[Guide/Theorycrafting] Jedi Shadow Tanking!

[hambone_hamlet]

Thanks for the guide, currently I am just lvl 20 but every piece of information helps me be a better tank.

[Shadess]

Correcting myself here, it would appear, that I was wrong, it does not proc at all, at-least it didn't the 20 or so times I just tested it.. so I am not sure why you would take the talent, unless of course you are using Mind Crush, which is unlikely unless you have the balance skill that makes it an instant cast, as it has a long timer + uses a lot of force to cast.

 

First off thank you so much for taking the time to at least test and try it out. I'm not sure why then either those points are in there. The only thing I can think of is that it was put in on an old calculator and they never got adjusted for say when that talent tree calculator was updated.

 

Either way it looks like this build is going to be the most plausible bang for tanking butt.

[Alexandertwo]

bump. this should really be stickied. Just the priority for single and AOE targets alone is worth it.

[flyersfan]

You've got Mind Crust instead of Mind Crush. Just a minor typo.

 

Looks like a good guide, but it's most of what I've been doing so far. The skill tree you posted is almost identical to mine. The rotation is also what I've been using. I'm rather surprised with myself, actually

[Elradrielas]

Good god, page 8? Why isn't this a sticky? Edited January 21, 2012 by Elradrielas

[Kelberos]

I was under the impression that you can proc shields off crits - but apparently thats not the case haha.

 

Though the two roll system seems rather primitive - putting everything on the same slider and all. What do you reckon?

[BroadStreetBully]

I was under the impression that you can proc shields off crits - but apparently thats not the case haha.

 

Though the two roll system seems rather primitive - putting everything on the same slider and all. What do you reckon?

 

I don't know. I like the two roll system. The two roll system give you closest to "real life". The first roll determines if you get hit. The second roll determines whether you shield the attack, take it normally, or are critically hit. A critical hit that is shielded wouldn't really be a critical hit. From a logic standpoint, it's hard to explain a critically damaging hit that was also blocked. That's just my 2 cents, but I'd be curious to hear your reasoning.

[Valkenheineken]

Assuming a 77% chance to do so (10% base dodge + 5% base block + 15% Combat Technique block + 15% Kinetic Ward + 2% Shadowsight + 4% Double-bladed Saber Defense + 5% Force Breach debuff + 6% combination (assuming a minimum of gear contribution) + 15% shield item bonus chance) and 1 attack per GCD, you can expect a further ~1 Force/sec, for a total of 11.4 Force/sec.

 

No no no. Multiple event probabilities aren't additive. You need to find the union of all the separate events in order to determine the chance of proccing that talent per hit.

 

Think about that for more than a second. Under your formula, you could have over 100% chance to do something. That violates the first rule of statistics (the probability of any given event is between 0 and 1, and the sum of the probabilities of all possible events = 1).

[Kitru]

No no no.

 

Dude, that was written back before the game was released and before I knew about the 2 roll system which is pretty obvious when you realize that I used 6% as an assumption for total gear contribution (which is about what my tank gets from Defense rating *alone*). The numbers actually *underestimate* the amount of Force gained however, so it's still completely applicable.

[Valkenheineken]

Dude, that was written back before the game was released and before I knew about the 2 roll system which is pretty obvious when you realize that I used 6% as an assumption for total gear contribution (which is about what my tank gets from Defense rating *alone*). The numbers actually *underestimate* the amount of Force gained however, so it's still completely applicable.

 

Even without the two-roll system, you still know that chance to dodge and chance to shield are separate events. If you couldn't tell intuitively, then you could tell from the beta.

 

I didn't see the timestamp, but it matters that one of the assumptions within the first few posts of a guide thread is totally wrong. Your "calculation" is no better than simply guessing that you recover about 1 extra force per gcd.

 

On top of that, the union of most probabilities are appreciably lower than their sum. So you're not underestimating, you're overestimating, and that's ignoring all of the other assumptions needed to make that conclusion (such as needing to be nearly constantly attacked by shieldable and dodgeable attacks, which is unlikely to occur every gcd).

 

It's quoted in the guide, but it's completely and totally wrong. OP, you really should remove that section and recalculate it on your own. Until then, the guide loses a lot of credibility.

Edited January 23, 2012 by Valkenheineken

[Kitru]

I didn't see the timestamp, but it matters that one of the assumptions within the first few posts of a guide thread is totally wrong. Your "calculation" is no better than simply guessing that you recover about 1 extra force per gcd.

 

And it's still based upon a fundamental assumption of 1 attack per second, which offers greater variability in the end Force/sec gained than that provided by the two roll system's percentage yield.

 

Honestly, you shouldn't get hung up on the passive Force/regen math. It's pretty much flawed by admission, but it's so close to actual play the the differences are within tolerance such that they can be used to determine results in real play. The math wasn't done to determine what our chances to mitigate damage were. They were done to determine what a realistic assumption of Force/sec regen was. I could redo the math completely but it wouldn't change the actual results of the conclusion at all: the use paradigms are still entirely applicable.

[Valkenheineken]

And it's still based upon a fundamental assumption of 1 attack per second, which offers greater variability in the end Force/sec gained than that provided by the two roll system's percentage yield.

 

Honestly, you shouldn't get hung up on the passive Force/regen math. It's pretty much flawed by admission, but it's so close to actual play the the differences are within tolerance such that they can be used to determine results in real play. The math wasn't done to determine what our chances to mitigate damage were. They were done to determine what a realistic assumption of Force/sec regen was. I could redo the math completely but it wouldn't change the actual results of the conclusion at all: the use paradigms are still entirely applicable.

 

That's funny because it's not even close to mine. In my experience, it so highly depends on what exactly is going on that you can't slap a number on it (let alone one based off of totally incorrect math). Further, the inability to do a simple probability calculation greatly undermines the rest of the analysis. That's why I'm suggesting to the OP to bring at least that portion down until it can be changed into something more accurate and useful.

 

Good thing the stickied guide is much more accurate. Shadow players can simply redirect new players to that one in the meantime.

[Arbegla]

the use paradigms are still entirely applicable.

 

Actually your 'use paradigms' are horribly wrong, as I've tried to explain to you time and time again. Here is the break down from a different thread.

So, because TK Throw uses double the animation time (3 seconds) then Slow time and Force breach, your suddenly going to be using it just as often? Even though you only want to use Slow time and Force Breach often enough for their debuff, and you want to use TK Thrust once every 30 seconds? Now I do understand what you mean by the fact that your going to use 3 seconds on Slow Time and Force breach within those same 30 seconds, its not a stable way to built a rotation on. At the end of the day, you used 2 Slow times and 2 force breach per 1 Tk Throw, so you have to consider that in your rotation. You can't just say 'well, 10% of the time your using Tk throw, and 11.1% of the time your using Slow time' when in reality your using Slow Time twice as often as your using Slow time.

 

This is where your 'percentage based' assumptions fall on their face.

 

Animation time consumed is not an accurate way to measure how often you use the powers in question, it’s a measurement of how much time it will take you to use the power. That doesn't tell you how 'often' you will use it, just how long it'll take you to use it. This is pretty obvious when you’re saying that Project, and Double Strike will use the same 'Animation time consumed' regardless of the fact that you will be using Double Strike A LOT more often you will be using Project.

 

You're right; you're not using a rotation. You're using animation time consumption, accounting for the fact that certain powers take longer to animate then others, and assuming that you will follow that same paradigm in a real world setting. In the real world (or at least the real in game world) you want something more akin to a rotation, so they people know exactly what to use. Your easy/easy rotation (which is sustainable forever, assuming single target tanking and only having to refresh Kinetic Ward when it drops) also assumes that over 35% of your animation time consumption will be Saber Strike. This will substantially lower your overall DPS (as Saber Strike is our weakest attack) and could potentially cause threat issues, which adding more willpower would not be able to fix.

 

Using your own RULE people have come to a rotation based on your priority system, which basically says 'Use Double Strike until you get PA, Use Project ONLY with PA, and use TK Throw only with HSx3. Keep Slow Time and Force Breach active at all times, and use Saber Strike when low on force.' In a 30 seconds cycle, that basically means X:3:1 (where X is the number of Double Strikes needed for PA procs) 2 force breaches, and 2 slow times, with (X-3) being the number of Saber Strikes you would use (as you only want 3 PA procs in 30 seconds)

 

Using a rotation based on actually using the attacks in question, and not based on animation time, as that could very easily be skewed not just in the fact that your accounting for 1:1 Double strike:Project, which will NEVER match in game experiences, but that you're assuming TK Throw won't be interrupted, by movement, is the best way to figure out your attacks as they allow for a better system of measurement. Accounting for pure animation time consumption, and then saying that “I operated under as assumption that PA would proc 100% of the time; it's close enough to reality to work perfectly fine” skews your numbers by a very large margin, as you are in fact nearly increasing your overall Project damage by 60%, as seen by your coefficient values (1.81 vs. 3.12) which you then multiply straight across, without regarding the 50% proc rate of your boosted DPS. Your just assuming the PA proc will always happen every time you use Double Strike, and thus you can just always account for it.

 

I understand what you’re getting at with your animation time consumption, but what you seem to fail at is understanding that comparing animation time consumption is an INSANE way to build a rotation. If you want to build a rotation that way, by all means, go for it, but it will be wrong. You want to build a rotation around your procs, how often you will be using them, and your highest damaging powers being used most, within the limits of your force regeneration. This is exactly what my rotation (and even your own RULE) considers.

 

Basically your saying that you're going to use X power (TK throw) 1 time every 30 seconds, and because it takes 3 seconds to animation, you’re going to use 10% of your time on X power, and that you're going to use Y power (Slow time) 2 times every 30 seconds, and because it takes 1.5 seconds to animate, your also going to use it 10% of your time, thus X = Y. Except, that is only a measurement of how 'long' the power takes to animate, not how 'often' you actually use the power. In reality, x = 2y as you will be using Slow time twice as often as you use TK Throw, regardless of how 'long' it takes you to use that power.

 

Your animation time consumption will NEVER match in game experiences, due to the fact it is accounting for a 100% uptime on PA, and that you will NEVER interrupt TK Throw for any reason at all; 2 things that just aren't going to happen. Heck, if you don't need the healing on TK Throw, but you need the debuff from Slow time, you would interrupt TK Throw and then use Slow time. Or if you needed extra AOE threat instead of extra single target threat.