Yes I remember this being reported back then too. I remember testing it with a friend of mine and as far as I remember it was bugged however it was not that big of a deal since Noxious Stings were only 3%.
Shortly after I switched to MM and never payed any attention to it anymore. Now that Noxious Stings are 10% we should definitely try to draw some attention.
I noticed that FD lists Haste as the most valuable secondary stat for my SV Hunter Twink (ilvl 383 atm). The SV Guide for Cataclysm claims crit is the best... im not sure what to do now. Crit or Haste ?
Im currently reforged on crit and its workin out pretty good, did 29k on Ultraxion today, which is pretty decent for that kinda gear imo... ^^
Oh, and im a little confused that FD spits out only about 28k for my char with best raid buffs enabled while Simcraft is at roughly 30-31k on a 6m Ultraxion fight.
Shouldnt a theoretical maximum be always higher than an actual fight ? At least Simcraft always was, i never managed to get above simcraft values. Or did i miss some option where i can set movement/encounter properties.
I noticed that FD lists Haste as the most valuable secondary stat for my SV Hunter Twink (ilvl 383 atm). The SV Guide for Cataclysm claims crit is the best... im not sure what to do now. Crit or Haste ?
[...]
Shouldnt a theoretical maximum be always higher than an actual fight ? At least Simcraft always was, i never managed to get above simcraft values. Or did i miss some option where i can set movement/encounter properties.
While it may get buried from time to time, it is a well known and acknowledged fact that haste is overrated in FD (I dare say in any simulation) - the guide is right. You can test it yoursen by reforging on your FD profile from haste to crit (or from mastery to haste, and compare it to a mastery to crit on the same item).
The issue is that FD (and most other simulations) are, because of their nature, not able to reliably reproduce human reaction time, and assume a perfect or near-perfect execution.
As to the theoretical maximum, yes it should be higher than an actual fight - if the theoretical is calculated from the actual realworld situation that the actual fight takes place in, including gimmicks that increase your damage, your specific raidcomposition, timing of phases, timing of bloodlust and your latency (this one is actually quite big, there is abot a 5k difference in my theoretical max as BM from using standard latency and switching it to my actual latency at around 25 ms). If you exceed the theoretical maximum, you probably haven't used the right variables.
Another thing to realize is that the stat table is an "estimate" of the DPS benefit of adding 1 point in a stat. This estimate is smoothed out over a small finite range of the stat as oppossed to be a single point sample.
It is not an evaluation of the value of all of the previous points that you have in each of those state.
As such, as you near a local DPS maximum of a stat, its value for stat + 1 can be higher since you are on a relatively high slope for the stat.
So If I read correctly, mastery is for lazy people and haste is for people with extreme reaction and rotation skills?
You do not read correctly no.
This is about theorycrafting, and theorycrafting with a bit of sense. For SV, haste can theoretically yield the same dps as mastery in perfect execution with no hesitation and perfect timing and absolute decision making. In reality (applying sense) haste is subject to human reaction, decision making process, and most of all, does not apply to anything more than Autoshots and CoS. CoS is also something used less and less the more haste you get, so its the first 100 haste that grants the most benefit, not the last 100 haste, due to switching to increasingly more AS as AS is more DPET than CoS. No human has the extreme reactions and "rotational skill" to gain full benefit from the extreme haste scenarios, since every relevant fight (including Ultraxion) requires decision making - not to mention the procced decisions (LnL) swicthing it up.
Mastery on the other hand linearly buffs all damage except autoshots.
In reality (applying sense) haste is subject to human reaction, decision making process, and most of all, does not apply to anything more than Autoshots and CoS. CoS is also something used less and less the more haste you get, so its the first 100 haste that grants the most benefit, not the last 100 haste, due to switching to increasingly more AS as AS is more DPET than CoS. No human has the extreme reactions and "rotational skill" to gain full benefit from the extreme haste scenarios, since every relevant fight (including Ultraxion) requires decision making - not to mention the procced decisions (LnL) swicthing it up.
Mastery on the other hand linearly buffs all damage except autoshots.
Just a bit of clarification:
- haste also directly affects pet attack speed and pet focus regen
- haste has an indirect effect on your other shots because faster casts on cobra can affect how many other shots you get in over the course of a fight or how much focus you have to use other shots. How large this effect is will obviously vary on circumstances
- mastery does not buff kill shot, the direct damage portion of multishot, or pet damage in addition to not affecting autoshots
The logic behind why we lean towards mastery over haste with SV is sound (although the gap really doesn't seem that significant in most fights) but I wanted to make sure all the points of consideration were phrased accurately.
Concerning the SV haste versus mastery discussion that has been going on.
First, saying that hunters that do not employ the same strategy as the top hunters are lazy is not only unfair but is also incorrect. The top hunters tend to have closer to best in slot gear. As such, their gear has the maximum number of stats on it that allows for some stat options to possibly increase DPS that may not provide increased DPS for people with lower total ilevel of gear.
Thus, maximum DPS options that may work for the best geared hunters may not work for hunters behind the gear curve – even by a little bit. For instance, for my hunter as currently geared (398.6 avg ilvl), the 2300 haste rating level mentioned (actually it is 2265) is not even possible for my character to currently obtain by reforging, regemming, and reenchanting without also losing agi. Getting as close as possible without losing agility is -580 DPS for my character in FD. Thus, not going with that option currently for me and others in similar gear is not being lazy, it is being smart and as effective as my gear will currently allow.
In addition, your argument eliminates the fact that different hunters have different playstyles, such that one build for one hunter may not be ideal for another. Your statement also overlooks the fact that some builds are more ideal for some fights and not for others. There may be a certain build for a hunter that is the best build for a standard fight. However, for a progression fight with which the guild is struggling, the hunter may make short term changes in order to optimize their performance for the mechanics of that fight.
For those that are interested the remainder of the post below the line is some SV rotation theorycrafting for pros and cons of possible SV base rotations, including the 2265 haste rating level rotation previously mentioned.
Now looking at the SV haste versus mastery discussion, let us look at some different SV rotations and the ideal haste amounts for each. This assumes the 10% haste buff and 3/3 Pathing, which all SV hunters should have. The base CoS cast time under these conditions is 1.765s. It also assumes an LnL cycle on every third ES rotation.
At the start of Cataclysm, a base ES-CoSx3 rotation was suggested since CoS did almost as much DPS as AS and since focus balancing was a large concern. We also had relatively less stats on gear with which to play around. To ideally accomplish this, we needed to perform 3 CoSs in 5s. This requires a CoS cast time of 1.666s. To achieve this. 5.914% haste on gear or 758 haste rating was required. This haste level was easy to accomplish early in Cataclysm and allowed good crit and mastery stats. The bad aspect of this approach is that whenever you needed to cast an extra instant cast shot (like BA, SrS, or AS to dump focus) that the rotation either gained 0.666s of dead time or you performed an extra shot pushing out your ES cast.
With changes in the game since the start of Cataclysm, that approach may no longer be ideal for the following reasons:
- We now have better gear available with a larger stat pool. This makes stat key points that were not feasible previously, feasible now.
- The T13 2P bonus makes casting more CoSs to gain focus no longer necessary. In fact, the original base approach discussed above, now wastes over 36 focus an ES cycle. Even with using the maximum DPS LnL cycle of ES-AS-ES-AS, the over cycle between LnLs wastes over 50 focus. Ideally, we should take advantage of the T13 2P and the additional focus.
- The damage difference between AS and CoS has grown a little that coupled with the shorter “cast” time makes casting AS more desired when have the focus available.
- Having an instance cast AS in the base rotation already allows for BA or SrS substitutions when needed without affecting cycle timing
So let us look at other options in a step by step approach.
This should be obvious, but a base rotation just replacing a CoS with a AS, requires even less haste. An ES-AS-CoSx2 rotation should do more DPS than original SV rotation due to the slightly greater damage of AS over CoS. For a tight rotation, you need to perform 2 CoSs in 4s, requiring a 2s cast time. Since the CoS cast time is 1.765s with no haste on gear, this requirement is already met with 0.47s of deadtime in the cycle even with no haste on gear. Any haste on gear makes the deadtime longer. The bad news here is that the cycle is focus negative by over 5 focus per cycle at 0 haste rating. In order to balance the focus on this cycle out, 21.34% haste on gear or 2733 haste rating is required. This situation results in a larger deadtime in the cycle of a little over 1s. When factoring in the LnL procs every third cycle, this cycle becomes viable. At the 5.914% haste on gear case, the ES-AS-ES-CoS LnL case gains almost 21 focus while this base SV rotation only loses less than 4 focus. Hence, the net cycle of 2 of the base rotations and one LnL rotation is positive by 13 focus. Hence, ideally for this cycle, even less haste on gear is required so that less focus is wasted and the cycle has less deadtime. Hence, the extra AS damage over the third CoS and being able to put more stats possibly into crit and mastery can increase DPS a little over the original Cataclysm case.
The next case to explore is what if we just add in an AS instead of replacing one of the CoSs for an ES-AS-CoSx3 rotation. For a tight 6s cycle, this rotation requires 32.392% haste or 4149 haste rating on gear and wastes about 21 focus per cycle. This extra focus per cycle does allow you to use the maximum DPS LnL case of ES-AS-ES-AS which even though it loses about 17 focus at this haste level still results in an overall net focus waste of 25 focus over the period between LnLs. We can throw this exact case out since the haste requirement is too extreme. However, with realizing that we have focus to spare, we can drop the haste requirement some if we do not mind a little delay in our ES casts. Even with no haste on gear, this rotation is still always focus positive with the maximum DPS LnL rotation, although your ES is being delayed by 1.3s per ES cycle at 0% haste. The key here is that this cycle is probably not ideal at any haste level since it always waste focus and results in significant ES cast delays at feasible haste levels. Even at 20% haste from gear, the ES cast delay is over 0.41s.
So with the ES cast delay and the extra focus, the next cycle to try is to replace the third CoS with a second AS, which is ES-ASx2-CoSx2. For a tight ES cycle, you need to perform the 2 CoSs in 3s. This requires a 1.5s CoS cast time, which occurs at 17.682% haste from gear or the 2265 haste rating previous mentioned as an option. This ES base cycle loses almost 23 focus per cycle. This focus deficit can be balanced out by a 50% mix of the ES-CoS-ES-CoS and ES-AS-ES-CoS LnL cycles. With factoring in BA casts, the ES-CoS-ES-CoS LnL cycle will have to be used a little more often than 50% of the time.
Hence, the 2265 haste rating case is an option to consider. It is a tight ES cycle. It is definitely feasible to obtain at BiS gear levels and a little below. It also can optimally utilize your focus under ideal situations with proper shot selection decisions. The potential problem with this haste rating requirement for the average hunter is whether the haste level is even feasible to obtain and if it is what other stats you have to give up to achieve it. A hunter in best in slot gear, can still have a decent crit rating while obtaining this level of haste. For hunters with lower ilevel, the sacrifice in crit may not be worthwhile.
As an example, my hunter has 398.6 average ilevel, which is basically complete DS normal with one heroic piece and the 391 BiS cloak. With no secondary stats to reforge on either trinket or either weapon, my current stat pool when unreforged is:
Hit:628
Crit: 1971
Haste: 1786
Mastery: 1067
Assuming that I can get exactly the 333 hit rating that I am short from mastery, that leaves only 734 mastery rating left available, with most of that being unreforgeable. With being 479 haste rating short of the 2265 and already being pretty limited on reforging my remaining mastery, much of the that haste rating has to come from my crit rating.
In fact, with quickly trying several reforge options, I can at best get to 2027 haste rating from reforging with this gear, which results in only about 0.05s of ES delay, which is not bad. However, FD shows the results as being -340 DPS relative to my current build that stacks crit and has enough haste to still support the minimum desired MM case. Adding the addition about 240 haste rating from my current stat pool is not going to be much of a DPS gain from tightening the rotation up by 0.05s, but it will be a further DPS loss from the loss of that much crit or mastery. Manually moving half of the additional required haste rating from both crit and mastery is another 71 DPS decrease for me in FD.
Since manual stat changes are not allowed in game, I have to obtain that extra haste rating by regemming and reenchanting. I can almost make it to that haste level without losing any agi (end up 25 short). Doing so is an additional -240 DPS for -580 DPS total in FD.
I have now done some considerable SV theorycrafting and now feel myself being pulled more to the darkside since MM currently lags in real play situations.
When looking at possible SV base rotations, one also has to factor in TotH proccs. I'm not sure, if you did this in your calculations Whitefyst, maybe I just overlooked it. My experience is, that these proccs can scramble the whole base rotation. Thus, my "base rotation" is changing on the fly, as you have to react on TotH and LnL proccs.
All in all it is hard for me to think in "static" basic rotations, when it comes to SV (in contrast to MM). That makes it hard for me to evaluate the value of haste. I also think, that haste can sometimes also be valuable in AoE or movement heavy fights, e.g. if you can get in that extra Multishot or finish the CoS-Cast before you have to move or rebuild your focus faster after wasting it during movement phases, where you did not switch to AotF. I have no clue how one can factor in this behavior as I think it is very difficult to model those situations, but may that's just my deficit.
I'm also not sure about the exceptional good required reaction time in the full haste scenario. With the queuing of shots this shouldn't be much of an issue or am I getting something wrong here?
When looking at possible SV base rotations, one also has to factor in TotH proccs. I'm not sure, if you did this in your calculations Whitefyst, maybe I just overlooked it. My experience is, that these proccs can scramble the whole base rotation. Thus, my "base rotation" is changing on the fly, as you have to react on TotH and LnL proccs.
All in all it is hard for me to think in "static" basic rotations, when it comes to SV (in contrast to MM). That makes it hard for me to evaluate the value of haste. I also think, that haste can sometimes also be valuable in AoE or movement heavy fights, e.g. if you can get in that extra Multishot or finish the CoS-Cast before you have to move or rebuild your focus faster after wasting it during movement phases, where you did not switch to AotF. I have no clue how one can factor in this behavior as I think it is very difficult to model those situations, but may that's just my deficit.
You are correct in that I overlooked the effects of TotH. That just embarrassingly shows my lack of SV theorycrafting recently.
One thing that I want to make sure you are aware if you are not used to my MM posts is that when I list a base rotation like ES-ASx2-CoSx2 that I do not literally mean that you should be shooting ES-AS-AS-SS-SS in that order. They are not “static” as you implied. I am just listing the ideal shot selection over the ES cycle. Which of the shots is used at a particular time is flexible and depends on several factors, such as your current focus level, including the effects from TotH (e.g., if focus is low after the ES, cast at least one CoS first). Also other instant cast shots you need to make, such as BA and SrS, take the place of an AS where you would have previously used it that cycle. Furthermore, LnL can occur at anytime. Regardless of these factors, to theorycraft, you have to assume a base cycle and make modifications from that.
Concerning lack of factoring in TotH in my analysis, in the on average point of view, TotH is only a gain of 2.55 to 4.95 focus per ES cycle depending on how many ASs you cast from 0 to 2. This is a gain of about 5 to 10 focus over the 2 ES cycles on average between LnLs. This really does not impact the analysis much. Considering that there is about a 42 focus difference between each successive version of the LnL cycle, an average change in 5 to 10 focus does not change things much. All it means is that you can use the little higher DPS LnL version more often.
From an individual cycle point of view during game play, a lucky TotH streak can save you up to 33 focus if you are using 2 AS and both of them and the ES proc TotH. In this extreme case, it is highly possible to waste some focus unless you are able to replace 1 or both of the CoSs with additional ASs depending on what your current focus levels are at. However, this situation is highly unlikely since in most cases you will only get 1 TotH proc an ES cycle. Regardless, when you get extra focus from TotH and are high on focus, you either cast your ASs earlier to better use that focus or cast an additional AS if possible (that additional AS cast can be during that cycle or during the LnL).
The key is that the base rotations I list are flexible and not hardcoded. They are just the base rotations that are performed in nominal situations (with flexible order of shots) and which are modified accordingly when procs occur.
Haste really has very little impact on SV AoE. Unlike caster DoTs, hunter DoTs are unaffected by haste. Furthermore, the little extra focus from more static haste does not provide much addition focus over a short time period. The significant benefit of haste in AoE situations is the shorter CoS cast times to regen focus, which with the T13 2P providing double the focus per CoS cast, this haste benefit has limited impact.
Between the traditional 5.914% and the 17.682% haste cases being debated, the difference in focus regen from haste on gear over a 6s ES cycle is only 2.91 focus. Hence, that 1804 extra haste rating only allows an additional AS cast instead of a CoS about every 14 ES cycles (or 84 seconds) from focus regen. Thus, over a 5 min fight, it only allows for 3 extra AS for CoS replacements. Assuming about 3.5K damage difference between CoS and AS, the DPS gain is only about 35 DPS from the 3 replacements. Also recall that the focus gain difference over a CoS cast it constant at a little over 25 focus, regardless of the haste level up to the GCD cap. Hence, in AoE situations, the only focus gain from focus regen due to haste is the additional 0.49 focus regened per MS, which requires about 80 MSs to be cast to regen enough focus to cast an extra. Thus, as can be seen, the difference in focus regen due to haste has very little effect on your DPS.
Between the traditional 5.914% and the 17.682% haste cases being debated, the difference in CoS cast time is 1/6s. That means in an AoE situation that you have to cast 6 CoS to gain the additional GCD for an extra MS with the reduced cast time due to haste. Considering that you do not gain any extra focus regen over those 6 faster CoSs, you do not have any more focus to cast the additional MS than you would have at the lower haste level beyond the small amount regened over each MS as mentioned previously. The only difference is that for each CoS cast, each subsequent MS cast occurs 1/6s sooner than it would have at the lower haste level, which is indeed beneficial.
Whitefyst, I believe you missed out on the fact that with the last patch LnL procs do not get used up by AS, only ES can use up LnL procs now. Recently there has been a pretty long discussion about the way to use LnL proc as well and it basically appears that to avoid the ES dot clipping you have to stick a single AS in between the ESx3. For Example ES-ES-AS-ES would avoid dot clipping. This however is only the case for low latency. On higher latency you do not need to have that AS or delay to avoid dot clipping.
While I might agree with you on the rotations, which are really a good starting point in MM, for SV - it really is quite impossible to predict the rotation and has to be done on the fly as you play. LnL procs, as you suggest every 3rd ES cycle and TotH procs, the two combined together and you may only see 1 in 3 cycles using the suggested rotations. That is a really low to base the haste breakpoints on. Quite honestly, I don't really know how else to approach this matter, but the 'rotational' theorycrafting is really not a good starting point for SV.
Recently there has been a pretty long discussion about the way to use LnL proc as well and it basically appears that to avoid the ES dot clipping you have to stick a single AS in between the ESx3. For Example ES-ES-AS-ES would avoid dot clipping. This however is only the case for low latency. On higher latency you do not need to have that AS or delay to avoid dot clipping.
Actually, the latency issue affects whether you need the slight (.1 sec or so) pause between 2 ES to avoid dot clipping, not whether you need to have an AS after either the 1st or 2nd ES (ES ES AS ES and ES AS ES ES are both viable options since AS was removed from LnL). Becuase of the way the dot refresh works, it only extends the last tick, so if you do ES ES, there will be more than one tick left if you do a 3rd ES, regardless of latency, since the partial tick from the first ES has to be added into the duration of the second ES. For example:
0s ES - expires at 2s
1s ES again - adds in the 1s left on the first ES and adds 2s duration, expires at 4s
Now if you ES again at 2s, you will lose a tick, because there are still 2 ticks left on your previous ES, if you add an AS in first, you will get the full ticks. The latency issue affects whether you need to wait till 1.1s to cast the 2nd ES to make sure that there is slightly less than a tick left to refresh instead of slightly over a tick.
Whitefyst, I believe you missed out on the fact that with the last patch LnL procs do not get used up by AS, only ES can use up LnL procs now.
While I might agree with you on the rotations, which are really a good starting point in MM, for SV - it really is quite impossible to predict the rotation and has to be done on the fly as you play. LnL procs, as you suggest every 3rd ES cycle and TotH procs, the two combined together and you may only see 1 in 3 cycles using the suggested rotations. That is a really low to base the haste breakpoints on. Quite honestly, I don't really know how else to approach this matter, but the 'rotational' theorycrafting is really not a good starting point for SV.
I took the updated LnL into account in my analysis but went with the "safer" always having an AS/CoS between the ESs. This is shown by where I reference the different LnL options like ES-AS-ES-AS. I do not show the third ES since that is the one that costs focus and is the start of the next ES cycle since it elicits the CD. If some hunters want to do the LnL cycle differently, it does not change my analysis much since it was more focused on haste effects on the non-LnL cycles.
Concerning your second paragraph, I agree with most of your statement, and tried to indicate that in my previous statement. To reiterate, the listed base rotation options are used for the purpose to theorycraft the optimal haste points. You cannot do the haste analysis without knowing what shots you want to ideally take on average and the ideal cycle time. If you do not make those type of assumptions, you cannot perform the analysis.
Furthermore, as I tried to indicate the base cycle is meant to indicate the rough average cycle you make on none LnL proc cycles and needs to be adjusted to the situation of procs and current focus. I agree that you will only actually perform that base cycle on a small set of cycles. However, when it is all said and done, your ratio of shots taken minus those involved in the LnLs should roughly align with the ratio for the base rotation. The extra focus from TotH should roughly make up for the extra focus cost for BA replacements of ASs. I also prefer to defer my focus balancing decisions when possible to the LnL procs when there is more freedom in deciding to perform either an instant or a CoS to balance focus without hurting your ES cycle timing.
I also think that we pretty much agree on what is going to happen, but I just prefer to have the ideal number and combination of shots in my head for an ES cycle so I can try to make it happen more often than I otherwise would without the goal. This is more of a preferrential play style on my part to help me better make shot selections. It is definitely important to make shot decisions on the fly with taking into account your focus levels and procs. But if you only take into account that information and not the ideal shot combination, you are more likely to have non-ideal cycles. There are many shot points in the fight where you are somewhat in the middle of your focus bar and can either take a focus costing shot or perform CoS. If you are not careful, you can take an extra instant on a cycle than what you had to and end up with wasted deadtime (this is not bad when you are using focus that would otherwise be wasted, but it is bad when you are not). On the other hand, you could take an extra CoS and end up with delaying your ES cast when you did not need to delay it. Furthermore, when you screw up the decision on one cycle, your focus levels will often result in you having to perform the other case to correct focus levels, meaning that unideal cycles often come in pairs - the initial offender cycle and the focus corrector cycle. Thus, for me it also helps to know ideally how many shots you want to take between ESs (3 or 4) and how many of those should be instants or CoS so that when I am not at either end of the focus bar I can make better decisions to increase my overall DPS.
Anyway that is the thought process that I like to use while playing and fully realize that there are other ways to skin the cat and still be effective. Hence, I see nothing wrong with my way to look at the actual SV shot selection decisions but fully recognize that others have different ways of looking at it, with some of them maybe being superior. It does not matter what method you use as long as you can make the right decisions more often and that method works better for your mental playstyle.
Let me illustrate my thoughts more fully in all of the cases that we need to adjust. Although my gear does not current support it, let's use the ES-ASx2-CSX2 base rotation as an example.
Lets assume that there are no TotH or hate procs and that all LnL procs occur immediately after the 2 ES cycle. In this case the base cycle can be performed forever with performing the shots after the ES in an order that balances focus and the shots chosen between the ESs during the LnL are used to help balance the overall focus and the net focus loss of the bas ES cycle.
Now what happens if there is one or more TotH procs? Even if you get 2 TotH procs on a single ES cycle, the about 25 focus just makes up a little more than the focus deficit for that cycle. You can either choose to make no adjustments that cycle and wait until the LnL to use the Maximum DPS LnL cycle or you can chose to cast an AS instead of a CoS that cycle since that roughly balamces the focus and then perform the whatever LnL cycle is necessary to balance the focus.
If you only have a single TotH proc on ES, the same decision process as above applies. If you only have a single TotH proc on AS, then no adjustements are needed in the base cycle.
Any other TotH case is exceedingly rare since with only 6 instants in the 2 ES cycle, the chance to have 3 or more TotHs is less than 5%. But even in the rare cases that it happens, you handle it similarly with a combination of performing ASs instead of CoSs as needed in either the ES cycle or during LnL.
Now let us explore dynamic haste cases. For T13 4P and Bloodlust, with two CoSs in the base rotation, 0.7s is saved in the ES cycle. For unglyphed RF, 0.86s is saved in the ES cycle. Even with normal Starchatcher Compass procs, you save 0.56s. In any of these cases, the best option is probably to perform the additional AS and to make up for the focus deficit during the LnL. With factoring in the increased chance for TotH procs too, taking the extra AS seems like the best option.
Now let us look at LnL procs occurring at anytime, as will usually be the case instead of at the end of an ES cycle. Since LnLs are the situation where you can make decision to balance focus one way or another, it really does not matter that they occur randomly. Your choose what shots to take in between the ESs dependent on your current focus levels and come out of the LnL with sufficient focus to perform the base cycle.
Multiple of the above situations can occur at the same time or within an ES cycle without it really impacting the situation any differently than that discussed above for each of the individual cases.
Thus, the complete philosophy is the base ES-ASx2-CoSx2 cycle on all non-proc/dynamic haste cycles, with possibly performing AS for CoS replacements for TotH procs depending on resulting focus level, performing an extra AS during dynamic haste, and using the shots between ES to provide focus correction as needed during LnLs.
Not only Latency, but travel time can also affect dot clipping. HC Warmaster P2 is a good example. At the start of the fight when downing Goriona to 80% you are mostly at max range, while when she comes down and when downing Warmaster you are very close to minimun range. In that case you´ll need to wait more than .1 sec to avoid dot clipping.
Said that, you can on paper (almost) always be on mid/high focus for LnL proccs. When above 60 focus, if LnL proccs you can do EX AS EX AS which is the highest DPS option out there. Since it is really difficult to have recurrent proccs each cycle, you will have a cycle at least between proccs to pool focus so you can do Ex As Ex As. that is on paper, ofcourse.
Actually, the latency issue affects whether you need the slight (.1 sec or so) pause between 2 ES to avoid dot clipping, not whether you need to have an AS after either the 1st or 2nd ES (ES ES AS ES and ES AS ES ES are both viable options since AS was removed from LnL). Becuase of the way the dot refresh works, it only extends the last tick, so if you do ES ES, there will be more than one tick left if you do a 3rd ES, regardless of latency, since the partial tick from the first ES has to be added into the duration of the second ES. For example:
0s ES - expires at 2s
1s ES again - adds in the 1s left on the first ES and adds 2s duration, expires at 4s
Now if you ES again at 2s, you will lose a tick, because there are still 2 ticks left on your previous ES, if you add an AS in first, you will get the full ticks. The latency issue affects whether you need to wait till 1.1s to cast the 2nd ES to make sure that there is slightly less than a tick left to refresh instead of slightly over a tick.
While you are correct in theory, I will link to a discussion on the wowhead hunter forums where a few people did some testing on the clipping of the ExS dot. Talk me down on SV vs MM in 4.3.2 - Hunter - Wowhead Forums which showed quite consistently (for the dataset) that the final of the three ExS shots (the one not under LnL) will clip a tick off the LnL-ExS applied just before. So even though there is no difference in theory between EAEE and EEAE, the former will pretty consistently loose you a tick.
(Screenshot provided in the wowhead thread)
I really hope this is the right thread, I thought it was maybe too much a question about details to put it into the "Simple Questions - Simple Answers" Thread.
So, first of all I do not play a hunter and have never done so, so please bear with me on abbreviations. We are currently progressing on Madness of Deathwing HC and our Hunter is having trouble proccing Spellweave on the Regenerative Blood adds for some reason. Now for us this is a mere cosmetic thing, because we do have enough AoE to bring them down, but I do understand that he feels it is a problem that he cannot contribute to a part of the fight as much, be it a minor one.
My question is this: What exactly do you have to do as an SV hunter to maximize your damage on the blood adds? I'd really appreciate it if you went into detail on this, because as mentioned I do not have a clue about hunter rotations, spell abbreviations etc at all.
Thanks in advance!
Edit: Just thought I'd mention again that this is by no means the same question as "What would our hunter optimally do with our raidcomp?" This is just about how to approach the blood adds in a way that allows you to top the meters, if you will, and burn them down as fast as you can. Just imagine we needed any last ounce of DPS for the sake of the argument.
Just simply multi-shot and use explosive trap as he should do in any aoe-situation. The Multi-shot, the serpent sting spread with it and the explo-trap all proc spellweave.
Simply put he has 2 choices:
a) Launch an Explosive Trap and then use multishot
or b) just multishot
In our kills (especially the last ones with the ongoing %nerf) the bloods do not live long enough and are not static enough to make use of an explosive trap, therefore simply multishotting is the best solution. Even though it requires only a few ticks on so many targets from the explosive trap to out dps a black arrow, the lost chances for lock and load should make it a net dps loss in comparison to just using multishot.
Pooling some focus might help to fire several multishots in the short burn period of the bloods without having to use cobra shot to regain focus.
Survival hunters with serpent spread are great for aoeing the bloods because there are multiple possibilitiers per blood to trigger spellweave. Our SV hunter is most of the times on of the top dps on the bloods.
There are only 10 types of people... those who understand binary and those who don't.
I've read in a number of places that snakes from the trap are able to proc spellweave (possibly from their attacks or their deadly poison dot), but I'm not sure how to confirm it as snakes are never credited with spellweave damage on world of logs. Is there a way to check the logs more carefully to be sure? Or has someone been able to drop a snake trap during the fight with a controlled environment to test it?
While the bloods going down quickly will lessen the damage of Explosive Trap, keep in mind that each tick still triggers Spellweave on each target, so it may still be more damage to lay down an explosive trap as opposed to just keeping BA on another target.
I'll be more specific with what I meant before. Some of the things in this post is wrong or out dated, in my opinion. One example of out of date is someone was theory crafting still with Arcane Shots between Explosive Shots. They now work as magical dots and won't be clipped if you re-apply after the 2nd to last tick. An example of "Wrong" is the idea of Lich King enchants being better than Cata ones because of the agility. 22 agility is 44 AP and I believe .07% crit. More AP with different specs, but with survival, it's 24 agility for 48 AP, plus 10% AP = 53 AP? Roughly.
I don't recall what the 65 crit added as far as crit percentage, but it obviously applies more crit than the 22 agility does, even though crit doesn't scale multiplicative...ly. Reasoning behind the choice, I believe, is "per point, agility is worth more than crit. So go with an agility enchantment." But even a simulator says the difference is -34 DPS. Real world conditions would favor crit.
This isn't to say crit is better than agility, but we deprive ourselves of crit options with gems. Yellow slot +10 agility bonus? Break it for more agility. So having a crit enchant on cloak balances that out. As far as the 20 Agility to glove vs. 50 haste. That's a different story. Haste stacks mutliplicatively. (apparently not a word) The 50 rating stretches farther. Another point is a secret I'd rather not share.
But I got an infraction from my previous post being to vague. Now I've given some examples. New to the forums, so yeah. Sue me. :[
I'll be more specific with what I meant before. Some of the things in this post is wrong or out dated, in my opinion. One example of out of date is someone was theory crafting still with Arcane Shots between Explosive Shots. They now work as magical dots and won't be clipped if you re-apply after the 2nd to last tick. An example of "Wrong" is the idea of Lich King enchants being better than Cata ones because of the agility. 22 agility is 44 AP and I believe .07% crit. More AP with different specs, but with survival, it's 24 agility for 48 AP, plus 10% AP = 53 AP? Roughly.
According to femaledwarf.com with my current mix of T14 and T14 heroic gear the 65 mastery enchant adds 12 dps over the Cata one. The 65 crit to cloak enchant adds a whopping 16.1 dps over the Cataclysm enchant. I consider both of those to be small enough to be statistically insignificant.
Edit: One thing you may be forgetting is that Survival (and BM) have talents that increase the agility enchants in addition to the raid buffs which increase them.
With damage over time abilities now refreshing without wasting a dot, "interleaving" a cobra shot between procs is less viable. Instead we can more quickly use our explosive shots without wasting charges. Simply spamming Explosive Shot will result in 8 ticks, with 1 lost. Before 4.2 this would have been 7 ticks. If you wait a small fraction of a second after the 1 second GCD, it should be sufficient to get all 9 ticks quickly. This would be 9 explosive shot ticks in about 3.3 seconds assuming you wait 0.1 between each. This is almost always the best usage on single target, although there may be times when filling in an instant cast ability such as kill shot is worthwhile.
I'm not sure what you find wrong here. If you spam 3 straight explosive shots, the second will not clip the first but the third will cause clipping. If you read the last several pages you'll find a discussion on the subject.
Last edited by Tphirey : 04/30/12 at 5:49 PM.
Originally Posted by Shaker
DBA interviews are simple. You have a choice - stick your face into this vat of unidentified chemicals labelled "Hazardous" or run this SQL script that a developer sent you and said "it's totally fine". DBAs are the guys rushing to go bobbing for second degree burns.
I'll be more specific with what I meant before. Some of the things in this post is wrong or out dated, in my opinion. One example of out of date is someone was theory crafting still with Arcane Shots between Explosive Shots. They now work as magical dots and won't be clipped if you re-apply after the 2nd to last tick.
Do you have any data showing how you're correct and everyone here on the last few pages is wrong? If you want to come and say everyone that has been discussing this with data is wrong 24 weeks into the tier that's fine, but please bring a lot of data to back yourself up if that's the case.
Originally Posted by Venekathas
This isn't to say crit is better than agility, but we deprive ourselves of crit options with gems. Yellow slot +10 agility bonus? Break it for more agility. So having a crit enchant on cloak balances that out. As far as the 20 Agility to glove vs. 50 haste. That's a different story. Haste stacks mutliplicatively. (apparently not a word) The 50 rating stretches farther. Another point is a secret I'd rather not share.
What are you "balancing out"? Crit is linear. Haste rating also isn't multiplicative but additive. Haste percentages (Berskering, Rapid Fire, Bloodlust, total haste rating pool all added together) are multiplicative. I'm two pieces off of full best in slot and 65 mastery is a 9.62 dps loss on FD over 20 agility and 50 haste is a 40 dps loss while the cloak is a 20 dps difference (65 crit vs. 22 agi). They are so close it's rather insignificant.
Originally Posted by Venekathas
I don't recall what the 65 crit added as far as crit percentage, but it obviously applies more crit than the 22 agility does, even though crit doesn't scale multiplicative...ly. Reasoning behind the choice, I believe, is "per point, agility is worth more than crit. So go with an agility enchantment." But even a simulator says the difference is -34 DPS. Real world conditions would favor crit.
Do you have reasoning why real world conditions would favor crit? The agility bonus is less subjective to bad RNG because you have a static gain from AP unlike crit. Is it merely your personal reasoning that real world conditions favor crit or do you have anything at all to back it up?
I'm sorry if this question has probably been asked a million times, but here i go
on heroic madness of deathwing, i keep hearing about spellweaving. the way i've heard about it is that I should tab over to the blood ( when next to the tentacles/claw/wing/whatever) and sting it, and just refresh sting with a cobra shot on it every now and then, and that the sting will proc spellweaves to attack the tentacle for max dps.
02/28/12, 5:04 AM
Similar Threads
