Remove Advertisements

[DK & Druid - Tank] Looking for input re:Theorycrafting

Ret, Holy, PVP, etc

Moderators: Fridmarr, Worldie, Aergis, Baelor

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby KysenMurrin » Tue Oct 27, 2009 10:36 am

Edit: A lot of this is wrong, and should be ignored.

As a dual-wield DK I did some quick testing on how the Strikes behave when hitting with both weapons. With the talent Threat of Thassarian, each Death Strike, Obliterate, Plague Strike, Blood Strike, and Frost Strike counts as two attacks calculated seperately, so each Strike as DW Frost has two chances to result in a parry.

I am not 100% certain, but Rune Strike behaviour with ToT appears to be that the off-hand strike occurs simultaneously with the main hand swing, resetting the swing timer on the off-hand.

Also notable for rotations is that if the main hand portion of the Strike is dodged, parried, or misses, the runes do not go on their full cooldown (instead, they refresh once the GCD is over). This is the same as for 2H DKs, and it appears that the off-hand Strike is not affecting the behaviour. This will increase the number of parryable attacks in a rotation as the DK can repeat the Strike imediately if it fails.
Last edited by KysenMurrin on Tue Oct 27, 2009 11:17 am, edited 2 times in total.
I don't play WoW any more.
Donnan - Nangun - Kysen - Kysen - Mardun - Timkins

Mostly-Book Blog.
KysenMurrin
 
Posts: 6673
Joined: Thu Jun 26, 2008 6:37 am
Location: UK

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby Splug » Tue Oct 27, 2009 10:43 am

Oh, just to muddy the waters up a bit more - any time a rune-based attack misses, the rotation shifts to re-insert that attack 2 GCD's later, as the runes come off cooldown early in that cycle. The exception is Howling Blast, which goes on cooldown for 8 seconds whether it hits or not. In that case, the runes would be used for Obliterate instead. This may cause propagated shifts in the rest of the rotation due to rune skew. You may want to just not include this factor since on average, it should just delay the same rotation, causing parryable attacks at the same frequency.

-Splug
Active raid character: http://www.wowarmory.com/character-shee ... an&n=Spyte
255 characters is not enough to fit my alts' armories in.
User avatar
Splug
Maintankadonor
 
Posts: 2381
Joined: Tue Oct 23, 2007 10:16 am

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby Splug » Tue Oct 27, 2009 10:48 am

KysenMurrin wrote:As a dual-wield DK I did some quick testing on how the Strikes behave when hitting with both weapons. With the talent Threat of Thassarian, each Death Strike, Obliterate, Plague Strike, Blood Strike, and Frost Strike counts as two attacks calculated seperately, so each Strike as DW Frost has two chances to result in a parry.
This is incorrect. If the main hand connects, the offhand automatically connects as well. If the main hand misses, the offhand attack is never rolled. They are parried or land as a unit.

I am not 100% certain, but Rune Strike behaviour with ToT appears to be that the off-hand strike occurs simultaneously with the main hand swing, resetting the swing timer on the off-hand.
I have not observed an offhand swing timer reset. The offhand autoattacks seem to be unaffected by the mainhand's rune strikes in any way; the damage just occurs. I am much less certain of this than I am of the previous point, and it is worth confirming; it may be an artifact of my weapons having the same swing speed.

Also notable for rotations is that if the main hand portion of the Strike is dodged, parried, or misses, the runes do not go on cooldown. This is the same as for 2H DKs, and it appears that the off-hand Strike is not affecting the behaviour. This will increase the number of parryable attacks in a rotation as the DK can repeat the Strike imediately if it fails.
Runes for missed (or parried/dodged) attacks get a 2 (or was it 3?) second cooldown instead of the normal 10 second cooldown. The attack cannot be attempted again immediately, but will instead be re-inserted two GCD's later. As mentioned, this behavior is exactly the same for either one or two weapons.

-Splug
Active raid character: http://www.wowarmory.com/character-shee ... an&n=Spyte
255 characters is not enough to fit my alts' armories in.
User avatar
Splug
Maintankadonor
 
Posts: 2381
Joined: Tue Oct 23, 2007 10:16 am

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby KysenMurrin » Tue Oct 27, 2009 10:59 am

I'll bow to greater knowledge on the latter two points, however the first was made having observed the behaviour on live just a few seconds prior to posting.

Just to verify, I went back onto my DK and tried single strikes until I observed it again. Screenshot attached shows 1 Blood Strike, Main Hand was parried, Off Hand landed.
Attachments
Blood Strike ToT.jpg
Blood Strike ToT.jpg (35.98 KiB) Viewed 1263 times
I don't play WoW any more.
Donnan - Nangun - Kysen - Kysen - Mardun - Timkins

Mostly-Book Blog.
KysenMurrin
 
Posts: 6673
Joined: Thu Jun 26, 2008 6:37 am
Location: UK

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby KysenMurrin » Tue Oct 27, 2009 11:04 am

On further observation I might not be right about that. I don't believe I've actually seen a single off-hand strike fail to land. Will continue testing.

Edit: I've now seen miss, dodge, and parry on the main hand, all accompanied by a successful off-hand strike. Still no failed off-hand hits.

I apologise for the incorrect post earlier, I obviously jumped to conclusions without looking closely enough.
I don't play WoW any more.
Donnan - Nangun - Kysen - Kysen - Mardun - Timkins

Mostly-Book Blog.
KysenMurrin
 
Posts: 6673
Joined: Thu Jun 26, 2008 6:37 am
Location: UK

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby Ballador » Tue Oct 27, 2009 12:14 pm

theckhd wrote:If my memory isn't too faulty, bears have a fixed base swing timer (as in weapon-independent), but I don't remember what it is. Does anyone has a ballpark estimate of the base swing timer, as well as a realistic estimate of the swing timer after haste buffs are taken into account?


The base swing timer for a bear is 2.5sec, with 3%buff(swift ret/imp mookin) and 20%physical haste it drops down to 2.023sec. I put together a quick BiS in Rawr since Bear Druid gear is just DPS leather gemmed/enchanted for stam. Druids will have somewhere above 300+ Haste in T9 Tank gear making there swing timer around 1.8sec.
Ballador
 
Posts: 17
Joined: Tue Aug 11, 2009 12:12 pm

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby theckhd » Tue Oct 27, 2009 1:12 pm

Thanks for all the information. I'm going to start sifting through it piece by piece to make sure I got it all right.

Since I think druids are mostly nailed down now, let me recap:
B_druid = 1 because both Maul and auto-attacks can be parried
A_druid = 0.5 because you get 160 GCDs in 240 seconds, 40 of which are Faerie Fire, and the other 120 of which will be parry-able abilities. Thus A=120/240=0.5. It's possible for A to go as high as 160/240=0.667 if the druid drops FF for more dps abilities.
druid_swing_speed = 1.8ish after haste buffs and gear.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Call to Arms 5.x, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7658
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby theckhd » Tue Oct 27, 2009 1:20 pm

Based on Splug's comments, I need to calculate the runic power expenditure of Rune Strike. If we have a Rune Strike rate of (Pp+Pd)/bss, then our expenditure is simply 20 RP * 20 sec * (Pp+Pd)/bss = 400*(Pp+Pd)/bss. Assuming that bss is 2, Pp+Pd is about 60 (ballpark guess?), that gives 120 Runic Power on average, leaving only 10-30 RP for Frost Strikes.

So I would guess that you'll get at most around 1 Frost Strike every 20 seconds then, for N=1?


Do I need to do a similar subtraction for the 11+N in 40 seconds Frost cycle, or are we making different assumptions about Rune Strike there?
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Call to Arms 5.x, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7658
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby Splug » Tue Oct 27, 2009 1:39 pm

The math looks correct, but there's two other points to throw in for Rune Strike frequency. First, Rune Strike is an on-next-attack ability. Back-to-back avoids, particularly with a 2h, can frequently result in wasted opportunities since it will be triggered twice before rune strike can get used.

Secondly, an optional talent that frost tanks can occasionally reach at varying levels (usually, there's only enough points available to put 2 points in at most, but I occasionally run specs that do none at all): Scent of Blood is a 3-point talent which triggers on damage intake (or on dodge/parry) which causes your next X attacks to generate 10 RP each, where X is the number of talents invested. So there may be some additional RP generation created.

So there actually may be a bit more RP available, depending on spec, which would be invested into frost strike. At the same time, Icebound Fortitude and Anti-Magic Shell cost 20 each, though AMS may return some runic power as well... so if cooldowns are getting popped, the math all skews for a bit. Since it's an optional talent, you may want to just ballpark it at some quantity of additional attacks per time per talent.

As for figuring out the attacks per cycle, it should just be plug and chug with the formula - RS is still the highest priority use of RP and is off the GCD. The formula you have would work for all three cases, though you may want to compensate for the points above. It's only going to sway it one attack per 20 seconds or so either way, so I'm not sure how worried about it you are.

-Splug
Active raid character: http://www.wowarmory.com/character-shee ... an&n=Spyte
255 characters is not enough to fit my alts' armories in.
User avatar
Splug
Maintankadonor
 
Posts: 2381
Joined: Tue Oct 23, 2007 10:16 am

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby theckhd » Tue Oct 27, 2009 3:16 pm

The way I see it now, many of the DK possibilities fall into a relatively narrow range (i.e. 6-9 attacks per 20 seconds). I have to sit down and do the math for the others, but my guess is that the easiest way to approach this is to calculate a "min" and "max" value for a DK's number of parryable attacks per second. That should give us rough estimates of the upper and lower bounds for expertise's effectiveness for a DK, which is probably enough for the article.

Since most of these numbers are dependent on gear, avoidance levels, and so forth, it doesn't seem to make much sense to generate anything more exact than "50-60% as effective." The goal is to show that it's not completely worthless as an avoidance stat, and to put a rough estimate on it so people can better consider gear upgrades.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Call to Arms 5.x, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7658
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby Splug » Wed Oct 28, 2009 9:38 am

Yeah, that's a reasonable approach. Since there are a wide variety of specs and rotations available, I think that keeping it a bit open is likely the best way to handle it. It compensates for factors like missed attacks and operator error as well, to some extent.

-Splug
Active raid character: http://www.wowarmory.com/character-shee ... an&n=Spyte
255 characters is not enough to fit my alts' armories in.
User avatar
Splug
Maintankadonor
 
Posts: 2381
Joined: Tue Oct 23, 2007 10:16 am

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby Doogiehowser » Wed Oct 28, 2009 10:19 am

As far as I know, the below is the best TPS rotation for our furry tanks.

1. Mangle
2. Feral Faerie Fire
3. Lacerate and/or Swipe (x2)
4. Rinse and Repeat

Every time Maul becomes available, use it. (However, most bears will have macroed Maul into every attack/special use/mount/eat food/log out button on their keyboard). I'd be curious to know if any bears out there have had any kind of rage issues when raid boss tanking. I know I haven't which is why Maul is macroed so heavily.

Lacerate should be used if your stack is less than 5, or if it is about to fall off. Otherwise, Swipe will be your bigger dps/tps move.

This rotation assumes that you do not have Imp. Mangle. If you do have Imp. Mangle, then you prioritize Mangle over FFF when they clash.
Doogiehowser
 
Posts: 722
Joined: Wed Aug 27, 2008 10:38 am

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby amh » Wed Oct 28, 2009 10:40 am

OT: Cheeky faction- and gender-change there, Doogie? :)
I used to play a paladin.
User avatar
amh
Maintankadonor
 
Posts: 3012
Joined: Fri May 02, 2008 6:25 am
Location: Oh hi

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby Doogiehowser » Wed Oct 28, 2009 10:45 am

amh wrote:OT: Cheeky faction- and gender-change there, Doogie? :)


Gah. I would have gotten away with it, too, if it wasn't for you meddling in my affairs amh.

At least someone acknowledges my presence here. Thanks amh. :P
Doogiehowser
 
Posts: 722
Joined: Wed Aug 27, 2008 10:38 am

Re: [DK & Druid - Tank] Looking for input re:Theorycrafting

Postby theckhd » Tue Nov 03, 2009 1:17 pm

Based on Splug's extensive breakdown, and using N=0-2 for frost strikes in 20 sec:
  • Blood - !Dis: 9 every 20, for A=0.45
  • Blood - +Dis : 9 every 20, for A=0.45
  • Frost - !HB !Epi: 7-9 every 20, for A=0.35-0.45
  • Frost - !HB +Epi: 7-9 every 20, for A=0.35-0.45
  • Frost - +HB: 11-15 every 40, for A=0.275-0.375
  • Unholy - +SS: 21 every 60, for A=0.35
  • Unholy - +Dis: 6 every 20, for A=0.3
  • Unholy - !Dis !SS: 7 every 20, for A=0.35

It looks like calculating lower and upper bounds at 0.3 and 0.45 would be appropriate, which is probably what I'll do. If you calculate the mean of all the rotations, you get something around A=0.375, which is coincidentally the middle point of this range.

Edit: Are any of those DW tanking specs, or is there another calculation yet for those?
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Call to Arms 5.x, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7658
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

PreviousNext

Return to Off-specs & Other Classes

Who is online

Users browsing this forum: No registered users and 1 guest


Remove Advertisements

Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 380 on Tue Oct 14, 2008 6:28 pm

Users browsing this forum: No registered users and 1 guest