Ret Gurthalak nerf

[Simca]

Are you analyzing logs using Gurth only or just overall? Pretty sure many would have switched weapons.

Overall. I'm not overly concerned about the Gurth nerf - it isn't going to break Ret. If they messed up all our other procs though, that would break Ret.

[Huntingbear_grimbatol]

I'm doing a 1000 auto attacks without seal of truth and then 1000 auto attacks with seal of truth, will post results to determine if they've completely put seals out of proc of gurthalak which was the intended nerf.

Something i've noticed is that apperatus and vessel drops off constantly, though that's to be expected with only auto attacks and 8% crit + vessel.
I also did madness LFR one time and had 12k dps less than i did last week, I'll do more tests etc but from what i've seen so far tentacles from gruthalak (normal) does around 6-9% my damage 9% being heavy uptime, 6 being the lowest I've seen today. Pre nerf it was 10% min and 20% max if that's a reliable number to go from.
What's sad is that arms warriors and fury warriors still have 15% + damage from the tentacle and nothing seems to have changed from them.

Have to say it again, why didn't they just slap a 20sec ICD on the proc and make it fair for everyone?

[Ryann]

I'm not sure how relevant how many raids "still" have reta in them are. No guild outside maybe world top 50s would drop a player overnight cuz of a nerf, no matter how steep.

[Simca]

Average fight length between the two weeks is only 80 milliseconds apart (in favor of week one, actually), so there should be very little impact from gear (besides people switching from Gurth to something else in week two) in these statistics due to the method of collection.

Last Week:

• 30 Rets in 27 parses. 1.11 Rets per raid.
• Average effective DPS: 40314.58
• Landslide uptime: 46.52%
• Creche: 17.13% uptime (only 6 Rets used this)
• AoK uptime: 95.05%
• AoK procs around 15.6 times per fight
• Trinkets used: 20% of Rets used Creche of the Final Dragon, 27% used Apparatus of Khaz'goroth, around 10-15 used Vessel of Acceleration, most used Spine trinket, 1 used Rotting Skull (from Stars-TW, and he was the fourth highest DPS in week one), and 1 used Varothen's Brooch (but I removed him from the average DPS from week one because he was screwing it up).

This Week:

• 30 Rets in 22 parses. 1.36 Rets per raid (a lot of these were even the same guilds as the first week, so I don't know where all these extra Rets came from tbh).
• Average effective DPS: 38827.96
• Landslide uptime: 47.95%
• Creche: 16.88% (only 4 Rets used this)
• AoK uptime: 94.18%
• AoK procs around 16.25 times per fight
• Trinkets used: 13% of Rets used Creche of the Final Dragon, 53% used Apparatus of Khaz'goroth, around 5 used Vessel of Acceleration, and majority used Spine trinket.

Comparing:

• The average Ret player who is killing Ultraxion 25H each week lost around 1486.62 DPS.
• Landslide uptime was well within the standard deviation for our sample size. There is no statistically apparent change to Landslide's proc rate.
• Creche's sample size was too small to tell (only 10 Rets used it), but we expected a small nerf, so hard to tell. If it was nerfed due to this change, it isn't by a large enough amount to matter.
• AoK uptime/procs are within the standard deviation, but it is possible there is a small nerf.
• Apparatus of Khaz'Goroth experienced a huge resurgence this week, and Vessel of Accelerations were replaced across the board.

TL;DR:
Trinket procs seem fine or maybe very slightly nerfed in the case of AoK/Creche.
Landslide is unaffected completely.
Average Ret DPS loss is 1.5k, a 3.69% decrease.

[OmegaManHester]

Im unsure if the censure Hotfix was just to the weapon proc itself or all chances to proc from the censure, I compared my last 4weeks of logs, prior to the nerf I maintained 18% art of war procs over 3 weeks. This week post nerf to censure AoW proc was at 10%. Just wondering if I had terrible RNG or ya's have the same issue. On a side note i come up with about an avg. of 6-7k dps loss.

[Simca]

If there is one thing the logs I look at show, it is that three things in particular are ridiculously RNG:

1. Landslide uptime (anywhere from 20% to 70%)
2. Divine Purpose (everyone already knew that though)
3. Art of War (I saw it as high as 55% on one log)

[Butosai]

Average fight length between the two weeks is only 80 milliseconds apart (in favor of week one, actually), so there should be very little impact from gear in these statistics due to the method of collection.

Last Week:

• 30 Rets in 27 parses. 1.11 Rets per raid.
• Average effective DPS: 40314.58
• Landslide uptime: 46.52%
• Creche: 17.13% uptime (only 6 Rets used this)
• AoK uptime: 95.05%
• AoK procs around 15.6 times per fight
• Trinkets used: 20% of Rets used Creche of the Final Dragon, 27% used Apparatus of Khaz'goroth, around 10-15 used Vessel of Acceleration, most used Spine trinket, 1 used Rotting Skull (from Stars-TW, and he was the fourth highest DPS in week one), and 1 used Varothen's Brooch (but I removed him from the average DPS from week one because he was screwing it up).

This Week:

• 30 Rets in 22 parses. 1.36 Rets per raid (a lot of these were even the same guilds as the first week, so I don't know where all these extra Rets came from tbh).
• Average effective DPS: 38827.96
• Landslide uptime: 47.95%
• Creche: 16.88% (only 4 Rets used this)
• AoK uptime: 94.18%
• AoK procs around 16.25 times per fight
• Trinkets used: 13% of Rets used Creche of the Final Dragon, 53% used Apparatus of Khaz'goroth, around 5 used Vessel of Acceleration, and majority used Spine trinket.

Comparing:

• The average Ret player who is killing Ultraxion 25H each week lost around 1486.62 DPS.
• Landslide uptime was well within the standard deviation for our sample size. There is no statistically apparent change to Landslide's proc rate.
• Creche's sample size was too small to tell (only 10 Rets used it), but we expected a small nerf, so hard to tell. If it was nerfed due to this change, it isn't by a large enough amount to matter.
• AoK uptime/procs are within the standard deviation, but it is possible there is a small nerf.
• Apparatus of Khaz'Goroth experienced a huge resurgence this week, and Vessel of Accelerations were replaced across the board.

TL;DR:
Trinket procs seem fine or maybe very slightly nerfed in the case of AoK/Creche.
Landslide is unaffected completely.
Average Ret DPS loss is 1.5k, a 3.69% decrease.

Nice work, just want to say one thing. Personally I went to 410 axe from gurthalak after the nerf and im sure im not the only one. That quells the dps loss by a good bit. gurthalak prenerf to postnerf we lost at the very least 3.2k ish dps

Which even then isnt the end of the world. You should still be able to pull 42-45k in like 396-400 ilvl gear on ultraxion using the yorsaj axe, which is still very good and better than most specs on that fight.(I was switching back and forth on our kill this week and managed to get the kill when i was using gurthalak, but my damage was much higher with the 410 axe)

I think its time to start looking at the positives again instead of the negatives. We are still very good DPS and in the upper half of the meters. Just will have to live with the fact of 416 gurthalak being a sidegrade if nothing changes.

[Tosu]

Because weapons that do 10% of someones total damage with a proc while the legendary staff proc is 3% isnt completely retarded

10% of a mediocre amount. 3% of a ridiculous amount.

It evens out.

[Kenjji]

10% of a mediocre amount. 3% of a ridiculous amount.

It evens out.

Don't forget that beside the prooc, the legendary still has all it's weapon stats which is quite significant, Gurth doesn't.

[Kisko]

The average Ret player who is killing Ultraxion 25H each week lost around 1486.62 DPS

And the weapons used were ... ?
And distance to Ultraxion was ... ?

You did a lot of work here, but since you somehow missed an elephant (weapon used), and you can't check that considered Rets were utilizing Gurth properly (standing close enough), your DPS measurements are flawed.

Uptime measurements should be good, but since some well-known Rets report problems with AoK stacking I would wait on that one too.

[pbad]

I still don't know why everyone is under the impression this SoT was stacking any trinkets from this expansion, it hasn't been. Since we can't go back in time to do testing I went to one of Butosai's logs from Dec (since I know he uses one). Here's some really solid data from the very first boss. It really doesn't get any cleaner than this, a seal and a CS crit right off the bat, and only 1 stack gain. The 2nd stack comes immediately after the CS crit later.

[17:55:34.154] Butosai Seal of Truth Morchok *955*
[17:55:34.154] Butosai Crusader Strike Morchok *30808*
[17:55:34.519] Butosai gains Titanic Power from Butosai
[17:55:38.626] Butosai Crusader Strike Morchok *32730*
[17:55:38.955] Butosai gains Titanic Power (2) from Butosai

And in case you want to argue there's some ICD on gaining stacks here's what happened after he used his first 5 stack. We have the fade event, 1 second later he gains 2 quick stacks, 7 seconds later a seal crit, 3rd stack doesn't come until 6 seconds after that seal crit.

[17:56:00.491] Butosai's Titanic Power fades from Butosai
[17:56:01.035] Butosai gains Titanic Power from Butosai
[17:56:01.427] Butosai gains Titanic Power (2) from Butosai
[17:56:07.071] Butosai Seal of Truth Kohcrom *9512*
[17:56:13.462] Butosai gains Titanic Power (3) from Butosai

SoT hasn't been exhibiting any on melee x,y,or z effects, until gurth came along, since essentially ICC when we had shadowmourne/bryn/tiaj that I can think of. If there's a case that you can prove that it was with actual hard evidence then I will reconsider. Anecdotes like "I feel like it's harder to stack now" don't count. Below is the link to the log I used if anyone wants to check it.

http://worldoflogs.com/reports/rt-y9alevyx4b7qvqsv/

[Ryann]

Not real sure how you can say, with a straight face, in response to a nerf to Gurthalak, in a 12 page thread complaining about how it's effecting our dps, that you spent all this time comparing this week and last weeks numbers and didn't take into consideration weapons used by all of these rets, "Because you're not concerned with the nerf". It's just baffling.

[Simca]

And the weapons used were ... ?
And distance to Ultraxion was ... ?

You did a lot of work here, but since you somehow missed an elephant (weapon used), and you can't check that considered Rets were utilizing Gurth properly (standing close enough), your DPS measurements are flawed.

Uptime measurements should be good, but since some well-known Rets report problems with AoK stacking I would wait on that one too.

Weapon used is irrelevant because even if they stopped using Gurth and started using a different weapon, that's part of the compensation for the reaction of the nerf. You don't compare pre-nerf Gurth to post-nerf Gurth, you compare the best weapon pre-nerf (Gurth) to whatever the best weapon post-nerf is. Comparing Gurth to Gurth is pointless if you're going to replace Gurth anyway.

My DPS measurements are right on - Rets responding to the nerf are part of the nerf DPS measurement. If Blizzard nerfed the damage done by Templar's Verdict by 50%, would you sit there using the old priority with TV high up there or would you use a new one that had TV as a very low priority? You would adapt first and THEN compare pre and post nerf numbers.

Edit: Distance to Ultraxion is slightly more relevant but still not important when you consider that 30 parses including 20+ top 200 Rets is a large enough sample size to eliminate Distance to Ultraxion as a variable. If you want to argue that a bunch of Top 200 Rets don't know where to stand, then that's fine - that just means your numbers are representing less than half of the top 200 Rets with Gurth. So you're speaking for like... 50 people at that point?


And tbh, most "well-known Rets" are overreacting at the moment and reporting typical RNG variances as problems both here and on EJ. That's the whole reason I had to do this in the first place. I doubt there is any issue with AoK.

Not real sure how you can say, with a straight face, in response to a nerf to Gurthalak, in a 22 page thread complaining about how it's effecting our dps, that you spent all this time comparing this week and last weeks numbers and didn't take into consideration weapons used by all of these rets, "Because you're not concerned with the nerf". It's just baffling.

It's like you guys want to compare pre-nerf Gurth to post-nerf Gurth just to say "we were nerfed by 5k DPS" when in reality you can switch weapons and make that nerf much smaller.


Would it have been nice to see pre-nerf and post-nerf Gurth numbers and H Slicer numbers to then use to make comparisons? Yes.
In fact, I'll go back and do it because I'm bored and it isn't hard to look through my browser history after already having done most of the work earlier.

[Simca]

Here is the best way to see the difference and see it properly.

The difference between what? Pre-gurth and post-gurth? That's a pretty awful way to see the difference, honestly, considering the amount of RNG 2 logs gives.

The reason I chose this fight to demonstrate the nerf is because not many kills are recorded which means you can easily see the difference.

That just means the potential sample size is smaller. I guess it is better if you don't want to actually look through a ton of logs.


I'm adding the Region, Server, Guild, Name, Weapon, and Mind Flay uptimes to my spreadsheet and then I'll up the whole thing to google docs and then that will solve this debate pretty much entirely.

Edit: replaced post

[pbad]

It's awful because it's 2 logs. You could find parses that were both pre-nerf with that large of a difference just because of the RNG involved with the weapon. You don't take a 35% crit rat on TV one week, compare it to your 25% TV crit rate the next and say you got a 10% crit nerf. Also, Vodka's rets do a significant amount of aoe in their strat, look at their damage breakdown.

[pbad]

First off, no one is debating that SoT no longer procs Gurth and the net result is a little less than about half the chances to proc, ie half the proc rate. Also, stop saying changed how seals are interacting with procs, it's how it was interacting with 1 weapon proc. Please read my post on the AoK stacking. The reason I mention the aoe is that if you divine storm 20 bloods that's far more chances to proc the weapon than single target. Therefore more procs and more uptime. The problem is that you're presenting the data as yours doing 1.2k eDPS and theirs are doing 4k/7k and then people scream OMG it procs 1/5 as much before.

[Deleted]

New blue post makes me angry.

http://us.battle.net/wow/en/forum/to...71?page=19#373

This is actually what troubles me more, can live with the gurth changes, but just openly admitting they want ret to be subpar compared to other melee specs.
disgusting really

We expect Retribution damage with this change to also end up where we intend it to be.

[Simca]

I'll publish my data in a bit (an hour hopefully) and it should have all the information needed to make any conclusions.

Edit: Around half done.

26/30 Rets used Gurth last week. Their average tentacle DPS was 4137.03. Their average tentacle uptime was 115.67 seconds.

Of the 4 Rets who did not use Gurth, one of them beat the average DPS, while the other three were below the average DPS. This isn't particularly telling since Gurth was so desirable that if you didn't have it, you were probably less geared overall too.

[spookyz]

I'll publish my data in a bit (an hour hopefully) and it should have all the information needed to make any conclusions.

Real, unabated, unbiased, information!?!?!? WE DON'T LIKE YOUR KIND 'ROUND HERE

(ty for being at least one sane voice of reason, instead of the knee jerk reactionary we've all come to know)

[Schattenlied]

I dunno where you read that ret is falling at the bottom... Me and another ret from the guild usually find ourselves first and second on dps meters... Never below the fifth place tho!

I dunno how this nerf will work out, but I honestly haven't seen enough weapon proc's to make it gamebreaking news....


EDIT: Question to anyone who does math: how do thing work out with t13 2p bonus? Is it worth skipping the haste on CS talent? I tried setting up a new rotation prioritizing Judgement and CS but I often found that due to the haste I got, it makes me lose a couple of milliseconds, up to a second on CS.... Any input would be nice!

Just because you're first and second in your group doesn't mean ret isn't middle of the pack, it just means the people you raid with aren't very good (I know that sounds mean, but it's the truth >.>). Or they all happen to be playing the couple of specs ret actually beats.