deal of accuracy what is likely to happen. If you make an accurate statistical model, it will predict what on average will happen very accurately. If you want to outdps someone again and again, your best bet is to do what the math predicts is better on average, not do something dumb and hope you get lucky again and again


I just wanted someone to admit luck is a part of the game, so thank you. I dont rely on luck, im just saying that if if Spec B has the potential to beat Spec A in any given situation, you cannot say Spec A is better flat out. You can talk all day about what supposed to happen, what should happen, and whats likely to happen. What actualy happens is all that matters.

You really should read the tooltips on the abilities you're talking about at least. You know those huge HS crits? Critting for 6k+? Well guess what, if you hadn't heroic stuck, you'd have crit for 5k anyway, not spent 2-12 rage on it, and generated 30-40 rage more from the white hit landing. You know why? Because heroic strike isn't an extra attack, it's a replacement of your next mainhand attack (which is your only rage generating attack as arms). All the heroic strike is doing damage-wise is adding 495 damage to your regular melee swing, and preventing it from glancing.

Sudden death on the other hand, is an actual extra attack. When you see it critting for 6-9k, it's an actual extra 6-9k damage, not just an extra 1k damage. And it absolutely does NOT take "a hell of alot more rage", because like I said in my first post about that, the rage cost of a heroic strike isn't the paltry 12 rage for pressing the HS button, it's that 12 plus the 30-40 rage you would have generated from the white attack you're replacing with a heroic strike.

You don't need any math beyond kindergarten to figure whether heroic strike or sudden death execute is a better source of damage as arms.


I know what the tooltips say and I know HS takes the place of your normal swing, which is why I wouldnt sapm it. I never said HS was a better source of damage than Execute, I know that its not. But improving HS greatly seemed better to me than improving execute a tiny bit. A 6k crit HS that costs 2 rage, is better than a 4-5k reg hit execute that costs 25 rage.
15% is pretty good to me, thats like 55%-60% crit raid buffed, damn good odds. More crits = more bleeds, + 10% damage.

Anyways im still trying to out and nothing is final, but honestly it seems damn nice so far. Ive had no rage issues and it really does crit most of the time.
Bundy
 
Posts: 171
Joined: Sun May 31, 2009 2:41 am

PreviousNext

Return to Off-specs & Other Classes

Who is online

Users browsing this forum: No registered users and 1 guest

Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 380 on Tue Oct 14, 2008 6:28 pm

Users browsing this forum: No registered users and 1 guest