Deadbox

I was under the assumption that was stated in reference to TC based on level 80, since we have no reliable way of knowing what gear stats and itemization will look like, however my question is in regards to gear and values we already have currently. So once again I pose the same question.

 

◊ Vontre

Did you click the link directly above you? Here let me help quote the first few words for you.

"Vontre's Magegraf
World of Warcraft version 3.0 at level 70"

www.magegraf.com

Raiding is full of challenge. Sometimes there is fire. You have to not be in the fire.

"We agree with Communism." - Greg Street 2009

 

Lhivera

No, I'm just not doing any at all until we see a new patch. I think Manly's correct that too much is speculative at this point; we've already pretty thoroughly analyzed what the individual talents and spells can do with the info we have now, and I don't think there's much point in putting any real time into more complex theorycrafting of whole specs at least until we see the next iteration of talents.

Even then, I doubt I'll put time into level 70 specs, since they'll have such a short period of relevance.

ETA: And besides, with Vontre's tool doing its thing, the demand is much lower

At Veridian Dynamics, we can even make radishes so spicy, people can't eat them. But we're not, because people can't eat them.

 

Muphrid

Well you know what my solution would be: lock in the m/r values, lock in the 1/b values, and convert stats to exponential terms so differences from talents would not cause varying effects at different gear levels. Anything else is just an approximate bandaid.

 

Roywyn

Flamestrike is *BEEP* because its scaling and its cap are terrible.
One reason is the DoT eating up scaling/cap which is wasted if you spam FS.

One is an instant cast 10 yard radius PBAoE with 21% scaling (392 base) and 10.1k cap.
The other is a 3s cast 5 yard radius targeted AoE with 23% scaling (532 base) plus 4 ticks over 8 secs with 3% scaling (106 base) and 7.8k cap.
Seed is a 2s cast mob-targeted 15 yards AoE with 16% scaling (pretty low), 1200 base damage (pretty high) and 13.6k cap.
(Assuming Seed's DoT doesn't tick at all.)

That means with 262% crits (that Burnout doesn't seem to give), 100% crit (which we can't guarantee) and World in Flames extending the cap (which no other talents do so far), Flamestrike spam just reaches the DPS of Seed spam with 0% crit.
Ignoring that Seed covers a slighly (800%) bigger area.

That's not a whine post, I know that AE spam and Seed spam are comparable.
But Flamestrike will need some tweaking (coefficients and cap; and then argue about the area), or it'll be labeled Lamestrike for a long while.


For the question list:
* Do the new ranks of mage armour affect all magic-dispellable debuffs, or all non-physical spell debuffs (like Cloak of Shadows affects them)?

It's most likely the first (the second would probably be too stong) but wording has been fishy with effects like these before.

The Blue Bar and you - the complete Fire Mage 2.4 mana compendium: http://elitistjerks.com/658230-post3191.html

DPS spec and class comparison in Naxxramas gear: http://code.google.com/p/simulationc...i/SampleOutput

And [Timbal's Focusing Crystal] doesn't proc on AM.
Neither does [The Egg of Mortal Essence] since 3.1.

 

Lhivera

Hm, I dunno. I mean, yes, those certainly reduce its scaling, but using my own stats and raid buffs, if I were fire spec, my DPS with Flamestrike spam would be about 98% what it would be with Arcane Explosion against as many as 8 targets, with much higher efficiency and at a nice safe distance. And that's now, with current talents at level 70, not including WIF and Hot Streak. That doesn't really seem terrible to me.

Now, yes, caps are a problem, but that's not a Flamestrike-specific problem; they're just plain implemented badly.

Got your question, will include it in next update.

At Veridian Dynamics, we can even make radishes so spicy, people can't eat them. But we're not, because people can't eat them.

 

Zaldinar

With a side note that the DoT is not capped (or atleast, wasn't last time I tested it), and has a convenient duration that matches Blizzards channel time, up to a certain point of haste you can rotate Flamestrike/Blizz from a safe distance (at less threat efficiency than AE, but you're safe from point blank concerns) without stomping on your DoT.

That point of haste being:
(3 + 8) / (1 + X) = 8
11 / (1 + X) = 8
11 = 8X + 8
8X = 3
X = 3 / 8 = 0.375 = 37.5%

To truly model the game, we first must research it.
http://forums.worldofwarcraft.com/th...icId=109841969
Proven TheoryCrafting Stuff, chain casting in a PTR near you soon.

 

◊ manly

Well, when I proposed the move, there was more than this involved. Personally I don't mind if people are willing to spend their time and work out all the possible outcomes of every possible talents and what is best and whatnot. It's awesome if someone is willing to do it. But the problem is that many people read those, and take them as absolutes. Now it doesn't become theorycraft stuff, it becomes whines, and QQs, and nerf threads, and countless other stuff that really have no reason to exist.

Truth is, if you look at how Water Elemental originally was planned to be pre TBC, and how it turned out, you can see just how far from the 'truth' any original speculations were - to the point where its so far apart you wonder how it ended up so much disjuncted. I view the current talents as something extremely close to that.

I think as players we end up either on the side of 'blizzard will nerf mages because its always been that way' or 'blizzard knows what they're doing, just wait and see'. I'm more of the latter opinion. They know they messed up. We know our vision of mages differs from what blizzard envisions, as such, we won't always agree with the results.

I am still impartial and undecided as to what my plan is when we have more conclusive data on wotlk talents. Will it be a unique opportunity to start a vast analysis and come up with cumbersome evidence or what works and what doesn't, in the hopes that blizzard changes the class the way we want it ? What I mean by that is that, if the numbers are better than my own expectations, if I chose to analyse the results, it would be my duty to report to blizzard that they need to scale us down. Most players just see it as an opportunity to whine until it bends, I always disagreed with the practice. Do we want to try and this time have a true impact on the class, be it positive or negative, or is this just a foolish idea ? I'm warning everyone in advance, if we end up in a massive whinefest, its not gonna fly. I have no interest in reading anyones QQ or whines, not even my own. Its rarely productive.

 

Arazan

Well manly the problem I have is that people who read something they don't want to recognize will just decredit it as "whining". I remember all that BS I had to deal with when I was doing my WWS analasis thing on the wow forums... where I was just trying to show how mages are being pressed out of the top dps role by the three other pure dps classes, and at least half of the responses that weren't supportive of what I was doing (whether they agreed with me or not) basically boiled down to the mindless "LOL QQ MOAR L2P!" crap that plagues the wow forums.

My point is that if there's something really wrong, if mages are going to be in the exact same position via TC on the next iteration of talents, or if they're going to be blatantly overpowered because of a stupid mechanic like the exponential stacking returns of WG, you can't really be too concerned about whether or not people just see it as whining and nothing more. If people don't want to hear it or want to be dicks about it they'll just discredit it as whining. While I totally understand what you mean by avoiding a massive whinefest, you've gotta see that some people will see anything as a whinefest... especially the people who are happy with their position in the game relative to the mage class.

Because of what was said at the WWI abotu re-establishing the mage class as a true dps class, I'm slightly optimistic, though obviously wary because it easily could all be lip service. What I want the most is for one of the lead developers to release an unambiguous definition of what the mage class is supposed to be. I think that's the biggest disconnect, because to me a mage should be the class that deals unrivaled dps if they're left alone, but must be constantly wary because they have little to no ability to withstand damage. Honestly, I'd like to see Ice Block removed because I don't think it fits the style that I'd see the mage class as, and I'd like to see stuff like Blink be buffed so that the mage is more about being able to get away and not being able to sit in a bubble saying "neiner neiner neiner...".

Sorry getting a little off-point. My point is that what I'd like most is for a developer to come out and give a clear, concise definition for what the mage class is supposed to be. If that doesn't fit what I want anymore, then sobeit... but all we really have to go by now are random ambiguous things like "kings of aoe" and "experimental class". I can't really argue for buffs to our direct damage capabilities because I have no idea if they actually want mages to be that good with direct damage. Maybe we're not supposed to be primary nukers anymore, maybe support/CC/aoe is what they want the mage class to be.

Also, another thing that I don't think anyone has pointed out yet is that mage coefficients and capabilities are meaningless in a vacuum. We can do some minor TC with warlock stuff to compare, but really what we need to make a meaningful comparison would be rogue and hunter TC, because if I'm right then it could very well turn out that mages in WoLK are on par with warlocks, but rogues, hunters, and possibly even fury warriors and enhancement shamans pull way ahead of casters. Unfortunately, it's incredibly damn hard to TC for enhancement shamans, rogues, and fury warriors, hell actually hunters too, simply because their dps changes greatly depending on the weapons used.

 

PSGarak

Running spreadsheets at this point is rather pointless, which is to say fun and possibly rewarding but the information will be obsoleted long before it matters. However, given that bliz reads this forum somewhat, we do have some input on the class design, and I might even go so far as to say we have a responsibility to provide feedback in some capacity. And by refining what that "some capacity" is, I think gives us a basis for how we should be approaching the talents and how they might change before release.

The first is theorycrafting. There's only very scant evidence*, but it seems that blizzard's internal balance stuff is based on testing rather than theorycraft. We can raise red flags about scaling issues that may not be apparent at the gear levels they test things at. Especially since we can figure out the optimal gearing for a spec to be tested at, rather than what a template character might do. An extension of this is the gap between vision and implimentation: my prime examples of this are Molten Core and Torture from the warlock trees, which obviously intend for fire-shadow synergy but simply don't work well enough to overcome DS. I suppose your equivalents would be how damage caps interact with World in Flames, and Burnout vs Spellpower.

The second is Linus' Law ("with enough eyes, all bugs are shallow") applied to specs. Letting them know all the crazy-ass builds and chickenhead playstyles we can come up with so that they can be balanced around. If we take it for a given that Demonic Sacrifice wasn't tested in a destruction build, wouldn't it be nice to at least have that talent interaction pointed out beforehand? Or arcane missles being non-viable, or arcane blast being viable, or fire being indefinitely-sustainable. Or some bass-ackwards frostfire bolt build getting 450% crits or whatever. This also applies on the higher level of bloat, and venue-specific talent distribution occur throughout the tree, making some abilities basically pointless.

So yeah, TCing out to find that deep frost outdoes fire/arcane by 2% at 2400 damage and 32% crit is not just useless, but it contributes to whining that drowns out discussion. On the other hand, TCing out WG's effect on mage DPS is useful, as well as hot streak and burnout because it's the sort of information that helps determine if a talent is performing its intended function and scaling.


[edit: forgot my footnote]
* The "jaw-dropping" comment is one, that shows they were either only testing at one gear range or failed to notice DS. The other evidence that really seals the deal for me is when Elemental was overhauled with changes to Lightning Mastery (cast time reduction reduced from 1s to 0.5s) and Lighting Overload (5% chance of extra bolt changed to 20% chance of half a bolt). Theorycraft said that the two changes together were an overall nerf, blizzard said that testing revealed they were an overall buff. It turned out on the PTR the half-damage bolt only halved the base damage, not the gear contribution, which explained the descrepancy but was Not Working As Intended and got fixed.

http://sig.gamerdna.com/quizzes/INFL...erks/Garak.png

 

Searix

On Incanter's Absorbtion:

Assumptions:
Sunwell is the current and future mindset of raid developers, so we'll use the fights as models.
Your dps can and does make or break killing the boss (Barring Kalecgos and Felmyst)
One discipline priest with the availability to cast any of his talents on you BARRING Power Infusion (aka divine aegis). This priest has just enough time to cast one GHeal on you for 7k when you have heroism and pop your cooldowns.
Incanter's Absorbtion does not stack on itself, you can only eat one large blast to make it work.
3.0.0 comes out really early and you are trying to decide if incanter's absorbtion is worth it or not, so you still are level 70 with 61 talent points.


So is it any good? Let's find out
Let's see what spec (fire/frost) it would work best with, its essentially MM spec with 3 extra points.

At lvl 70 it's not worth it (over 200 dps loss, can't imagine it would make up for it), at lvl 80 it might work but for the sake of arguement let's look at lvl 70 versions of the trees so Rawr wont cry sweet tears.

That leaves going frost, or something with 2 peice t5.

Best frost and arcane build i can find: War Pirate :: Mage WotLK Alpha Talent Tree

I'm going to test it pure frost spamming, and 2 peice t5 arcane blast spamming during cooldowns.


Now quickly to model Incanter's Absorbtion. I'm going to test it using the Major Protection Potion of the respective school, Mana Shield, Fire Ward/Frost Ward where i can, in addition to Divine Aegis (which i model at as a 2100 shield at level 70).

Modelling the Sunwell fights:
Brutallus, use on 3rd meteor slash, hits for 8.7k before absorb. I'm figuring with Fire Ward i can absorb ~8k of that. That translates into a bonus 1200 damage incanter's
Twins, use on a sear (4800 damage). 640 damage incanter's.
Entropius p2, pop mage with frost nova when spellstealing buff, eat an instant cast fireball for 9k (Living on the wild side, i do this anyways), again 1200 damage incanter's.
Kil'Jaeden eat a shadow spike, 6k damage. 900 damage incanter's.

I'm currently modeled at 2400 dps using 40/0/20.
I'm going to model this as an activateable trinket (adding the bonus damage on to the on use) as a 2 minute timer trinket with 10 second duration.

Off to rawr! (Again, im testing this without knowing the result).

Modeling this as a 1200 damage trinket, incanter's absorbtion adds 210 dps (Note: factored in costing 2 globals to fire ward/mana shield).
Modeling this as a 900 damage trinket, adds 160 dps.
Modeling this as a 640 damage trinket, adds 100 dps.

Using 2 peice t5 and spamming arcane blast vs. frostbolt during you popping "all your cooldowns" adds roughly the same dps.

On a final note: I factored in the 2 globals needed to mana shield and fire ward as taking 2.5 seconds. Quick math shows a ~40 dps hit over the entirety of a fight. This has already been subtracted from the above totals.
Final final note: Did not factor in potent spirit, Presence of Mind is unmodeled in Rawr

Interesting stuff

 

Densor

You're assuming the use of absorption shields only when cooldowns are up. That greatly increases the effect of IA at those time, making almost any absorption effect useful, but it neglects the random shielding you'd benefit from during a fight.

Priests could also be using PW:S more on fights. A holy priest's PW:S is a more reliable source of damage absorption than a disc priest's GHeal 7 crit, and provides about 2400 absorption at 2.7k +heal. On fights with scheduled burst damage, a shield wouldn't be a bad idea for priests to be tossing on mages with IA. It would be less proactive healing than a renew, but unless things are tight, it wouldn't be too big a deal, especially if the shield is better for DPS. And since you can get a shield every 15 seconds, you could reliably have 360 extra spell damage for 60+% of the fight if 2400+ damage bursts happen frequently. And if damage bursts happen less often, but in larger amounts, Major Protection potions would result in being better than Destruction Pots, as they average 540 extra spell damage.

For wards, is 240 spell damage for 10 seconds worth a GCD? During fights where there is downtime, like Najentus, it would be worth it, for sure. In other cases, I don't believe so. 240 spell damage translates to like 364 extra damage per frostbolt (using 1600 spell damage and 35% crit), while a frostbolt would be averaging 2209 damage over the GCD required by the ward. So, you'd need to get off just over 6 bolts in 10 seconds, requiring over 50% haste. Maybe when PoM is up, since that cuts your required haste by a bit.

Mana shield is worse than wards. Might be worth it when cooldowns are up, if you have time before the damage hits and can spare the mana.

That's all for level 70 builds (which would be giving up cold snap and frost channeling/arctic reach from the standard 40/0/21 build). At level 80, the increase in the wards' protection will be over a 50% gain. That should push things in favor of warding whenever possible. Mana shield still looks like it will suck.

 

Gediablo

I'm pretty sure a 33/?/? fire spec still be safer and do more damage spamming AE than deep fire AoE. Seems like 33/?/? is more mana efficient and even higher single target damage - both sustained and controlled burst.

 

Skallewag

Dont forget the reduced threat. The true damagecap of many AoE fights is aggro. Also if we imagine a lv 80 version of mount hyjal or any extended AoE fight there will be times when shit happens. Tank dies or you just misjudged threat a bit, in those cases arcane mages will have a 1 sec invis. Not instant aggrowipe but with tools like blink and frostnova one seccond is easy to buy. While Prismatic cloak only requires 25 points in arcane it shares tier with both arcane instability and arcane potency, thats about 6% crit you haveto start ignoring if you want prismatic cloak as early as possible, not to mention that if youre far enoughe up in the arcane tree to start getting arcane instability you would be mad not to pick up AP as well. Long story short it looks like AE spam will be our best bet for WotLK. The only scenario arcane AoE would get beat in is Living bomb + firetalents being able to massively outdamage AE and threat at the same time not being an issue.

Another interesting thing I kinda hope for is the new mage armor, magic absorbtion and incnaters absorbtion making mages a good choise for magic tanking. We get 120 resistance to all schools above what gear can do and when we do resist stuff we proc mana that can be used for a bit of shielding that boosts our damage. Not a major deal I willingly admit but I have always been fond of mage tanking in any form and I kinda like this potential new concept for the arcane tree of absorbing and dodging magic. It feels very fitting.

 

Zeldyrr

Given the debate over how much Blizzard plans and "does the math" I thought this quote (via mmo-champoin) was revealing.

Emphasis mine. This is pretty much how I imagine Blizzard liked to operate. They don't do hardcore math on things like Winter's Grasp before beta. They think of a cool idea that accomplishes a design goal (in this case, make frost mages be able to use talents related to being frozen on bosses) and stick in some numbers that seem reasonable. Release it to beta, let people give feedback, "watch numbers", and then tweak as needed.

This is how something like DS-destro builds slipped through. It just never came up in a beta/PTR cycle.

 

◊ manly

Oh sweet so they totally ignore scaling, or better yet, we can't test it.

 

Muphrid

That doesn't surprise me. There simply isn't the mathematical foundation in the mechanics to theorycraft class balance to arbitrary gear levels. Balance that holds at one gear level seldom implies balance at another sufficiently distant gear level. If Blizzard devs were doing theorycraft instead of tests, they would see immediately how poorly designed so many of their mechanics are and how much, for example, proportional scaling would fix that.

Not to say proportional scaling is perfect. There does seem to be a nontrivial tradeoff between mathematical simplicity and game complexity, but there is no doubt in my mind that applying proportional scaling to very simple and clear-cut mechanics like +spell damage and attack power would greatly simplify the job of theorycrafting balance at very little cost of complexity.

 

Skallewag

What really suprises me is why they dont take advantage of the fact that 10 million people are playing and loving this game. They could have so many testers and theorycrafters working for them worldwide entierly for free. Thousands of hours of work people would perform gladly since the reward of having actual devs read everything you write would be reward enoughe. Or am I way off *smiles at Lhivera* Wouldnt you enjoy having devs read and consider your calculations rather than getting forum bans as thanks? If they spent a bit of time to wash up some good number crunchers and some devoted hardcore pvp and pve people they would be able to test both gameplay and estimate scaling. The chanses of being notified about certain spec and item combos they didnt intend and dont want would improve. Not that I think any of that is gonna happen but all the same its a valuable resource they could have for free.

 

◊ manly

Well, what I meant to say with my comment is that all the gear available in wotlk should be available for testing. balance dps around the best in slots rather than entry level gear. Somehow I don't expect that to happen, particularly since it would have to cover gear that doesn't exists.

 

Lhivera

Anyone noticed that some sites are reporting a version of the Living Bomb description text as including a knockback effect on the detonation? Is this a change that was spotted at WWI?

At Veridian Dynamics, we can even make radishes so spicy, people can't eat them. But we're not, because people can't eat them.

 

PSGarak

Skallewag: Then you need someone to read through ten million submissions, and identify the three good ones in the deluge of crap. Getting any sort of meaningful results requires an acceptable signal-to-noise ratio. The official forums, the WoW community, and humanity in general don't have that. EJ does, because they have a strict policy of permabanning retards and deleting shit posts. Sorting through the general forums for the diamond in the rough would likely require more skill and labor than actually theorycrafting it for themselves. The obvious solution, of course, is to come to EJ instead of the general forums for TC analysis, and to listen to the general forums only for the opinions of the playerbase (QQ included). But not publically.

http://sig.gamerdna.com/quizzes/INFL...erks/Garak.png

 

Actovision

Hah, a little from column A, a little from column B.

They've done it in the past, and probably still continue to do it. Some of the more prominent members have a hotline to the devs when needed, as well.

 

Skallewag

Um... no you dont need to delve through 10.000.000 pages of crap to find the nuggets in there. A large part of the player base would be completely unable to contribute to the game development. But it would be far from impossible to find the right kind of players to put to work. For example they allready know that Lhivera is a good theory crafter. All blizz would really need to find more is one single emplyee whos apt at tossing math around and has good insight in the current and planned game mechanics. Such a person could spend time looking for other good theory crafters. In just a couple of hours reading on these forums for example you would be able to find 10 people who know how to juggle numbers (and who do so wether asked to or not). Thats an afternoon spent by one person that could be turned into hundreds of hours of valuable theorycrafting time. How about asuming that blizz would give five people two months to scour the internet for enthusiastic gamers who love to wrestle with numbers and tend to make good estimations from the numbers they have to work with. Doenst that seem like a potential goldmine in eager theorycrafting time they could get for free? Not to mention that it would be theorycrafting that could be done with way more accurate numbers than datamined scraps.

As for actual playtesting I really dont see a problem there either. Hardcore raiders and pvpers love new content. If you were to allow a bunch of nihilum-type players fiddle around in various planned raid isntances you would have very dedicated people who are used to figuring out the optimal way to beat content and to turn all talents and gearpices from every angle to make them opted.

Im not suggesting blizzard actually emply 10.000.000 people, but finding sharp minded good players and offering them the honour of weighting new content could be a potentially very powerfull developmenbt tool. Kinda like Torrent downloading. Throw a chunk of data out there and let lots of people nibble and learn both from what they find and from what others find.

 

PSGarak

Considering that we're already here theorycrafting in public, how much of a benefit is there to bringing us in-house? There's faster response between the development and the theorycrafting, but the people they bring on would be under NDAs, meaning that the community we have here would be thinned out and less productive. And that's assuming that said people contribute pro bono rather than as employees. Believe me, I'd love to be a paid theorycrafter. It would be a consumately awesome way to put my math degree to work =P. But what they really need is more TC- and math-inclined devs, not a dev team with some math bitches they go to after they've gotten something off the drawing board.

http://sig.gamerdna.com/quizzes/INFL...erks/Garak.png