Savage Defense Nerf Discussion

85 Tauren Druid
3190
So I was willing to give the changes to Savage Defense a shot, I thought that the way it was seen on the PTR might just be because it was the PTR and it would be corrected (giving Blizzard a lot of credit here, especially concerning druids). But in practice, I'm seeing that savage defense is significantly worse on live than it has ever been in the past.

In comparing SD now to previously it appears the damage absorption is much much lower overall than it has ever been. I went through whole Atramedes fights where SD absorbed maybe 10% of the damage that it used to, or much much less...

It was far more apparent on Chimaeron. Now, not that I'm saying this is right, but simply using it as an example, I used to rank above the healers in the raid simply because of how effective SD was because we'd basically always have full absorption and it was up pretty much all the time.

Any thoughts on this? As I can tell we're now taking much more damage than most other tanks in the same tanking situations. I've compared logs on WoL and now vs pre-patch in a similar atramedes heroic fight I've taken almost 1 million more damage in almost an identical time period.
Edited by Nareena on 5/5/2011 10:15 AM PDT
Reply Quote
90 Night Elf Druid
CFT
10670
Wtb combat log showing before and after.

Also Chimaeron is a terrible example.
Reply Quote
85 Tauren Druid
4010
I am confused by your statement. SD absorption was changed to use the full shield before disappearing, instead of being lost after 1 attack regardless of damage. This was made so that aoe-mitigation was stronger. This made no change to bosses, as their attacks would go through the shield immediately.

I guess I don't understand what you mean?
Reply Quote
85 Tauren Druid
3190
The issue with combat log showing before and after is WoL simply counts the number of times SD procced pre-patch, however it didnt have an absorption number. Post-patch it actually shows how much damage was absorbed. However, in comparing damage taken with my gear significantly better than pre-patch I'm taking much more damage as I mentioned. About 1 million more damage in the same time period on Atramedes.

Go through the shield immediately maybe, but it still reduces the hit by as much as the SD proc is worth.
Edited by Nareena on 5/5/2011 10:21 AM PDT
Reply Quote
85 Tauren Druid
4010
As far as I know, nothing about SD was nerfed. it's still 35% of your AP modified by mastery. You're gonna have to show Arielle some logs.

Hm. Did you recently switch over to total stamina gems? I'm not saying that's the issue, but it would have taken a chunk of your AP out (offset by vengeance in some cases).
Edited by Ssunflash on 5/5/2011 10:25 AM PDT
Reply Quote
90 Night Elf Druid
CFT
10670
1) Where are the combat logs.
2) If you're in Supernova you're posting on an alt or had a name change since 4/28.
3) It might have something to do with going from normal to heroic.
4) Even if I do compare to the previous kill log, the physical damage taken on H-Atra went down.

Show combat logs.
Reply Quote
85 Tauren Druid
3190
This is an alt, I wasnt even paying attention to which character I had selected... Happens when you have 4 feral druids. My main is Beradon. We've killed heroic Atramedes a few times at this point. Unfortunately last night, which was our most recent kill, our regular WoL logger was not online and therefore it was not posted.

I dont appreciate the attitude, I'm simply bringing up something that appears to be the case since the patch, I could very well be incorrect in my observations and admittedly I'm not the best at reviewing WoL information, but I figured I'd come here for verification.

Anyways, logs:

Atramedes Heroic, 2 different kills, 1 pre-patch, 1 post patch.

Pre-Patch:
http://www.worldoflogs.com/reports/osi4jr51h4fxye52/?s=7838&e=8315

Post-patch:
http://www.worldoflogs.com/reports/o9np4dr3sv9nmekq/?s=10272&e=10677
Edited by Nareena on 5/5/2011 10:54 AM PDT
Reply Quote
90 Night Elf Druid
CFT
10670
PW:S and Illuminated Healing amounts were cut in half.

No Divine Aegis either.

Nothing to do with SD.
Reply Quote
85 Tauren Druid
9245
Okay, let's comparalize them:
http://raidbots.com/comparebot/4dc2e6e64b1cfb2970000ddd

One fewer barkskin, one fewer frenzied regen - and TWO fewer SIs. You did no SIs in the second parse at all. That alone would cause you to feel significantly squishier.

Savage defense uptime on the first parse was 38.3% compared to 37.5% on the second. Not a lot to base anything on, there. SD absorbed 500k on the second parse; hard to say how much it did on the first. However, we can make some guesses; PW:S absorbed 520k on the first, 240k on the second, so chances are that's where SD's absorbs are partially. In any case there's no reason here to believe that SD absorbed less in the second one from this; you had almost identical amounts of SD procs and close to the same uptime.

So...let's dig into the logs!

Here are some examples from the first log:
[00:04:20.967] Atramedes hits Beradon 60258 (A: 15750)
[00:04:21.097] Beradon's Savage Defense fades from Beradon
[00:04:23.855] Beradon gains Savage Defense from Beradon
[00:04:25.707] Atramedes hits Beradon 53715 (A: 20055)
[00:04:25.933] Beradon's Savage Defense fades from Beradon
[00:04:26.687] Beradon gains Savage Defense from Beradon
[00:04:27.418] Atramedes hits Beradon 37266 (A: 18471)

So far, doesn't look too weird.
From the second one:
[00:53:52.187] Atramedes hits Beradon 75144
[00:53:52.474] Beradon gains Savage Defense from Beradon (Remaining: 17257)
[00:53:52.474] Beradon's Savage Defense fades from Beradon (Remaining: 0)
[00:53:54.012] Atramedes hits Beradon 56542 (A: 3540)
[00:53:57.457] Atramedes hits Beradon 59741 (A: 8381)
[00:53:59.273] Atramedes hits Beradon 66489
[00:53:59.382] Beradon gains Savage Defense from Beradon (Remaining: 18267)
[00:53:59.382] Beradon's Savage Defense fades from Beradon (Remaining: 0)
[00:54:00.903] Beradon gains Savage Defense from Beradon (Remaining: 18267)
[00:54:02.476] Beradon's Savage Defense is refreshed by Beradon (Remaining: 18267)
[00:54:12.451] Beradon's Savage Defense fades from Beradon (Remaining: 18267)
[00:54:37.385] Beradon gains Savage Defense from Beradon (Remaining: 7407)

This is from later in the fight. Gotta love those 'combat log errors', if they are indeed that. Because otehrwise those just...suck.

This actually worries me a bit:

[00:55:48.039] Beradon gains Savage Defense from Beradon (Remaining: 18267)
[00:55:54.475] Beradon's Savage Defense is refreshed by Beradon (Remaining: 15009)
[00:55:56.929] Atramedes hits Beradon 54922 (A: 21286)

So you gain SD, then refresh it and lose 3k in shield after 6 seconds. My only guess here is that vengeance decay kicked in like a bastard.

There is some weirdness here, but it's tough to say whether it's just from the combat logs or if it's actually bad. What bothers me is that the shields are fairly weak in the second parse; they look to average around 15k, which is pretty low for a raiding bear who is taking that much damage.
Reply Quote
85 Tauren Druid
3190
I appreciate the response, I'll take a look at some of my choices to see what I can do myself
Reply Quote
90 Troll Druid
11445
I would say the low values on SD have to do with all the stam gems. it wouldn't account for a difference pre-4.1 and post but it definitely is making your block weaker.
Reply Quote
85 Tauren Druid
9245
I would say the low values on SD have to do with all the stam gems. it wouldn't account for a difference pre-4.1 and post but it definitely is making your block weaker.
Not really. It'll make pre-vengeance weaker but post-vengeance about the same as agility. The difference between stam and agility as far as SD goes is pretty minor; the main difference is simply uptime on SD. But the actual shield size? It's about the same when hitting vengeance caps.

It doesn't explain why shields look a bit smaller in one parse vs the other.
Reply Quote
90 Night Elf Druid
CFT
10670
This actually worries me a bit:

[00:55:48.039] Beradon gains Savage Defense from Beradon (Remaining: 18267)
[00:55:54.475] Beradon's Savage Defense is refreshed by Beradon (Remaining: 15009)
[00:55:56.929] Atramedes hits Beradon 54922 (A: 21286)

So you gain SD, then refresh it and lose 3k in shield after 6 seconds. My only guess here is that vengeance decay kicked in like a bastard.

He wasn't attacked for nearly 10 seconds during that time, vengeance decay is a likely cause.

05/05/2011 11:30 AMPosted by Fellhoof
Not really. It'll make pre-vengeance weaker but post-vengeance about the same as agility.

No, still worse post-Vengeance.
Edited by Arielle on 5/5/2011 11:43 AM PDT
Reply Quote
85 Tauren Druid
9245
About the same. Not significantly worse. It's certainly not going to be a difference of 20k to 15k in shields between a bear that goes full agi vs a bear that goes full stam.

That being said, nothing actually explains a 3k drop off in shield strength from 6 seconds. It's not 10 seconds, it's 6 - and vengeance doesn't decay THAT quickly in 6 seconds' time. I'm not concerned about the final absorb, just the refresh message. My understanding is that when SD is applied is when the shield's strength is calculated, so it shouldn't matter if they're not hit for 10 seconds or whatever; all that matters is that at second 0, they had a 18k shield, and at second 6 they had a 15k shield reapplied.
Edited by Fellhoof on 5/5/2011 11:59 AM PDT
Reply Quote
90 Night Elf Druid
CFT
10670
05/05/2011 11:58 AMPosted by Fellhoof
it's 6 - and vengeance doesn't decay THAT quickly in 6 seconds' time.

Still 30% of max vengeance achieved in 6 seconds. Or around there. Definitely explains the drop.

However, it does pretty conclusively show that a smaller vengeance-reduced shield can replace a larger one.
Reply Quote
85 Tauren Druid
9245
Yeah, no matter what else I think that's clear - and I wonder if that was the case before 4.1. That would explain shields feeling a bit smaller than before.

It also really means that SD modeling is very optimistic, making SD significantly worse than before. Before this things like Rawr and Tangedyn's spreadsheet assumed SD's value to remain basically capped and basically staying that way, but it's clear that SD isn't going to work that way - and more reasonably that the higher your uptime on SD thanks to avoidance, the lower your actual SD shields are going to be.
Reply Quote
85 Tauren Druid
4010
so we go back to the problem that Katarn posted in his blog a couple weeks back. Avoiding damage lowers our vengeance and makes our shields smaller, and the more me dodge the smaller our shields will be. >.< Is there any reason vengeance should not proc off of dodged attacks? We have natural reaction (although lackluster in its current implementation) because they realized too much dodge was hurting our rage generation. Maybe make it only 4.5% of the AP instead of the full deal, but don't let it drop so quickly.
Reply Quote
85 Tauren Druid
9245
Really, as long as ferals are tied to vengeance for their mitigation and as long as vengeance has these issues, ferals are going to have some odd issues. Vengeance works fine when you're talking about giving tanks some way to do more damage/threat that scales with defensive stats and incoming damage, but when you then link that to mitigation it becomes problematic.

That SD still remains a static boost is similarly problematic.

I'll see if I can add some measure of SD scaling and decay to the sim I wrote. The short answer is that this won't make dodge bad, just slightly worse, but it will make SD significantly less valuable than what we thought.
Reply Quote
90 Worgen Druid
12790
05/05/2011 12:17 PMPosted by Arielle
However, it does pretty conclusively show that a smaller vengeance-reduced shield can replace a larger one.

Time to see how many times I got screwed on Sinestra. :)

http://www.worldoflogs.com/reports/rt-ryzaw2u75zpixz5w/dashboard/?s=2330&e=2779

If you wanna poke around Kalon.

Beginning of P3, I have a Twilight Drake and Sin on me. Midway through P3 I pick up 2 sets of whelps, then tank Whelps + Sin for the second round of whelp deaths. Vengeance is typically fairly high for this fight. If I'm not getting melee'd, I'm getting a breath to the face, so it rarely drops low when I'm tanking her.
Edited by Reesi on 5/5/2011 12:51 PM PDT
Reply Quote
90 Night Elf Druid
CFT
10670
05/05/2011 12:41 PMPosted by Fellhoof
I'll see if I can add some measure of SD scaling and decay to the sim I wrote. The short answer is that this won't make dodge bad, just slightly worse, but it will make SD significantly less valuable than what we thought.

That's just a nice way of saying this behaviour of SD is effing retarded.
Reply Quote

Please report any Code of Conduct violations, including:

Threats of violence. We take these seriously and will alert the proper authorities.

Posts containing personal information about other players. This includes physical addresses, e-mail addresses, phone numbers, and inappropriate photos and/or videos.

Harassing or discriminatory language. This will not be tolerated.

Forums Code of Conduct

Report Post # written by

Reason
Explain (256 characters max)
Submit Cancel

Reported!

[Close]