Updated for 5.2
Yes, that habit.... break it with a sledge hammer..... or dynamite would be good if you have it. Not necessarily because Mr. Robot is terribad, but because it is always good to see a second opinion, if only for the fact that it makes you think about what you think you know a little harder and maybe come to a better understanding of it.
Had a few talent changes, mainly buffs to less popular ones, shockwave got a nerf on single targets. Shield slam and Revenge got their base damage increased +150% but their AP scaling reduced by ~10% it will make questing/PVP/dungeons nicer but above ~80K AP (so for raid bosses) it will be a nerf. Warriors now get 50% more haste from haste rating, so more white damage.
Thanks - on another matter, what's the basis for preferring the eternal meta-gem to the austere one? I've been following your advice but was surprised today to see Ask Mr Robot wanting me to switch to the austere one. I know it's hard to put stamina and parry on a common metric, but I guess it's the other parts of the metas that are more important in your recommendation. What does the maths say about comparing the the 1% BV and 2% armour for a typical prot warrior nowadays? (Sorry for asking an old question - I did a search but my google fu failed me; a link would be fine).
Also, am I right in thinking the 1% BV does less if you are using Sbar rather than if you are using SB?
Askmrrobot's default weights are shit for tanking. The only way to make that tool worthwhile is to put in your own weights. The creator hangs out on the MMOC forums and asks how he can improve all the time, but he hasn't done dick to make it something worth using with his defaults.
The Austere is the same old argument; Armour is always there, however in terms of TDR and spike prevention the Eternal is better unless you are exlucisvely using Sbar as your rage dump. 1%BV does nothing for Sbar, but really the only fights where you should be using Sbar over SB are Magic fights, where you would prefer the efflugent anyway
actually his stats are based on thecks latest work. the current stats weights (hit=exp > mastery > avoidance) on amr work quite well .tbh. It is true though before that it was making weird suggestions with strength gems preposterous things like that.
AMR prioritizes Stamina (and thus EHP) above all, and unfortunately that's just not the case as how much stamina you require depends on what level of content you're doing (25man HC requires more than 10 man Normal) whether you're 2 or 3 healing encounters, whether you 1 or 2 tanking encounters, if you're just dropping dead to mechanics, or if your healers are ooming, if you're good and can get an Sbar up for every breath, or if you're bad turn your back to the boss. You just don't need as much stamina as AMR suggests, unless you are really bad at active tanking, or are in greens.
You can comfortably clear Terrace with 1 stamina trinket and not a single solid stam gem. If you're not a dick on Lei Shi.
EDIT: AMRs default weights are far to "round" to be actual stat weights with any sort of math behind them. thye look more like arbitrary weights to fit a generalized priority.
EDIT2: and it also still defaults to being under the hit/exp caps slightly. Myself and Agg have argued against this before and Theck's most recent math seems to support it as it indicates that if you're going to go into hit/exp you go all the way as even a small deficiency opens you up to fatal spike. Missing one Firebolt may be annoying as DPS, missing a shield slam as prot can be deadly.
Last edited by Tengenstein; 02-22-2013 at 07:31 AM.
so if your hit capped and 7.5 + lets say 10% expertise you are already in that <0.01% part you don't have to go "all the way"Curiously, the “uncapped” control/balance set is an improvement over the regular control/balance set. This suggests that being exactly at or over cap isn’t as critical for warriors as it is for paladins, who saw a noticeable loss in this same comparison.
Last edited by gom; 02-22-2013 at 04:15 PM.
Don't Quote things out of context. THeck hates that, just like everyone else, If you read the whole article, or at least the bit pertinent the graph it does back up "all the way to cap"
so 7.5% hit 10% expertise is nowhere near as close to the caps as you need to remove spikes and having a less than 0.01% chance of taking a >90% health in 4 hits is a fairly high percentage. in a 5 minute fight you've got 144 opportunties to get take a 4hit string.In particular, what we’re really interested in is how much hit and expertise we really need to maintain Shield Block. We know that 6.667 is the theoretical limit, but we also know that it isn’t sufficient in practice because rage generation isn’t a nice, smooth, continuous thing. So let’s try and determine exactly how much hit and expertise it takes to push 90% spikes below an acceptable threshold – let’s say 0.001%. To do that, we’ll take the control/balance gear set and just hack away rating until we get to zero. Technically we’d be able to allocate that itemization somewhere else (say, mastery), but for this calculation we’ll pretend that we can’t, and that we’re simply losing itemization. Maybe it’s being shifted to Stamina, for example. The reason is that it will give us a better “worst-case” estimate on our hit and expertise needs. Since the 4-attack moving average seems to have the worst time with 90% spikes in the avoidance gear set, we’ll use that.
If we perform that calculation, it looks something like this:
Spike damage presence vs. combined hit and expertise percentage. The inset shows an expanded view of the results above 15% combined hit and expertise. The dotted line on the inset is 0.001%.
With no hit or expertise, our spike presence (the percentage of events exceeding 90% throughput) is a little over 3.5%. It drops fairly dramatically with hit and expertise up until about 10% on the x-axis, where it’s starting to slow down in effectiveness. The surprising part here is just how much hit and expertise it takes to eliminate those spikes. Even at 15% combined hit and expertise (5% hit, 10% expertise), about 0.03% of all events exceed the 90% spike threshold. While that’s certainly not a lot of spikes, it’s still well above our self-imposed 0.001% limit. It isn’t until around 21% combined hit and expertise (7% hit, 14% exp) that we consistently drop below 0.001% spike presence, and it’s only above 22.3% hit and expertise that we finally hit zero. The behavior in that final section of the x-axis is a little easier to see on this semilog plot, for those that are interested.
So while it’s not critical to be exactly at hit/exp cap, this plot tells us that we start opening ourselves up to 90% spikes if we’re even 1% shy of the cap. It also suggests that warrior rage generation is “bursty” enough that it pays to be well above the theoretical minimum rage threshold for Shield Block, because a missed Shield Slam at the wrong time is all it takes to create a dangerous spike event. As such, it’s probably not a bad argument to maintain caps if you’re going with a control strategy.
Last edited by Tengenstein; 02-23-2013 at 03:42 AM.
I can't help but feel that's taken out of context Gom.
The graph, from what I understand, is from the "spam shieldblock, and fuck everything else"-strategy, which only requires sufficient rage to maintain shieldblock (or rather, use sheildblock on cooldown). Theck himself mentions that hit/expertise will fall drastically behind when they exceed rage regen requried to keep up shieldblock.
I'll have to study the post more closely in it's entirety (and steal some of that delicious matlab code), but I'd be careful drawing a lot of conclusions from the that graph.
edit: yeah ok, so Tengenstein was faster and more precise
Last edited by Selendis; 02-23-2013 at 02:49 AM.
It infers to me that becuase rage comes in large blocks that vary significantly from a mean RPS (we get 20 rage one second, 0 rage the next, 10 rage the next) that being over the minimum 6.67rps required for Shield block on CD was a good thing.
It only refers to the SB-only priority, and it's not surprising in the least. Your basic thought is correct, however because we look at relatively big timeframes, those little variations become somewhat irrelevant, especially as you lose absolutely no uptime, as long as you get those 60 rage while at least 1 charge of SB is on cd, meaning roughly 18s.
For small(er) timeframes your assumption would be correct (if you get 3 or 4 SBs in 30s it's a difference, but if you get 399 or 400 in 3000s... who cares). (And if you take Theck's way of 'i haz single fight with 10000000s' (yes, the number isn't exact, but I seem to be unable to find the correct one rigth now), those little variations get even less relevant)
You didn't understand me. In Theck's simulation (and for that matter every similar simulation, meaning only(!) SB & a relatively long time) there is a softcap at roughly 6.67 RPS. It just doesn't matter for anything ... 'more advanced' (simulations including SBarr/for shorter times... not to mention the actual game).
I'm confused Quietsch, lots of pronouns, I'm not sure who you're trying to respond to. Is what you're saying that over a long time it doesn't matter? Since when do tanks care about what happens over a time average? It's all about damage smoothing and preventing spikes.
I'm trying to say that Tengs assumptions are correct for the game or any complex simulation, but they aren't correct for Thecks or similar Simulations, as those are much simpler (It would probably be more accurate to say the part which is quoted above).
(And I'm responding to Teng)
The point which some people miss is, that those are only simulations with certain conditions, thus they aren't (directly) applicable to ... basically any more complex stuff (mainly the game, but also more 'advanced/complex' calculations/whatever, based on Thecks work).
Which is happening here. Of course there exists no actual 'softcap' (which is somewhat inaccurate, but I'm missing a better term) at roughly 6.67 RPS in the game (or more 'complex' simulations for that matter), however in the simulation to which Selendis is referring, there exists a 'softcap', simply because it's a only-SB-over-a-single-long-fight-simulation (and because of this, the value of hit/expertise drops after you reach roughly 6.67 RPS, simply because you can't spend the additional rage).
Actually, Theck's simulations (as can be read in his blog) may use a large number of swings over time, he DOES look at burst streaks and not the Total Damage Reduction. I doubt TDR really needs a sim, as I actually did pretty much all the math on that in my spreadsheet.