Thread of Ultimate Stupidity - Page 1968

• Jessamy · 08/30/12, 10:13 PM

Originally Posted by Oriondk

With the 4p t13, wouldn't it be better to go for runic corruption instead as frost dw? The chance to get 710 mastery 4p proc is 15% higher with runic corruption.

There is a conversation about this in the unholy thread. My dummy test is a very small sample size, if you want to contribute more data please do.

mistic · 08/31/12, 1:49 AM

This reset the pacth hit live servers and i was checking some armorys and forums to try to figure out which is the best way to set stats for affliction. I'm sorry in advance if this is not the correct place to post about this doubts,

I saw some people going for full haste and other people going for full mastery, so i came here to ask if someone could tell me which is the best way and if there is a threshold of haste we should be respecting with the new talents.

Another question is, since our affliction mastery got changed, is it still worth to use cunning of the cruel or insignia just became better?

Exoduz · 08/31/12, 3:49 PM

but in both is the molten oder mage Amor (critcap) better than frost amor

abs0lut · 08/31/12, 4:05 PM

edit for deletion.

Muphrid · 08/31/12, 4:34 PM

Originally Posted by Exoduz

but in both is the molten oder mage Amor (critcap) better than frost amor

Yes, the weights would suggest that, but 5% of either stat is a lot--enough that the linearized weights (which are what we have) no longer apply. The result is suggestive, though, and I think it would be valid to sim Molten or Mage armor as a check.

bromli · 09/01/12, 9:35 AM

Originally Posted by Lorayth

You're not just losing J's HoPo generation for one cycle only; by delaying J and using TV prematurely so you can use CS and Exo, you have a J off of cd sitting there doing nothing. The name of the game is to generate as much HoPo through the ENTIRE fight, not during a 9-second window. In the scenario you spoke of above, if you follow the suggested rotation, you do J-TV-CS-Exo-J-TV-CS-Wait-J-CS-TV..., pushing both CS and Exo back 1 gcd, but able to use J twice without delay. Delaying gives you this: TV-CS-Exo-J-CS-TV-Wait-CS-J-Wait-CS-TV. By delaying J, you inefficiently managed J's cd, resulting in two waits instead one.

All of our abilities now generate HoPo; there is simply no reason to wait to use one in favor of a harder hitting ability, because the more you wait, the less HoPo you will end up having overall.

i'm not talking about a 9 second window. oddly, you describe a slightly longer window as evidence. incidentally, it's incorrect. the first gap in the suggested rotation would have TV3 instead. the second gap in the alternate rotation would have TV4 instead.

taking these out a bit further,
suggested on top, alternate underneath:

(EDIT: there is a mistake in these sequences as well. additionally, this sequence describes only the situation where you have 4 HoPo, J is off CD, CS is off CD in 1 global, and Exo is off CD in 2 globals. not sure why i just went with Loryath's list instead of the base case, but i did. please see post below).

J TV5 CS Exo J TV5 CS TV3 J CS x x CS Exo J TV5 CS TV3 J CS

TV4 CS Exo J CS TV5 x CS J TV4 CS x Exo CS J TV5 CS TV3 J CS

compare these two. both cases have 2 open GCDs about to happen. both cases leave the cd on J and CS at the same place. both cases have the same HoPo (2). case 2 has exo coming off CD 1 GCD earlier. both cases have 2 open globals. both cases use 5 TV, 2 exo. case 1 has an extra J, case 2 has an extra CS. we have just turned a J into a CS, with the additional benefit of a slightly sooner next exo, and since exo is 1 spot sooner the entire time, we are allowing for more AoW procs as well.

just to show that i'm not ending the rotation at a convenient place for my theory, here's what comes next.

x x CS Exo J TV5 CS x J CS TV3

x x Exo CS J TV5 CS x J CS TV3

everything is identical here, other than the early exo from the proposed rotation. also, note that the proposed rotation could in fact simply revert to the accepted rotation at that point, as exo and CS come off cd at the same time. this isolates the lists above. doing the proposed sequence above turns a J into a CS, with zero negative impact on the rest of the rotation.

(EDIT: again, there is a mistake in the sequences. my apologies, please see post below.)

obviously, this ignores AoW procs (which favor the proposed rotation), it ignores avenging wrath and sub-20% (which also favor the proposed rotation), and it ignores lv90 talents (which again favor the proposed rotation).

the reason HoW and lv90 talents benefit the proposed rotation is that they consume GCDs. the real thing we're worrying about here is preemptively stopping ourselves from delaying a good HoPo generator (aka not judgment) in favor of TV5. this is a situation which does in fact occur, even in the normal rotation (non-wings, non-execute phase). the more abilities we add to the rotation, the more likely it will be that we become "handcuffed" with TV5 and CS/Exo/HoW available at the same time. delaying J can loosen this, and when you get proccy you can sometimes go for quite a while without needing J to fill the globals.

Arcane36 · 09/01/12, 4:35 PM

I would love to have a BM spec as my main, but currently in DS and overall in raids, I can do more damage in Survival. It may be just the case that I am not used to it. The last time I played BM was on wotlk in Naxx, so returning after such a long time to BM is pretty hectic for me. Otherwise I heard that the BM is best both for dmg and for survivability. I am trying getting used to it in PVP

well with the recent gold change, there is really no reason for me to try and solo the whole Ulduar, even though it is fun. I would love to learn the tactics and try to overcome the obstacles, but without the reward, it is useless for me

So the only thing I am soloing now is Onyxia, Kara and TK. Other mounts I either have, or are impossible to solo (25hc versions)

Paliousxl · 09/01/12, 7:22 PM

Apologies in advance if I missed them, but were there any posted stat weights for 5.0.4 for the 3 specs? I found some rough numbers on other sites, but always use EJ for the most accurate numbers.

Thx in advance

Emphanis · 09/02/12, 4:25 AM

Originally Posted by Paliousxl

Apologies in advance if I missed them, but were there any posted stat weights for 5.0.4 for the 3 specs? I found some rough numbers on other sites, but always use EJ for the most accurate numbers.

Thx in advance

Discard this reply

lightedge · 09/02/12, 4:44 AM

hi
i know
At L85 relative values:
Str > Haste > Mastery > Crit

anyone know
Stat Rating for 1% Value compared to Str for lvl 85?

Durendil · 09/02/12, 5:16 AM

Well, the nerf on survival was hard - I went from 43K dps on ultraxion to 36K (and yes, I had spent time getting used to the new talents). Fun fact, my first 5-minute try on dummy in BM got me to 35K without buffs (used cat for mastery buff).
BM rotation is really close to survival, albeit simpler (spam arcane and save 40 focus for a 6-sec CD signature shot, but no black arrow/LnL), and the burst phases are basically spamming arcane/kill command.

bromli · 09/02/12, 5:50 AM

Originally Posted by bromli

i'm not talking about a 9 second window. oddly, you describe a slightly longer window as evidence. incidentally, it's incorrect. the first gap in the suggested rotation would have TV3 instead. the second gap in the alternate rotation would have TV4 instead.

taking these out a bit further,
suggested on top, alternate underneath:

(EDIT: there is a mistake in these sequences as well. additionally, this sequence describes only the situation where you have 4 HoPo, J is off CD, CS is off CD in 1 global, and Exo is off CD in 2 globals. not sure why i just went with Loryath's list instead of the base case, but i did. please see post below).

J TV5 CS Exo J TV5 CS TV3 J CS x x CS Exo J TV5 CS TV3 J CS

TV4 CS Exo J CS TV5 x CS J TV4 CS x Exo CS J TV5 CS TV3 J CS

compare these two. both cases have 2 open GCDs about to happen. both cases leave the cd on J and CS at the same place. both cases have the same HoPo (2). case 2 has exo coming off CD 1 GCD earlier. both cases have 2 open globals. both cases use 5 TV, 2 exo. case 1 has an extra J, case 2 has an extra CS. we have just turned a J into a CS, with the additional benefit of a slightly sooner next exo, and since exo is 1 spot sooner the entire time, we are allowing for more AoW procs as well.

just to show that i'm not ending the rotation at a convenient place for my theory, here's what comes next.

x x CS Exo J TV5 CS x J CS TV3

x x Exo CS J TV5 CS x J CS TV3

everything is identical here, other than the early exo from the proposed rotation. also, note that the proposed rotation could in fact simply revert to the accepted rotation at that point, as exo and CS come off cd at the same time. this isolates the lists above. doing the proposed sequence above turns a J into a CS, with zero negative impact on the rest of the rotation.

(EDIT: again, there is a mistake in the sequences. my apologies, please see post below.)

obviously, this ignores AoW procs (which favor the proposed rotation), it ignores avenging wrath and sub-20% (which also favor the proposed rotation), and it ignores lv90 talents (which again favor the proposed rotation).

the reason HoW and lv90 talents benefit the proposed rotation is that they consume GCDs. the real thing we're worrying about here is preemptively stopping ourselves from delaying a good HoPo generator (aka not judgment) in favor of TV5. this is a situation which does in fact occur, even in the normal rotation (non-wings, non-execute phase). the more abilities we add to the rotation, the more likely it will be that we become "handcuffed" with TV5 and CS/Exo/HoW available at the same time. delaying J can loosen this, and when you get proccy you can sometimes go for quite a while without needing J to fill the globals.

my mistake was not putting the first Exo ahead of CS in the accepted rotation. both rotations get the same benefit from Exo and AoW, there is no edge for the proposed rotation in this situation, as far as Exo/AoW goes (actually i suspect there is still a slight edge due to the 'handcuffing' effect, but it's not something that's easily demonstrable with a simple spell sequence).

the proposed rotation still turns a J into a CS, however. both rotations still end up at the same place (regarding spell CDs and HoPo available).

accepted rotation on top, proposed rotation underneath:

J TV5 Exo CS J TV5 CS TV3 J CS x x Exo CS J TV5 CS TV3 J CS x x

TV4 CS Exo J CS TV5 x CS J TV4 CS x Exo CS J TV5 CS TV3 J CS x x

this sequence can be truncated after 11 GCDs, the rest is just for show.

bromli · 09/02/12, 7:44 AM

the base case:

you have 4 HoPo, J is off CD, CS is off CD in 1 GCD. obviously, we don't know how long Exo is on CD at this point. in order for this situation to occur, we must have generated the 4th HoPo through some means other than CS (otherwise CS would be on CD for 2 GCDs, not 1) OR we used the previous GCD for a non-HoPo generating ability (lv90 talent, divine shield, sacred shield, selfless healer, etc). since this thread is for level 85, and we are ignoring execute-range and avenging wrath, the most likely scenario is that we generated the 4th HoPo via Exo.

accepted rotation on top, proposed rotation underneath:

J TV5 CS TV3 J CS x x CS Exo J TV5 CS TV3 J CS x x CS Exo J TV5 CS TV3 J CS x x CS Exo

TV4 CS J TV3 CS x J CS TV3 Exo CS J TV3 CS x J CS TV3 x Exo CS J TV3 CS x J CS TV3 x Exo

starting with the first Exo, both rotations repeat infinitely.

the accepted rotation has a double gap, the proposed rotation has 2 evenly spaced gaps (on the 6th and 10th GCD of the repeating sequence). perhaps it's just me, but it seems advantageous to have the gaps spaced out. it could be easier to use non-dps spells in the proposed rotation, and an AoW proc could be less likely to delay other HoPo generators. i offer no proof of this; this is purely conjecture.

the accepted rotation delays a HoPo generator in favor of TV5. in and of itself, this is fine. some HoPo generators will be delayed in some fashion regardless of your priority/rotation. the problem is that when you start proccing, and if HoW is available, you will end up delaying more generators than just CS. the proposed rotation never even uses a TV4 after the first one, let alone TV5. this leeway will be highly beneficial imo, both for quality of life reasons and for DPS.

if you delve into comparing the 2 sequences, you will note that the dps is identical over infinite time. however, real fights are not infinite. either you or the boss will eventually die, and we don't know exactly when that will be.

that in mind, i started comparing the sequences like this: "if the fight ends at x time, which sequence is ahead?" and applied that to each GCD. as it turns out, the only time the accepted rotation comes out ahead is after exactly 6 globals. any other endpoint the best case for the accepted rotation is a tie. the 10th GCD begins the repeating section of the rotation, which ends on the 19th GCD. the third column is the advantage rotation 2 has over rotation 1 at that moment.

1 GCD r2 TV-J
2 GCD r2 CS-J
3 GCD tie
4 GCD tie
5 GCD r2 CS-J
6 GCD r1 -J (this is the only time rotation one is ahead)
7 GCD tie
8 GCD r2 CS
9 GCD r2 TV
---------------------------
10 GCD r2 TV
11 GCD r2 TV+CS-J
12 GCD r2 CS
13 GCD r2 TV
14 GCD r2 CS
15 GCD r2 CS-J
16 GCD tie
17 GCD r2 CS
18 GCD r2 TV+CS
19 GCD r2 TV

ignoring the non-repeating section at the beginning, the worst case for rotation 2 is a 10% chance of a tie. 50% of the time rotation 2 has at least a TV over rotation 1, 80% of the time rotation 2 has at least a CS over rotation 1. conclusion: use rotation 2 unless the boss will die in 3-7 GCDs. even then, the gains are only (2*J - CS) / 5. the gains for rotation 2 on a fight with an unknown endpoint are (.5*TV) + (.6*CS) - (.2*J).

when we start factoring in HoW and AoW procs, the edge for rotation 2 should only grow. in fact, i am starting to question whether the priority list should be TV5>HoW>Exo>CS>TV4>J>TV3. that is a MUCH more difficult question for me at least, as i am semi-retarded when it comes to computers.

Arcane36 · 09/02/12, 8:55 AM

Okay I made a little test on the little dummy. I did not have any buffs or anything. The first one was BM with Dire beast and Lynx Rush. I popped all my CD at the start for max burst and made 49k dps and then it slowly went down to 35k and after a while to 30-32k dps. I was using Spirit beast for mastery buff. My pve gear I used is mostly HC DS and reforged mostly in mastery.

The second attempt was in Survival. I used Skarr from Molten Front for the same mastery pet buff and popped all my CDs at the start for max burst. I used dire beast and Murder of Crows. Well the max i was able to do was 36k dps and then it was stable all the time at 33-35k.

So overall it seems that BM is best because of his burst and because mastery does the greatest damage. The problem comes when I join the raid. I am slowly getting used to BM, but Survival is still best for AOE and since I am used to the rotation, I can make higher numbers even in surv other than BM. But as I said, I am getting used to it, so it gets better every day

Draniia · 09/03/12, 9:35 PM

Does anybody have any solid math confirming that hurricane is better than landslide until leveling?