Jump to content
FORUMS
Sign in to follow this  
Sukata

Protection - Performing evaluation on ToS N

Recommended Posts

Hello all fellow warriors!

I play Protection Warrior since the beginning i came across Azeroth (somewhere in 2006). Always have been my main despite all the buffs, nerfs etc they suffer with the pass of all the xpacs. My other alts are also but i never felt keen to reroll other class to be main char. Used to be core raider back on Burning and Lich King days, heavy consumer of Tankspot and Elitist Jerks when then were up but short after Cata was released the usual rela life commitments made me pass to a relaxed casual way of play. Until Legion where i wanna burst a bit my fingers! :)

After this short introduce i submit to your aprecition my last week reset of ToS N and my armory link.

After some close looks i do realize i might be underperforming mainly cause i am not using properly SB/IP. I am IP way too much thar using Shield Block, sometimes or most of the times a full health which is a complete waste.

Was searching for some opinions so that i can improve on that and other things i might be doing wrong.

https://worldofwarcraft.com/en-gb/character/lightnings-blade/sukata

https://www.warcraftlogs.com/reports/jvHDKAh4r32yTzw8

Thanks all for the help and keep on the good work! :)

 

Share this post


Link to post
Share on other sites

Welcome Sukata!

I'll try to get to this in the next few days, probably on Friday due to work and raid. Were there any fights in particular that you felt especially weak on? 

  • Like 1

Share this post


Link to post
Share on other sites

Hi sukata and wellcome back

I try to give you some hints and my pov. I'm estarriol will give you some more (and better) advice. 

Gear

If your arsenal is up to date, you have to much crit. If you feel comfortable with 20% haste, then try to raise your mastery and versatility. I recommend to play around 25% haste (or even more if it is worthwile - depending on your looted pieces). First thing you should do: change your enchants and gems to haste. If you need some deeper explanation why you should do that, let me know. 

Talent

Well, i prefer to play with devestator instead of indom as long boss ability can't onehit. Again, if you need some deeper explanation, feel free to ask. If you're afraid to use it because some guys stated that "swing-timer" playstile...just ignore it. What they did is a theoretical mess which doesnt work fine at all and cause worse performance (plz dont ask for a proof). 

Playstile

I just had a short look in your logs. I'm not a pro in logreading but i think there are some points. 

- your sb-uptime seems low. But this can have different causes. If you play with heavy rep. i recommend to use sb even if your not tanking (more dps). 

- if you play with vengeance, i should use ravenge more often to fully benefit from this talent. There is a ragecap on which you should allways use ravenge before IP if you dont have the buff for reduced IP. Im not sure, but i its somewhere around 58. 

- you should try to just cast full IP's if you use vengeance. There are several situations where you dont have to pay attention to that. But you gain the most survivability if you do. 

- casting IP on full health isnt wasted at all. I know there are players who try to smooth the inc as much as possible...and loosing so much survivability. Using IP in a right manner is a own chapter with different opinions. My advice: use full IPs when they're ready, pool rage if its usefull due to mecanics 

  • Like 1

Share this post


Link to post
Share on other sites

Hello Estarriol!

It's fine when you can take a look either on Friday or another suitable day. 

Considering all the fights and also considering that after those Normal run i was able to step in HC mode and kill 3 bosses (Goroth, Harjatan & Inquisiton).

Confirmed weeknees on Inquisition mainly due to Scythe Sweep. Found myself getting way more damage than the other tanks i came across on Pugs (mainly Guardians tbh) and using Cds somehow earlier than i was expecting. On Normal, despite taking more damage then the Pally with, never really felt in danger.

The other, or perhaps the biggest weakness i detect in  myself on Normal mode, was the felclaws on Kiljaeden. If you observe the logs we wiped twice mainly because of me not holding Felclaw damage. We managed to kill him but...disliked way too much the damage my warrior took in comparison with Paladin. Cant really say if it is meant te be like that due to the diferrences between the 2 classes or if it was my fault as i think.

Thanks for taking your time into this! :)

Edited by Sukata

Share this post


Link to post
Share on other sites
4 hours ago, Allseye said:

Hi sukata and wellcome back

I try to give you some hints and my pov. I'm estarriol will give you some more (and better) advice. 

Gear

If your arsenal is up to date, you have to much crit. If you feel comfortable with 20% haste, then try to raise your mastery and versatility. I recommend to play around 25% haste (or even more if it is worthwile - depending on your looted pieces). First thing you should do: change your enchants and gems to haste. If you need some deeper explanation why you should do that, let me know. 

Talent

Well, i prefer to play with devestator instead of indom as long boss ability can't onehit. Again, if you need some deeper explanation, feel free to ask. If you're afraid to use it because some guys stated that "swing-timer" playstile...just ignore it. What they did is a theoretical mess which doesnt work fine at all and cause worse performance (plz dont ask for a proof). 

Playstile

I just had a short look in your logs. I'm not a pro in logreading but i think there are some points. 

- your sb-uptime seems low. But this can have different causes. If you play with heavy rep. i recommend to use sb even if your not tanking (more dps). 

- if you play with vengeance, i should use ravenge more often to fully benefit from this talent. There is a ragecap on which you should allways use ravenge before IP if you dont have the buff for reduced IP. Im not sure, but i its somewhere around 58. 

- you should try to just cast full IP's if you use vengeance. There are several situations where you dont have to pay attention to that. But you gain the most survivability if you do. 

- casting IP on full health isnt wasted at all. I know there are players who try to smooth the inc as much as possible...and loosing so much survivability. Using IP in a right manner is a own chapter with different opinions. My advice: use full IPs when they're ready, pool rage if its usefull due to mecanics 

Hello Alseye!

 

Thank you so much for you review! So much appreciated!

Starting by the begginning, there was a mistake on the trinkets of my normal gear as the 2nd trinket i use is Phantasmal Eccho because of the huge ammount of Haste it provides.

I am also eager to learn from you the main reasons to swap all gems and enchants to haste. I am here to learn and i really want it! :)

I use indomitable over Devastator because i was struggling with Rage. I used to talent Devastator during Nighthold but i somehow found myself lacking rage. Didnt had a good reason for that (and also didnt research as much for it) so i just swap back Indomitable for the nice extra IP. But would like to hear from you also on this matter.

Shield Block is way to low as my oppinion goes and need to improve on that...a lot! I play with heavy reper. and vengeance and will check why Revenge is not properly cast.

Another thing that comes across is that i think i am making Low DPS. If we compare to the Paladin that tanked with me i am making about half his DPS. Seems too much and there must be something wrong.

Always played with 15button mouse and slowly trying to keybind but it is been a bit of a struggle as it takes time to "forget" so many years of mouse playing! But we can talk about this other time!

Thanks for your time onto this! Cheers

Edited by Sukata

Share this post


Link to post
Share on other sites
10 hours ago, Sukata said:

The other, or perhaps the biggest weakness i detect in  myself on Normal mode, was the felclaws on Kiljaeden. If you observe the logs we wiped twice mainly because of me not holding Felclaw damage. We managed to kill him but...disliked way too much the damage my warrior took in comparison with Paladin. Cant really say if it is meant te be like that due to the diferrences between the 2 classes or if it was my fault as i think.

Prot-Warrior seems to be one of the best tankclass on KJ. The reason why you got so much dmg is mainly due to missing SB. You can also use your artefact-ability if sb should fall off or on high stacks (crit-block 4 the win :-))

10 hours ago, Sukata said:

I am also eager to learn from you the main reasons to swap all gems and enchants to haste. I am here to learn and i really want it! :)

Basically is this (copied from guide on icy veins): Haste lowers the cooldown of your auto attacks and of Shield Block IconShield Block, Revenge IconRevenge, Shield Slam IconShield Slam, and Thunder Clap IconThunder Clap, which indirectly boosts your survivability and damage output.

Beside lesser CD  on the abilities, you will get more shield slams (with devestator because more auto-attacks -> more SS proccs), which leads to higher uptime of SB and more rage (rage is not that significant because most rage we generate by damage taken).

The second reason why you should not use versatility: it's scaling is much worse than haste/mastery. That means: You need more points in versatility to gain 1% than you need for haste/mastery.

10 hours ago, Sukata said:

I use indomitable over Devastator because i was struggling with Rage. I used to talent Devastator during Nighthold but i somehow found myself lacking rage. Didnt had a good reason for that (and also didnt research as much for it) so i just swap back Indomitable for the nice extra IP. But would like to hear from you also on this matter.

Unfortunately, this is a wrong approach. With devestator you WILL earn more rage because of different reasons (no GCD waste for devestate -> more TC/Revange, more SS proccs). Ok, the ragegain should not be a big difference while your tanking (due to rage per damage taken). But in my opinion devestator is a survivibility gain because of higher SB-uptime (due to more SS proccs with heavy rep.).

 

I hope this gives you a deeper understanding and more room for improvement :-)

Share this post


Link to post
Share on other sites

Before I dig into the logs themselves, I'm going to add a few comments related to others in the thread already. I very much like the back and forth the two of you have had!

 

Quote

casting IP on full health isnt wasted at all. I know there are players who try to smooth the inc as much as possible...and loosing so much survivability. Using IP in a right manner is a own chapter with different opinions. My advice: use full IPs when they're ready, pool rage if its usefull due to mecanics

I both agree and disagree with this and that's largely due to situational usage. I'm going to talk mainly about two scenarios first.

1. You're about to take a large hit.

In this case, you would want to get as close to IP-capped as possible. It is not at all a waste to be using IP at full health here and not using IP could be fatal.

2. You're taking relatively consistent damage (melee hits, DoT ticks, etc).

In this scenario, it can be worthwhile to wait. For example,  If I'm taking just regular boss melee and my life isn't in danger (e.g. not Felclaws!) and I know that the next hit will put me down a bit of health if I don't use IP but not enough to be dangerous, then I will hold the IP. Then I take the hit and I'm down to let's say 75% HP and at this point I use my IP. In my gear, this will give me about a 1.2mil shield.

So why would this be desirable over using the IP before this first melee? With Mannoroth's Bloodletting Manacles, you'll be getting 6% healing back from that IP, which puts you back at 81% already. Additionally, healers need a certain amount of time to react to your damage taken. Taking a bit of damage first and then putting up the shield allows them time to react to that first hit. When the second hit lands (assuming roughly same size hit), my IP is about 1/5 of my health per usage so that would absorb 20 out of that 25% hit, leaving me with a hit that take 5% of my total health. At the end of the two hits, I'm left with 76% HP.

If I don't do that and I pop my IP right away, I lose the 6% healing, but I'm at 120% effective HP before that first hit lands. It hits, takes the same 25%, and leaves me with 95%. Because I used my IP already, I don't quite have enough rage for another one before the second hit. Then the second hit lands, takes that 25%, and I'm left with 70% HP.

Overall, it is a small difference, but one that can be significant over the course of the fight. It's just a small consideration that can help you optimize a bit. It's not something to really worry about and at the end of the day, if you think there's a chance the next hit might kill you, absolutely use that IP and don't wait. Similarly, don't wait if doing so will lead you to cap your rage bar.

Quote

There is a ragecap on which you should allways use ravenge before IP if you dont have the buff for reduced IP. Im not sure, but i its somewhere around 58.

When playing with Vengeance, if you don't have the buff to reduce IP's cost, then Revenge first regardless of where your rage bar is at so you get that benefit (this is almost always the case, I'll go into exceptions shortly). Mathematically, this will give you the most bang for your rage.

No buffs: IP first, then revenge

60 (IP, no discount) + 20 (discounted Revenge) = 80 rage

No buffs: Revenge first, then IP

30 (Revenge, no discount) + 39 (discounted IP) = 69

So to get those buffs rolling, using Revenge first will net you a savings of 11 rage.

That's just my advice for if you have no buffs up. Once you do, use the thing that is discounted first. Personally, I like to use my abilities in combos. If I want an Ignore Pain up, I'll use both IP and Revenge so that I reset back to the same point and don't have to remember or check which buff I have. My preference is to leave it so that my next IP is discounted so that I can use a discounted IP in a snap if needed and revenge when convenient.

Okey, so when is it not advisable to make sure you're getting that combo right for maximum rage savings? Basically, when you're about to take a big hit in a short period of time or if you're about to take that hit with no ability to generate more rage. Let's say I have saved up 130 rage and I'm running out to take Odyn's spear. As I'm running out, I'll hit IP, then Revenge, then IP, then IP again. This gives me 1 full IP from the first cast and including Revenge's cost, leaves me with 71 rage. The second IP cast gives me a full IP for a cost of 39 rage, leaving me with 32 left. The remaining IP using 32 rage gives me about half of a full IP. My total here is 2.5 times my normal IP size.

If I keep the combo rolling the second time (IP, Rev, IP, Rev), then I get a full IP with the first combo for 59 total rage, leaving me the same 71 left. The second combo gives me a second full IP, but costs 59 rage. This leaves me 12 rage. The minimum discounted IP cost is 13 rage. So I get a total of 2 IPs out of this and end up taking about 600k more damage (in my current gear).

Why bother with the combo at all since in this scenario it wouldn't even hit something? Assuming I start with the discounted IP, then I'd get a full one for 39, leaving me 91 rage, a second full one for 60, leaving me 31 rage. And just about half of another IP. So this total is around 2.5 as well. It is also a valid method; however, if you had gotten a Revenge proc right before you ran out, you wouldn't benefit from it this way.

Quote

you should try to just cast full IP's if you use vengeance. There are several situations where you dont have to pay attention to that. But you gain the most survivability if you do. 

Firm agree! To go into it a bit deeper, let's look at some numbers.

A discounted Revenge is 20 rage.

A discounted max IP is 39 rage.

A discounted min IP is 13 rage.

So comparing combos, a max IP+Rev costs 59 rage. Without Vengeance, these same two abilities would together cost 90 rage, a savings of 31 rage.

A min IP+rev combo costs 33 rage. Without Vengeance, these same two uses would cost 50 rage, a savings of 27 rage. But that's not the right way to think of it when using Vengeance. When you're using Vengeance, think of it in terms of cost to put up the same amount of IP if you weren't using Vengeance at all.

With Vengeance, the cost of putting up a full IP is 59 rage, compared to 60 without Vengeance. If you happen to get a Revenge proc, you get a full IP for 39 rage, neat!

With Vengeance, the cost of putting up a min IP is 33 rage, compared to 20 without Vengeance. This dramatically increases the cost of your survivability if you're using small IPs.

So why is Vengeance often used for survivability, then? Revenge procs lead to rage savings and even if you don't get Revenge procs, you can still output respectable damage with no cost to survivability. The rage savings you're likely to get will often exceed the 60 rage every 1.5 minutes from Booming Voice, especially if you're getting meleed by multiple mobs.

Another consideration for prioritizing max IPs is Dragon Scales, one of our golden traits. If you have this proc, you gain more by using a max IP.

140% of a max IP is... 140% of your max IP. 40% free IP!

140% of a min IP (which is 33% of a max IP) is 46.2% of a max IP. This is 13.2% free IP. =(

Quote

(rage is not that significant because most rage we generate by damage taken)

Very disagree! To the numbers!

For Sukata's N KJ kill, 1675 rage was generated between Intercept, Thunder Clap, and Shield Slam. 1075 rage was generated from damage taken.

For his (please correct me if I'm wrong) Harjatan kill, 1093 rage from active abilities and 1031 from damage taken.

For his Maiden kill, 948 from active abilities, 563 from damage taken.

Keep in mind this is without Devastator, so the divide is even more significant with it.

If I look at my own logs, I'll look at Harjatan mythic as it hits extremely hard and I'm taking high damage nearly the whole fight. This is kind of a best case scenario for RFDT. I have 3966 rage generated between Shield Slam, Devastator, Demoralizing Shout (from Booming Voice), Thunder Clap, and Intercept. I have 2568 from damage taken.

Haste is a large contributor to our rage generation because it affects the cooldown on Thunderclap and Shield Slam as well as reducing the amount of time between auto attacks (and therefore Devastates with Devastator). More Devastator auto attacks and more Thunder Claps lead to more rage by themselves but also leads directly to more Shield Slams, which are a huge amount of out rage generation by themselves. In most fights, the rage generation from Shield Slams alone will be near or even exceeding the rage generated from damage taken even on heavy-damage fights.

Quote

The second reason why you should not use versatility: it's scaling is much worse than haste/mastery. That means: You need more points in versatility to gain 1% than you need for haste/mastery.

I very much disagree with nearly all of this as well. Our stat priority is Haste>Mastery>=Versatility>Crit. This means that a piece of gear with haste/mastery is very nearly equal to a piece with haste/verse. The reason mastery>=versatility is that mastery gains more benefit in fights with heavy melee damage. Versatility closes that gap on fights with high magic, bleed, or otherwise unblockable damage fights.

While it's true that you need more stat points to gain 1% for vers than you do for mastery, haste, and even crit that doesn't make it inherently weaker. 1% mastery does not have the same benefit as 1% versatility as 1% vers will give you a flat 0.5% damage taken reduction and a flat 1% increase to IP. The effect of 1% mastery will be much less. However, you need significantly different amounts of actual stat points to reach 1% mastery vs vers.

That said, our stat weights are actually pretty even. This means that ilvl dominates stats in most cases. Get the highest ilvl gear that you can. If that gear has haste, even better.

Crit is our weakest secondary, so in some cases you would want a slightly lower ilvl piece with better stats. If for example, you had the choice between a 905 belt with high crit/ low vers or a 900 belt with high haste/ low mastery, go for the 900.

But really, don't worry too much about it. Go for highest ilvl in general, bonus cookies for haste.

On 7/6/2017 at 3:18 AM, Allseye said:
On 7/5/2017 at 4:39 PM, Sukata said:

I use indomitable over Devastator because i was struggling with Rage. I used to talent Devastator during Nighthold but i somehow found myself lacking rage. Didnt had a good reason for that (and also didnt research as much for it) so i just swap back Indomitable for the nice extra IP. But would like to hear from you also on this matter.

Unfortunately, this is a wrong approach. With devestator you WILL earn more rage because of different reasons (no GCD waste for devestate -> more TC/Revange, more SS proccs). Ok, the ragegain should not be a big difference while your tanking (due to rage per damage taken). But in my opinion devestator is a survivibility gain because of higher SB-uptime (due to more SS proccs with heavy rep.).

Yep, Devastator absolutely gives you more rage than Indomitable. With Indomitable, you get exactly the same rage generation as you would if you had no talent selected at all. The baseline, actively used Devastate provides no rage by itself and only serves as a filler to proc Shield Slams. Devastator, on the other hand, removes the need to fish for Shield Slam procs via Devastate spam in addition to making the Devastates that proc from your auto attacks generate 5 rage on their own. On top of that 5 rage per auto, you also significantly increase your number of proc opportunities because you're having proc chances both with your active abilities (Thunder Clap and Revenge) and your inactive ones. With my gear, I get roughly (very roughly) 4-5 proc chances from active abilities in 5 seconds and about 3 (nearly 4) proc chances from inactive abilities in those same 5 seconds, for a total of about 8 proc chances per 5 seconds. Without Devastator, I'd get just about 5 proc chances per 5 seconds. So that's (very, very, very) roughly about a 60% increases in proc opportunities. This doesn't directly translate to 60% more Shield Slams as RNG is RNG, but you can see that it will lead to significantly more procs over the long run.

If we look at that same Harjatan kill of mine that I mentioned earlier, I got 740 rage from Devastator from the 5 rage/auto attack alone. Even if we assume pretty bad procs and say that only 20% of my Shield Slams were extras procced by having Devastator, then I gained another 364 rage from Devastator, for a total of a rough 1100 rage. My total rage gain on that fight was 6534.

TL;DR Devastator gives you a ton more rage. That extra rage gives you more IPs and those IPs are more distributed throughout the fight, leading to better damage intake smoothness than you would get from the 25% increased IP size from Indomitable.

So when is Indomitable better? Basically, when you're in danger of getting 1-shot killed. If not, you'll get more mileage out of Devastator.

 

Okay, this got a lot longer than I was anticipating, so I'm going to post this one and then make a separate post for the direct logs analysis.

Share this post


Link to post
Share on other sites

Nice, i really like the detailed answer you gave Estarriol ...like expected :-). There are some aspects i'd like to clarify...and perhaps we wont stay at the same opinion (what isn't a bad :-))

14 hours ago, Estarriol said:

When playing with Vengeance, if you don't have the buff to reduce IP's cost, then Revenge first regardless of where your rage bar is at so you get that benefit (this is almost always the case, I'll go into exceptions shortly). 

I have to partiallly disagree with this. If i'm sitting on low rage and don't got a ip-reduce buff, i try to avoid casting Revenge.

So lets say i'm sitting on 30 rage. What will happen if i would cast Revenge? Casting Revenge (lets say with buff for reduced costs) will bring me to 10 rage -> no rage to even cast a smaller IP. If i feel necessary to cast IP, then i cast a 30 IP and gain effectively survivability (because it was necessary). But if i simply cast Revenge to obtain the buff, this will lower my flexibility in using IP and this will lead to less survivability. This is why i recommend to  cast Revenge not until 58 rage. Ok, we won't gain the most benefit of vengeance...but i'm interested in most survivability and not in some fancy values in wrong situations.

As i said...eventually we aren't the some opinion. But yes, if i got more than 58 rage i'm using Revenge...i'm also using it if i already got the IP-reduce buff if i'm not pooling rage and inc-dmg is irrelevant.

14 hours ago, Estarriol said:

Very disagree! To the numbers!

For Sukata's N KJ kill, 1675 rage was generated between Intercept, Thunder Clap, and Shield Slam. 1075 rage was generated from damage taken.

For his (please correct me if I'm wrong) Harjatan kill, 1093 rage from active abilities and 1031 from damage taken.

For his Maiden kill, 948 from active abilities, 563 from damage taken.

Keep in mind this is without Devastator, so the divide is even more significant with it.

If I look at my own logs, I'll look at Harjatan mythic as it hits extremely hard and I'm taking high damage nearly the whole fight. This is kind of a best case scenario for RFDT. I have 3966 rage generated between Shield Slam, Devastator, Demoralizing Shout (from Booming Voice), Thunder Clap, and Intercept. I have 2568 from damage taken.

Ok, i have to relativze my statement. Ragegain from abilities is significant. But keep in mind: you don't find yourself always tanking. I'm not sure if we should compare rage from abilities with RFDT over a whole fight (evtl. this is why you wrote "to the numbers").

14 hours ago, Estarriol said:

I very much disagree with nearly all of this as well. Our stat priority is Haste>Mastery>=Versatility>Crit. This means that a piece of gear with haste/mastery is very nearly equal to a piece with haste/verse. The reason mastery>=versatility is that mastery gains more benefit in fights with heavy melee damage. Versatility closes that gap on fights with high magic, bleed, or otherwise unblockable damage fights.

While it's true that you need more stat points to gain 1% for vers than you do for mastery, haste, and even crit that doesn't make it inherently weaker. 1% mastery does not have the same benefit as 1% versatility as 1% vers will give you a flat 0.5% damage taken reduction and a flat 1% increase to IP. The effect of 1% mastery will be much less. However, you need significantly different amounts of actual stat points to reach 1% mastery vs vers.

That said, our stat weights are actually pretty even. This means that ilvl dominates stats in most cases. Get the highest ilvl gear that you can. If that gear has haste, even better.

Crit is our weakest secondary, so in some cases you would want a slightly lower ilvl piece with better stats. If for example, you had the choice between a 905 belt with high crit/ low vers or a 900 belt with high haste/ low mastery, go for the 900.

But really, don't worry too much about it. Go for highest ilvl in general, bonus cookies for haste.

Ok, i don't want to start a discussion about state-weights for prots. This always ends in a mess :-).  You can disagree with my statement but i wont change my recommendation to not enchanting and gemming versatility (specially not in high-blockable content). As long as someone has a low haste-rating there is basically NO reason to not enchant haste.

The second thing what you pointed out is not sacrificing ilvl due to crit. So, on ring, neck and trinkets there should not be crit at all. Even your crit-piece is 10 or 15ilvl higher. We just don't really need the stamina on it...if we would need it, then we go and play with indom. Don't got me wrong - Stamina is nice but if there are wrong states - nope thx. For other gear-slots where i got armor, strength ... yes, crit don't bother me that much. I absolutely agree the part you wrote with your example of a 905 and 900 belt :-).

 

At the end i have to say that i really appreciate each post of you, they always give me some points to think about!

Share this post


Link to post
Share on other sites

Estarriol and Alseye!

 

Thank you so much for your replies, so full of details and "pro" approach to our class! Will read it very carefully to fully understand and will reply shortly!

Very nice to know how you guys excel on Warrior! :)

Share this post


Link to post
Share on other sites
6 hours ago, Allseye said:

I have to partiallly disagree with this. If i'm sitting on low rage and don't got a ip-reduce buff, i try to avoid casting Revenge.

So lets say i'm sitting on 30 rage. What will happen if i would cast Revenge? Casting Revenge (lets say with buff for reduced costs) will bring me to 10 rage -> no rage to even cast a smaller IP. If i feel necessary to cast IP, then i cast a 30 IP and gain effectively survivability (because it was necessary). But if i simply cast Revenge to obtain the buff, this will lower my flexibility in using IP and this will lead to less survivability. This is why i recommend to  cast Revenge not until 58 rage. Ok, we won't gain the most benefit of vengeance...but i'm interested in most survivability and not in some fancy values in wrong situations.

This doesn't contradict what I was saying at all. It is a basic tenant of tanking that if one has a scaling ability wherein it is optimal to wait until a certain value to use it, but you need that ability now, you use it. If you're going to die, you hit IP. End statement. I didn't feel that important to include in regard to a question about the rage breakpoint to hit when using Vengeance. Furthermore, if you were going to choose a breakpoint where you'd use Revenge first in order to ensure you're following it up with an immediate IP, then that breakpoint would be 69 rage. If you need an IP now and don't have the buff to reduce its cost, hitting Revenge first at any point less than 69 rage would net you a lesser shield than just outright using IP.

Additionally, if you're in a situation where you need an IP now and a delay risks your life, the GCD itself may be a consideration. If you'll need to wait for the next GCD to hit Revenge to get the buff up regardless of rage status, skip it and just hit IP.

This, of course, ignores the possibility of a proc and it only relevant in a situation where you have to have that IP now. Realistically, you should largely be able to avoid these "oh, shit" moments.

But, again, my advice for running Vengeance was to make your combos IP-> Revenge after the first in order to keep that IP cost reduction buff up so you could immediately hit IP when needed ↓

21 hours ago, Estarriol said:

That's just my advice for if you have no buffs up. Once you do, use the thing that is discounted first. Personally, I like to use my abilities in combos. If I want an Ignore Pain up, I'll use both IP and Revenge so that I reset back to the same point and don't have to remember or check which buff I have. My preference is to leave it so that my next IP is discounted so that I can use a discounted IP in a snap if needed and revenge when convenient.

6 hours ago, Allseye said:

i'm also using it if i already got the IP-reduce buff if i'm not pooling rage and inc-dmg is irrelevant.

 

In a discussion on how to optimize Vengeance, this is superfluous. If incoming damage is so irrelevant over the course of the fight as a whole, you should be running BV/AM. If you're talking about situations where you're not actively tanking only, then this should be axiomatic.

 

6 hours ago, Allseye said:

Ok, i have to relativze my statement. Ragegain from abilities is significant.

If the contribution to overall rage generation from abilities is significant and haste increases the frequency with which we can use those abilities, then the contribution to rage generation from haste is significant.

6 hours ago, Allseye said:

But keep in mind: you don't find yourself always tanking. I'm not sure if we should compare rage from abilities with RFDT over a whole fight (evtl. this is why you wrote "to the numbers").

I'm honestly not sure where you're going with this. You said rage generation from haste is not significant. It is. If you're trying to make the case that:

On 7/6/2017 at 3:18 AM, Allseye said:

(rage is not that significant because most rage we generate by damage taken).

then we would compare the rage generated from damage vs. the rage generated by ability uses. In order to make the case, I cherry-picked a boss fight where I was taking full or nearly full damage for the whole fight. This fight is Harjatan mythic. Regardless of whether or not you're actively tanking on that fight, you are taking extremely significant damage. That means this fight is pretty much the best case scenario to support your statement that most of our rage is generated by damage taken and it notably failed to support that statement. Abilities were still a far higher contribution to rage generation than RFDT.

6 hours ago, Allseye said:

i wont change my recommendation to not enchanting and gemming versatility

I don't think my statements contradicted that at all. Haste is our best secondary point-for-point. I was clarifying that versatility is not is not "much worse than haste/mastery" and why the example using the amount of stat points to get 1% is not a good metric. Versatility is significantly less valuable than haste, but it is equal or nearly equal to mastery and is overall a valuable stat. However, I agree that haste gems and enchants are the way to go. The breakpoint at which you wouldn't want to do so is 45% haste, which is not realistically reachable right now, so the recommendation stands as enchant/gem haste, regardless of how much you have right now.

6 hours ago, Allseye said:

The second thing what you pointed out is not sacrificing ilvl due to crit. So, on ring, neck and trinkets there should not be crit at all.

Yeah, I left that out due to the lazies.

On 7/5/2017 at 4:27 PM, Sukata said:

Confirmed weeknees on Inquisition mainly due to Scythe Sweep. Found myself getting way more damage than the other tanks i came across on Pugs (mainly Guardians tbh) and using Cds somehow earlier than i was expecting. On Normal, despite taking more damage then the Pally with, never really felt in danger.

Okay, now to the actual logs. I'm going to look at Inquisition first.

Given that you said Scythe Sweep hits you hard, I took a looksee at that first. You were hit by the ability 3 times and have no blocked hits. That's pretty much your smoking gun right there. You took 5.06m (before absorbs) from that ability, but if you had prioritized having Shield Block up for those hits, you would have taken a maximum of 3.1m (before absorbs). You also absorbed 3.5m of the damage you did take from Scythe Sweep. If you had used SB, then your health lost due to that ability would have been zero. Given that you were running Heavy Repercussions, Shield Block would have been up most of the right with proper use. Making sure you will have either SB or Neltharion's Fury up for Scythe Sweep is very important - more important than overall uptime. That said, I reiterate that with HR, you should be able to have it up for most of the fight, so you shouldn't have to worry about saving it. If Scythe Sweep is coming up and you don't have a SB charge, then just pop NF.

I know you know that your Shield Block uptime is low, but I do really want to emphasize your usage. Your uptime on this fight was 13.98% and you used it only 5 times total. Your kill was 4:17 long. You could have had about 60% uptime even without running Heavy Repercussions. With your usage, you are simply not getting anything from that talent. Further, to put that 13.98% uptime in perspective, you had a higher % uptime on Demo Shout than you did on Shield Block. 13.99% to 13.98%. Your uptime on your neck enchant was 15.88%. Your uptime on Concordance was 32.51%. I know I'm hounding this, but this is huge. Getting that Shield Block uptime up is going to give you a huge boost to both survivability and damage. Don't let charges of SB go unused. You get a minimum of 38% less damage taken from melees and 30% (or 60% with HR) bigger Shield Slams.

Another big thing, that Allseye already mentioned, but I want to give attention to as well, is your use of Revenge. If someone is using Vengeance, I would expect to see their number of Revenge uses be equal to or greater than their number of Ignore Pain uses. You had 5 Revenges and 37 IPs. So you're not really utilizing Vengeance either. This is another source of your low damage output. Keep in mind that Revenge and Thunder Clap can also proc Shield Slam. You're not losing out on a proc chance by using these abilities. It doesn't look like you're refraining from using Thunder Clap, but if you are, keep that last sentence in mind but also remember that you take 1% less damage from the boss per stack of Neltharion's Thunder which is applied by TC. It's not a huge difference, but it is a contribution.

Also notable is that you took a LOT of damage from Bone Saw. You should be kiting that ability. The biggest concern there is the amount of torment you take from it, but you also took 2.5x more damage from standing in Bone Saw than you did from Scythe Sweep.

Overall, your CD usage looked fine. You may have been able to get in another use each of Shield Wall and Last Stand, but I wouldn't expect it.

At your ilvl, I would expect to see an average IP size of closer to 1mil, your was a bit shy of 800k. It's not too far off, but it's a thought to consider, especially when using Vengeance.

On 7/5/2017 at 4:27 PM, Sukata said:

The other, or perhaps the biggest weakness i detect in  myself on Normal mode, was the felclaws on Kiljaeden. If you observe the logs we wiped twice mainly because of me not holding Felclaw damage. We managed to kill him but...disliked way too much the damage my warrior took in comparison with Paladin. Cant really say if it is meant te be like that due to the diferrences between the 2 classes or if it was my fault as i think.

Shield Block is the #1 reason here. You have Shield Block falling off and staying off during Felclaws and that leads to death, no question about it. You also took Felclaws about 8 times and only had 3 NF casts. I try to use it on every Felclaws set I take. I wait until there's 4 seconds left on the Felclaws buff on the boss and then pop it. He often doesn't melee in that last second of the buff, so this catches the last two most of the time and it is a huge boost to survivability.

Your Shield Wall casts didn't really seem to line up with Felclaws well. I also think you could have gotten at least one more Shield Wall in with your Felclaws. You Last Stands did seem to line up well. Your Demo Shouts seemed to be used on CD rather than aligning with Felclaws. I'd recommend saving it for that ability as it also provides the extra rage generation from TC and Shield Slam, which is great during Felclaws.

Speaking of rage and Felclaws, I do think I should emphasize that if you can do the Vengeance combo safely during Felclaws, great. If not, don't worry about it, hit that IP!

Again, Shield Block uptime and Revenge usage are two of your biggest factors here and I suspect that'll be the case overall. So focus on getting that uptime up as high as you can on every fight and on keeping your Vengeance combos rolling (if using Vengeance). This is provide you the most benefit for both your survival and your damage.

Overall, I think your CD usage looks fairly solid. There are cases here and there where you could align your CDs better, especially NF. And in NF's case, keep in mind that it blocks 76% of any blockable ability. It is by far best used to mitigate specific abilities in raids.

  • Like 1

Share this post


Link to post
Share on other sites

Before giving some short remarks and "defending" my input: your analization is great, it's awesome how you read the logs. 

10 hours ago, Estarriol said:

This doesn't contradict what I was saying at all. It is a basic tenant of tanking that if one has a scaling ability wherein it is optimal to wait until a certain value to use it, but you need that ability now, you use it. If you're going to die, you hit IP. End statement. I didn't feel that important to include in regard to a question about the rage breakpoint to hit when using Vengeance. Furthermore, if you were going to choose a breakpoint where you'd use Revenge first in order to ensure you're following it up with an immediate IP, then that breakpoint would be 69 rage. If you need an IP now and don't have the buff to reduce its cost, hitting Revenge first at any point less than 69 rage would net you a lesser shield than just outright using IP.

Well, it was just an exemple why we should take care about how much rage we got (and i think i contradicted the statement about regardless using of Revenge :-)). I absolutely agree, if wedont have the reduced-Revenge-cost-buff, the breakpoint is 69 rage. I forgot to mention that. As you like we can kill this discussion at that point, i think we both agree that weaving Revenge is a must with vengeance :-).

10 hours ago, Estarriol said:

In a discussion on how to optimize Vengeance, this is superfluous. If incoming damage is so irrelevant over the course of the fight as a whole, you should be running BV/AM. If you're talking about situations where you're not actively tanking only, then this should be axiomatic.

Yes, thats true, nothing to defend :-)

10 hours ago, Estarriol said:

If the contribution to overall rage generation from abilities is significant and haste increases the frequency with which we can use those abilities, then the contribution to rage generation from haste is significant.

I'm honestly not sure where you're going with this. You said rage generation from haste is not significant. It is.

 

18 hours ago, Allseye said:

Ok, i have to relativze my statement. Ragegain from abilities is significant. But keep in mind: you don't find yourself always tanking. I'm not sure if we should compare rage from abilities with RFDT over a whole fight (evtl. this is why you wrote "to the numbers").

I already admitted that i was wrong. If you took a suitable fight for the comparison, perfect.

 

Share this post


Link to post
Share on other sites

I think I was just confused, then. 

If you want, I could try to go a bit more in depth on the process I use when reading logs. Took me a while to learn what to look for and how to use the program, but I think it's fairly easy to communicate. 

Share this post


Link to post
Share on other sites
2 hours ago, Estarriol said:

I think I was just confused, then. 

This hasnt to be the case. Actually, english is not my nativ language. Sometimes it feels difficult to express something in details. 

2 hours ago, Estarriol said:

I think I was just confused, then. 

If you want, I could try to go a bit more in depth on the process I use when reading logs. Took me a while to learn what to look for and how to use the program, but I think it's fairly easy to communicate. 

Oh, this is a nice idea. Eventually it would be a usefull sticky thread. I would say i got decent knowledge of reading logs but there is space for improvement :-) 

Share this post


Link to post
Share on other sites
4 hours ago, Allseye said:

This hasnt to be the case. Actually, english is not my nativ language. Sometimes it feels difficult to express something in details.

I think this is just an interpretation error on my end. Your English is fine and I have no room to judge someone on their second language when I only speak enough of another language to ask for a beer.

Share this post


Link to post
Share on other sites
On 7/5/2017 at 0:24 PM, Allseye said:

If you're afraid to use it because some guys stated that "swing-timer" playstile...just ignore it. What they did is a theoretical mess which doesnt work fine at all and cause worse performance (plz dont ask for a proof). 

I've been browsing the web trying to find some information on this very statement.. I went to update my WA swing timer & saw a note from the maker stating that the swingtimer was proven insignificant to rage gain after sims... Is this really true? So it's safe to take devastator & not worry once about a swing timer?!?!

Share this post


Link to post
Share on other sites

I can't give you a prove that using a swing timer doesn't lead to more survivability...but i haven't seen a prove that it does either. The reason why it should has always been very theoritical and was based on assumptions.

Just think about the downside the swing timer creates. Its not just the loss of dps and rage by delayed thunderclaps, its also the loss of attention to mecanics. Its getting difficult if you have to focus on a swing timer during while paying attention to fight, timers, other WA displays, rotation and so on. If you're doing wrong, the loss is even greater. This is something the maker of WA swingtimer hasn't considered. You see, this is why i stated it is very theoritical and not really proven... 

Edited by Allseye

Share this post


Link to post
Share on other sites
On 7/31/2017 at 1:14 PM, Gevo said:

I've been browsing the web trying to find some information on this very statement.. I went to update my WA swing timer & saw a note from the maker stating that the swingtimer was proven insignificant to rage gain after sims... Is this really true? So it's safe to take devastator & not worry once about a swing timer?!?!

I believe the maker of the WA stated that delaying gave little return or even leads to losses only in the warrior discord. From what I remember, there wasn't really a deep discussion and tons of information given. However, you can confidently take it to be true that delaying ability uses isn't worthwhile. Even the theoretical benefits with unrealistic assumptions, as Allseye stated, were minor. Thus the reality is that you do not want to delay. 

Share this post


Link to post
Share on other sites

Thank you two so much for your response.. I absolutely hated devastator because you couldn't play without it but having the play with a swing timer was detrimental to my rotation & awareness, just as you stated in your response. So, it's good to hear the timer isn't needed, although it may have never been needed in the first place.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...