Hrm... fascinating. Thanks for clearing those things up.
Oh, yeah I did a lot of testing on the SBV talents and stuff, all of the modifiers are additive to Each other, and then obviously applied to the base sbv.
so if you have 1000 showing on your char sheet, like Delmonte pointed out, you divide by 1.3 to get your base sbv, then go through and calculate all of your multipliers, also i believe SBV trinkets do just get added to your base SBV so that keeps things simple (been a while since i used a sbv trinket though so sorry if this information is incorrect). So say you hit shield block, have shield mastery, and ah what the heck, you even have the 5% meta gem, well (1000/1.3) * (1+1(shield block) + .3 (mastery) + .05(meta) = (1000/1.3) * 2.35
Hrm... fascinating. Thanks for clearing those things up.
oh, and don't forget, that once you start pushing into higher levels of SBV, the shield slam SBV diminishing Returns formula kicks in. HOWEVER, it has been shown, that sbv trinkets, and possibly even Shield Block exists somehow OUTSIDE of this DR. The theoretical maximum damage sbv should be able to boost your shield slam by is the upper limit of the DR curve which I think is like 2610 or something given by Blizzard. this would mean + SS regular damage, and crit damage + other modifiers your maximum shield slam damage should be around 9000 or so, but we know this is false because both in and in the case of bosses (who have armor) you can shield slam for well over 10k.
Some minor ones:
1. In your prot tree discussion you say "Nothing, ignore this tree". "tree" should be replaced by "tier"
2. You point out that bothering your 5 stacks of sunder as fast as possible isnīt as mandatory as most people think. I would make it a bit clearer that 5 stacks of sunder do increase your own tps by reducing the targets armor, not only by using the devastate itself.
3. I think you could add the glyph of cleave to the glyph choice. Agreed that you wonīt need it for boss tanking (or at least rarely) but it can be a very good glyph for trash oder add tanking.
4. Altough you skipped mentioning minor glyphs I would mention the glyph of thunderclap which increaes your TC radius, which is also a nice addition when tanking adds / trash. And minor glyph choices are simple enough anyway.
Yes, I understand this completely.
My point is simply, that if you write in a guide, that your initial burst threat is more important (which I do not question), you should still also mention that the sunder will also increase your own threat. Guides like this are usually for less experienced players and those might get drawn away form the information that the sunder debuff itself also improves your threat. Even if it s not the most important thing for the initial threat lead
Just for the sake of completeliness.
This is true, but it is REALLY suboptimal for your burst threat at the beginning of a fight to get the sunder stacks up. Like yes it does increase your TPS to have 5 stacks up, but the TPS loss from getting 5 devastates out first before unloading your big guns is too big. Let the sunder stacks get up as you go or make a fury warrior do it or something =P.2. You point out that bothering your 5 stacks of sunder as fast as possible isnīt as mandatory as most people think. I would make it a bit clearer that 5 stacks of sunder do increase your own tps by reducing the targets armor, not only by using the devastate itself.
Glyphs that might work:3. I think you could add the glyph of cleave to the glyph choice. Agreed that you wonīt need it for boss tanking (or at least rarely) but it can be a very good glyph for trash oder add tanking.
Taunt: increased chance to make sure taunt lands if someone does pull aggro
Vigilance: If you are a prot with Vig then this one can help, but I personally think other glyphs are better.
Mocking Blow: Great snap threat and has a fixate if you need it to wait for taunt to come back up.
Cleaving: for AoE
Already got ya =D it's a very good point though
True, I do cover minor glyphs in my prot war guide though... I think.4. Altough you skipped mentioning minor glyphs I would mention the glyph of thunderclap which increaes your TC radius, which is also a nice addition when tanking adds / trash. And minor glyph choices are simple enough anyway.
Idk this is why it's a quick and dirty guide just to get people on the right track, I typed it out pretty quickly without really proof reading it =/. Admittedly not my best work ever, lol.
Last edited by Aggathon; 07-30-2010 at 07:25 AM.
As long as it getīs the job done...
Ah and if you use glyph of devastate (which you already recommend), then weīre just talking about 3 Devs, not 5.
The dynamic of stacking Devastate is funny, though.
Think of it this way: you're also increasing the TPS of all the physical DPS classes while you do it--so you don't get as much out of it as you think.
The static increase per Devastate doesn't care about the time it takes to stack, because you always have the same penalty either way--5 Devastates with slightly lower stack-based damage. These being early or these being late in the fight makes no difference at all.
So, the only real gain to be had from using Devastate quickly is a slight TPS increase in relationship to caster classes. Even then, this gain is not worth the loss of maintaining a proper rotation. (Shield Slam is too powerful to ignore.)
Aye aye, good points.
I could say now, that it also improves the threat you get tricked to or mislead, but I am absolutely willing to accept (and never said anything else), that it will not change your initial threat lead style priorities.
BTW, in reading the OP I slightly disagree with your priority model. I think you may be suggesting something that is non-optimal.
As far as the numbers I have seen indicate, the optimal rotation is:
Shield Slam -> Revenge -> Devastate -> (Thunder Clap if needed > Demo Shout if needed > Shockwave > Devastate)
With this cycle restarting on every Shield Slam cooldown cycle or SnB proc.
You never want to delay your Shield Slam cooldown and you never want to miss early-cooldown SnB proc opportunities. The 3rd ability in the cycle is the least valuable as SnB does not provide a TPS increase if it procs on that GCD. (Shield Slam will come off cooldown for the next GCD anyway, so it is only a rage cost decrease.) This makes it perfect for using Shockwave or refreshing raid debuffs.
An optimal cycle is not just about damage or threat per execute, but also about opportunity cost. Leaving Shield Slam on cooldown or not attempting to proc SnB during the most valuable times can result in a significant net TPS loss.
Performing any ability that can proc SnB when Shield Slam is off cooldown is a huge decrease in TPS. Performing any ability that cannot proc SnB when Shield Slam has > 1.5s cooldown on it is also a major decrease in TPS.
Edit: And, yes, all things Kaze says about Block Value% modifiers are correct. They are all additive with eachother. (This is a bit inconsistant with many %-based statistics in the game, but oh well.)
It is also worth noting that if you have Deep Wounds, it is likely that Shield Slam will always be your highest threat per execute attack anyway.
Shockwave will deal more base damage and threat on average, however it has a lower crit rate. If you have a decently high DPS weapon, it will likely be marginally more threat to use Shield Slam than Shockwave--even discounting SnB. At the very worst, they are quite close in average threat per execute either way, thus the additional value provided by SnB makes using Shield Slam first the optimal thing to do.
Shockwave deals more damage per execute by a fair margin, but that's where SnB comes into play. The fact that they are so close together on threat per execute simply makes it a much simpler decision.
(I may have skimmed over it, but I didn't see any mention of the fact that Shield Slam gets 30% bonus threat in addition to its innate value of 770.)
Glyphed Devastate only deals 70% of the threat of a Shield Slam (and far less damage) and thus having the opportunity to proc SnB is one of the key things that raises its average threat per execute.
If you think about it this way:
Shield Slam: 16730 threat
Devastate: 11758 threat
Without SnB, you would need to do 3 Devastates, at a loss of 3314 TPS. If you proc SnB on the first Devastate, you are saving yourself the loss of that 3314 TPS over 3 seconds (two GCDs), which is a gain of 9942 threat. You have a 30% chance of this, so on the first GCD you have, on average, 2983 additional threat (25% increase), and 1492 additional threat (12.7% increase) on the second GCD.
If you use a Devastate when Shield Slam is on cooldown, not only are you losing the 4971 threat per second for delaying your Shield Slam cooldown vs. Devastate, but you are also losing the 2981 average additional SnB threat from doing it after instead of before. So, if Shield Slam is off cooldown and you ever do Devastate -> Shield Slam instead of Shield Slam -> Devastate, you are losing 2650 TPS--which is probably 20-25% of your total average rotational maximum.
Last edited by Kojiyama; 07-30-2010 at 08:08 AM.
I think you make a really good point about using SnB proc'ing abilities in the first 2 globals after shield slam. I've made that a habit of that since I read a post here on tank spot about a user comparing his shield slam qty to that of his co-tank. He noticed his co-tank had marginally more Shield Slam's then he did, yet he was convinced he was doing his rotation correctly. He was using shockwave as per the threat/use priority. His co-tank was making a habit of doing as you suggest, using non-snb proc'ing abilities on;y on the 3rd global after Shield Slam. IIRC he noticed approx 10% difference in number of Shield Slams. I switched my priority around a bit after reading that and noticed that my shield slams / min went up (although I don't recall how much).
Basically I concur with what you posted
If shockwave is better average threat then wouldn't it be optimal to include it in your rotation once you have a threat lead but then for burst threat/at the begining use shield slam?
I'm really lazy so unless someone wants to write a more complete guide then just "oh hey this is the raw basics and this works" I'm just gonna put a disclaimer at the top that suggests to read the entire article because you guys are giving way better explanations than I could. Threat is not my area of expertise (no pun intended). All I know is that when I tanked I did that and never had any threat problems and geared soley for survivability.
Even though technically you are losing TPS by delaying a Shockwave in some ways, using it when it is a larger threat boost than the global when you use it.
For example, technically one could argue that as shown above the Devastate following a Shield Slam is only worth 2983 additional threat on average, which is less than the difference between Shockwave and Devastate. (Which is probably around 5000 threat per execute.)
However, that's not the complete story. Really, what you should be looking at is how much threat you lose from not using Shockwave at that time.
Using Shockwave instead of a Devastate is only an increase of 249 TPS over its entire cooldown. Delaying this by 2 globals, or 3 seconds, lowers it to a 216 TPS, which is only a loss of 33 TPS. However, I believe that ignoring a SnB proc opportunity would result in a much greater loss of threat.
The main issue with SnB is that it is cyclic. One SnB resets your cycle and gives you more opportunities to get SnB procs. By ignoring opportunities to use SnB, you aren't just losing threat on that GCD but missing an opportunity to get additional higher-value SnB procs. (e.g. if the first Devastate following a Shield Slam is the highest value, the more SnB procs you have the more time you will spend converting those GCDs into Shield Slams rather than spending time in the lower end of the SnB GCD opportunity spectrum.)
Thus, using the last GCD in a cycle for Shockwave has zero loss in SnB efficiency with minimal loss in Shockwave efficiency. Delaying a Shockwave by 1.5-3 seconds only represents 7.5-15% of its total cooldown, which is often a worthwhile tradeoff.
(I may double-check these numbers this evening to make sure this is all 100% accurate...it's been a little while since I analyzed this!)
So, here's another way of looking at it.
I'll just approximate here (because it's close enough to accurate) that Devastate and Revenge have roughly the same threat and Shield Slam and Shockwave have roughly the same threat.
Shield Slam/Shockwave: 16730 threat
Devastate/Revenge: 11760 threat
Average SnB cycle is 4.7844s and contains the following distribution of abilities:
1 Shield Slam
= 42480 threat or 8878.8 TPS
Over 20 seconds, that would be 177577 threat.
First Post-Shield Slam GCD
By inserting Shockwave after a Shield Slam instead of a Devastate, you have having a 30% chance to extend the length of a single cycle by 3 seconds. So, it takes 20.7656 seconds to deal 177577 threat plus the bonus 4970 threat of a Shockwave instead of a Devastate.
So, in that case, you would be dealing 182547 threat or 8791 TPS over 20.7656s seconds. This is 87.8 TPS lower than a normal cycle.
Second Post-Shield Slam GCD
By inserting Shockwave in the 2nd GCD after a Shield Slam instead of a Devastate, you have having a 21% chance (.7 * .3 ) to extend the length of a single cycle by 1.5 seconds. So, it takes 20.315 seconds to deal 177577 threat plus the bonus 4970 threat of a Shockwave instead of a Devastate.
So, in that case, you would be dealing 182547 threat or 8985.8 TPS over 20.315 seconds. This is 107 TPS more than a normal cycle.
Third Post-Shield Slam GCD
However, if you were to insert the Shockwave into a non-critical GCD you would be keeping the same average cycle length (because you weren't altering the SnB situation) with the bonus 4970 threat of the Shockwave at some point in the natural course.
So, in that case, you would be dealing 177577 + 4970 threat in 20 seconds, or 9127 TPS over 20 seconds. This is 248 TPS more than a normal cycle.
Third vs. Second GCD
To solve for the cross-over point for using it in the 2nd GCD rather than the 3rd GCD, you could simply calculate:
8878.8 + 4970 / S = 8985.8
4970 / S = 107
4970 = 107 * S
46.44 = S
So... provided you are never delaying your Shockwave for more than 26 seconds beyond its cooldown (highly unlikely! You have a 49% chance on every cycle to get to the 3rd GCD without a SnB proc. So even in a worst-case alignment, you should not see a delay greater than 9.76 seconds on average) it is best to wait for the 3rd post-Shield Slam GCD opportunity to use the ability.
Even in the worst case situation of a 9.4 second delay, you would be dealing 9047.8 TPS, which is a 169 TPS increase over a normal cycle and still greater than even the best-case 2nd GCD scenario. (Keeping in mind that even a 2nd GCD cycle could be delayed by a few seconds as well if misaligned, although by a lesser amount.)
Disclaimer: I'm doing this math a bit quickly on my breaks, so it may or may not have grevious errors in it. I will double-check it later... or you can feel free to yell out should you find any grevious errors and I will correct it!
Last edited by Kojiyama; 08-02-2010 at 01:13 PM.
Really makes me glad to see that i'm not alone in my only shockwaving for threat 4.5secs after an SS.
Back to the original post (really good BTW), you do mention the correct timining of whne to pop SB is crucial, but in the "On the pull section" i don't see any mention of popping SB to increase the intial SS's threat and to force revenge to light up. Thus giving you an almost certain intial rotation of SS->Rev. As well as increasing the threat from damage shield. Likewise it doesn't mention that if you shouldn't really pop SB just after stunning something for optimal Threat gen.
I tend to also pop Beserker rage on the pull as well aslong as i'm not gonna need it soon after. Allows more HS queue at the start and stops the "charge->sapped->wipe" phenomenon.
very good points tengenstein.
I haven't worked through the whole thing Kojiyama but a few things little things I spotted were:
1) S&B is a 30% chance to proc when you deal damage and depending on your gear, even in full ICC gear you might have as low as 80% chance to connect with revenge or devastate. Therefore your chance to proc S&B might be as low as 24% per revenge/dev.
2) It won't affect the conclusion but in your scenario of using Shockwave on the first global after shield slam, you also have a chance to proc S&B on the second GCD after using SS. So (assuming you connect 100% of the time with dev) you'd have a 21% (0.3 x 0.7) chance to delay SS by 3 seconds and a 9% (0.3 x 0.3) chance to delay it by 1.5.
3) The chance to get to your 3rd GCD without proccing S&B (if you use an ability that can proc it each global and connect 100% of the time) is 49% (0.7 x 0.7) not 51% but that's not going to make any difference to your point there. 51% is the chance that you will get a proc from either of the 2 GCDs following SS.
Er, I think that's right anyway, I also worked through that pretty quickly
I also get quite different ratios of threat for dev vs. SS vs. Revenge to you but maybe there are mistakes in my working, I get revenge quite close to SS in my gear in ICC fully buffed at around 14k threat and devastate at around 10.6k. I'd say the ratios depend a fair bit on your gear/buffs etc but it seemed a bit unusual to have revenge so low compared to SS regardless of setup in a usual ICC setup. No-one's given me much feedback on my working though so I could just be way off on that.
Last edited by Delmonte; 08-02-2010 at 01:09 PM.