World of Logs

[Dzt]

WOL has (finally) been updated for SOO.  The parses are now correctly split between attempts.

 

For those of you who don't have a link to the guild profile:  http://www.worldoflogs.com/guilds/33619/

Edited September 15, 2013 by Dzt

[Varibash]

we have far to many people that are doing terrible dps, like even less dps then they were doing in ToT. People need to seriously pick it up, are healers are doing great, same with tanks, it's pure dps that will hold us back at this point. 

 

It's also a mechanics issue, people aren't comfortable with the fights, so their dps will be very low, once they learn the mechanics to where it becomes muscle memory, they will do better. But come tuesday, everyone needs to be on top of their game. I do not want u to get stuck on a single boss for weeks like we did with Tortos in ToT

[Dzt]

A few points on that:

 

 

1.  People need to spend more time learning the fights BEFORE we do them.  LFR comes out next week.  Everyone had the opportunity to at least do the first four bosses in on Flex difficulty. 

 

2.  Read the established boss guides and obtain a level of understanding above cursory.  Learn when to pop cooldowns.  Learn what glyphs and talents can be used for individual fights to maximize output.

 

Link to the boss guides:  https://www.icy-veins.com/siege-of-orgrimmar-raid-wow

 

Each individual boss section has an area specific to every specialization.  This is the section for DPS Warriors:  http://www.icy-veins.com/forums/topic/3051-siege-of-orgrimmar-warrior-dps-style/

 

 

 

 

This is an indication that people have chosen to rest on their laurels.  Competition for raid spots would fix a significant portion of this issue.  This is effectively a recruiting issue, IMO.

Edited September 15, 2013 by Dzt

[Viledead]

I'm going to bet there were a lot of adds not dying fast enough on norushen in 25m?  DPS on essences and manifestations is low (quite a few people aren't switching), which then is also why overall dps is low (debuff for dps on boss before the test cleanse).  

Also correct me if I'm wrong, but Flex has all the same mechanics just not as punishing for mistakes.  And there have been tons of groups clearing flex for the past week, so kind of crazy if people didn't go into one.  For instance, Sha of Pride will be a lot easier if people have seen it before, and norushen will make more sense.

[Lyntha]

We know how to do Norushen, people were just making individual mistakes.  The fight is tuned tightly enough that we couldn't have people dying and people would fail a trial or get meleed by the add or something similar.  A little bit more DPS would be helpful but it wasn't a huge issue.  Just making it through flawlessly was the demand, it's a tough fight.

 

@Bash:  WoL is helpful but take it with a grain of salt.

 

For Immerseus, there is a huge period of burst at the beginning followed by nothing, followed by add burst (And sometimes you'd be able to AOE adds, favoring classes with quick AOE, like shammies), followed by burst, etc.  Classes with burst potential would do phenomenal here whereas dot classes would do poorly.

For Fallen Protectors, multi-dot classes have a big advantage over classes that can't multi-dot.

For Norushen, the guys who go in to do DPS trials first then tunnel the rest of the fight have a HUGE advantage over the people who go later, so DPS doesn't even matter.

 

 

If you want to take a look at valid DPS numbers, take a look at fallen prot but it's still not a perfect representation of everyone's DPS potential and their individual contribution to the raid.  Also, remember that not all of our members are equally geared.  Some people are missing things like set bonuses, legendary cloaks etc.  Compare similarly geared people in the same class or with the same playstyle, not the top dps to the bottom DPS.

 

I'm all for constructive criticism but remember that data alone without knowing how to analyze it correctly is NOT useful at all.

[Varibash]

While yes, peoples gear aren't exactly the same, there is NO reason why our top dps should be SO FAR ahead of our lowest. Now, I am in no way perfect and I have a lot i need to do to increase my dps, but looking at those meters is pretty sad. No reason why some of our dps should be close to the dmg output of our disc priest. And what looks even worse, if 3 of our bottom dps are our officers and GM....

[Mastric]

In our defence, we were only active about 50% of the time

[Lyntha]

While yes, peoples gear aren't exactly the same, there is NO reason why our top dps should be SO FAR ahead of our lowest. Now, I am in no way perfect and I have a lot i need to do to increase my dps, but looking at those meters is pretty sad. No reason why some of our dps should be close to the dmg output of our disc priest. And what looks even worse, if 3 of our bottom dps are our officers and GM....

 

On which boss?  On fallen prot, unholy and gump were both low but they both were dead for a good portion of the fight.  Gump's DPS was very good when he was alive but he was dead for 40% of the fight.  Unholy's DPS should have been higher given that it's a DOT friendly fight, but he was still doing OK when he was alive but he was dead for 18% of the fight.  Mastric was the only one that was too low for his gear and up for the whole fight.

 

It's important to remember that this is the first week.  We're learning new fights, learning new mechanics etc.  Let's see what next week's fallen prot kill looks like and we'll go from there.

 

The other two fights aren't even worth looking at because DPS between characters isn't comparable at all because of the fight.

[Varibash]

I am talking about the fight we are currently slow on dps with, Nourshen

[Lyntha]

Yeah that fight is 100% useless as far as DPS meters go.  For one, there's different realms.  For two, the people who go down first not only do full damage to the boss, they also do extra damage to the adds, compared to the people who don't go down as soon and do half damage to the boss and normal damage to the adds (They'll do 75% less if they soak a void orb then someone else grabs their portal down).  They get a huge dps buff when others don't.  So it's a useless comparison.

[Varibash]

what's not useless is the fact that we didn't have the raw dps output to down him, and we have a massive number of people doing terrible dps

[Anddrong]

Actually the debuff that you have due to sanity is a DPS loss to the boss, it does not reduce any damage done to adds, and since it is an add fight for the most part, some people need to know what to attack and when instead of tunneling. Another thing is some people may not be using the correct rotation for there class and spec which could be why the DPS is as low as it is.

[Dzt]

If you are looking for specific information that can be assessed on the third boss in SOO, I would argue that DPS meters are absolutely valuable.  Evidence of this is clear:  we are hitting the enrage timer.  For the sake of argument, I will be using a single parse for this; unless otherwise specified.  A direct link to the parse in question (10 Man Norushen kill):  http://www.worldoflogs.com/reports/rt-bhxibfuv8s7s975n/sum/damageDone/?s=16415&e=16822

 

Facts:

1. The DPS for this fight is SLIGHTLY skewed because the personal instance (the portals) removes you from the parse while you are inside.  My personal damage is going to be higher because I am the one running the WOL client and the data it is still recorded while in the personal instance.
2. This, however, does not remove the value of a timeline-based parse for this fight. You will note that players flatline once during the fight (unless they die)  this happens when someone enters the personal instance.  The amount of time spent in the flatline section can be used to assess how well someone is doing in there.  If a player is constantly spending the full 60 seconds in the personal instance, they can be singled out for retraining or could be barred from entering and forced to do add duty all fight.  This is a clear leadership call that can be made by proper analysis of the data.
3. Who is killing the adds?  Well, that data is here:  http://www.worldoflogs.com/reports/rt-bhxibfuv8s7s975n/details/107/?s=16415&e=16822 (Essences only - click the "Damage by Actor" tab and look at the section "Damage taken from") and http://www.worldoflogs.com/reports/rt-bhxibfuv8s7s975n/details/114/?s=16415&e=16822  (Manifestations only - click the "Damage by Actor" tab and look at the section "Damage taken from")
4. Who is doing subpar DPS?  OK.  Here is a good way to take that into account that is not AS SKEWED by mechanics.  There will still be some variations on this specific fight due to when specific actors enter the portals.  Go to WoW Progress and look at the guild:  http://www.wowprogress.com/guild/us/skullcrusher/All+That+Rem%C3%A0ins# and go to the bottom and sort by SIM DPS.  It has a section that runs a character's profile through SimCraft and produces what is essentially their spreadsheet DPS.  Make note that this must be done in a short time after the boss fight because upgrades can inflate the SIM DPS; changing the sample data as a result.  My recommendation is to conduct such analysis within 24 hours.  For this, I will use two specific characters:  Anddrong and Phobia because I previously acquired their permission to post this, so please do not interpret this as an attempt to call anyone out.

 

SIM DPS

 

Phobia:  216,782.6 DPS

Anddrong:  213,527.6 DPS

 

WOL Parse DPS  (Parse used:  Fallen Protectors 25 Man Kill http://www.worldoflogs.com/reports/rt-rpwcev56wokq0rjq/sum/damageDone/?s=7635&e=8196)

 

Phobia:  150,938.6 DPS

Anddrong:  181,792.5 DPS

 

% of Total SIM DPS

 

Phobia:  69.6% of SIM DPS

Anddrong:  85.1% of SIM DPS

 

Consider that this data has value because it compares the average of what a player is doing to what they are mathematically capable of.  This effectively adjusts the numbers, relative to the mechanics, because everyone is subjected to the same scenario.  The exception to this, as mentioned above, is special assignments.  The guy babysitting nests on Ji-Kun will obviously be lower than he should be.  Norushen is another example of the exception to the rule specifically as a result of the personal instance.

 

 

Unfortunately, I do not have the time to pick through all the arguments that have occurred in this thread and provide appropriate responses to them all.  I would, however, suggest that leadership take the time to truly understand the analytics that can be gathered from a tool such as WOL and avoid using conjecture to refute scientific data.

 

If any of you need a hand understanding the data or would like to learn more about some of the advanced features of WOL; let me know and we can set a time to get some learning done.

Edited September 17, 2013 by Dzt

[Iammoot]

Dzt dropping the knowledge bomb.

[Varibash]

nice post DZT. Like I said, we just need to have our raiders step it up a bit since Nourshen is  a dps race

[Dzt]

nice post DZT. Like I said, we just need to have our raiders step it up a bit since Nourshen is  a dps race

 

Let me make one flat recommendation then:  look at the various attempt parses for the 25man.  Any DPS who is in the portal for the full 60 seconds needs to be put on add duty. 

 

 

Why - if they are in there for the full 60 seconds, they are not gaining the full benefit from killing the adds.  They are also increasing the time it takes to get back on the boss (a player who kills their adds in 45 seconds will have 15 seconds more uptime  than someone who goes the full duration and will 25% or more damage to the boss than their underperforming counterpart).  This fight has a low enrage timer and those spare seconds are the difference between a kill and a wipe for this raid group.

[Varibash]

we'll take a look at it.

[Lyntha]

Excellent post DZT!  This is exactly what I meant when I said that you need to analyze the data and not just look at it at a glance.

 

The one thing that's excellent analysis is taking a look at how long they flatline because that's how long they're in the trial.  This is VERY relevant information and one piece that we're actually behind on.  Unfortunately, we can't just "Not send them in" because we need to send them all in to make the enrage.  All of our DPS need to be out and at 0 sanity for the burn phase.  However, if we can send them through faster, we not only do more DPS (We get more adds out and the adds take full damage from 75 sanity players - 0 Sanity players do extra damage to the adds too, and all damage the adds take get xferred to the boss) but we also start the burn phase sooner.  It's a win/win.

 

I can't check it now because graphs are disabled but it's worth looking into later.

[Viledead]

FYI the class specific stuff for SoO on icy veins has a lot of BS.  Multidotting adds on immerseus is better then demo?  LOL.  If that's true then aff is even better for norushen as adds live longer.  Icy veins even says affliction or destro is best for norushen over demo.  So...yeah.  odd.

[Mastric]

Both Held & Snug out dps me on it :\

[Lyntha]

It depends on when you go down, again.  The only applicable fight to look at a glance and say "That person did better than me on DPS" is fallen protectors, and that fight is still skewed because you have 3 bosses up the whole time making multi-dot classes/specs shine.  I was just dotting everything up and otherwise doing terribly (as it was a new fight and I was trying to watch and make RL calls) and did almost 150k dps with no legendary cloak or anything as afflic.

[Viledead]

Of course - but there are baselines and if dps tbh is below 100k then a correction/improvement is needed as even with debuffs/etc that is quite low.  If someone just sat there and tunnelled boss with 75% corruption they still would do more than 100k.  

 

So for example if you take the fallen protectors fight linked here from tonight:

http://www.worldoflogs.com/reports/rt-ifgkouebgmxecyw3/sum/damageDone/?s=2475&e=3098

 

Two things stand out - seems that mark of anguish isn't transferring (causing 2 to die), and dps across the board is low - people should be around 170k generally.