Thanks alot for your tests, that's really an interesting developement. I wouldn't have guessed that ArP would scale THAT good. I'll keep looking at this. It's really interesting, because a former guildmate mentioned the discussion the rogues had about ArP a few days ago and Ulduar really has some interesting ArP gear.
If it is true that ArP gets better as gear improves overall, then it looks decidedly possible to reach the breakpoint. 800 something ArP is not a crazy amount. And if that falls lower then I'm sure some MM Hunters will eventually pass it.
The big question for me is, is it worth it. So we have found a breakpoint where ArP becomes better, but until then it isn't. So unless it is the best gear we have at that time, going for ArP before the breakpoint might be a really tough job.
I feel obligated to point out that in a real world situation like Vezax or Yogg where you move a lot and ISS procs/JoW are going to be hard to come by, the 7/57/7 max spec performs poorly due to how fast you burn through your mana. You're better off taking all the points out of Imp Barrage and Imp SS and going 5/5 in Efficiency in any situation where you don't have JoW to keep the mana flowing.
Remember to take the spreadsheet with a grain of salt - it only simulates a straight up tank and spank fight.
I would think using an efficiency spec for 3/14 fights (and maybe vezax, although he is a humanoid) that involved taking the points out of Improved Tracking and putting them into Efficiency, would be a better dps trade off. Because Yogg, XT, and Mimiron are all "untrackable," and thus those five points into improved tracking are only spent to go for survival Instincts for these three encounters. So you would kill two birds with one stone, in getting rid of five points useless to these fights and decreasing your viper up-time.
Since looking into the increasing value of ArP yesterday, I've had two thoughts:
1) Was I correct to dismiss the effect of trinket procs on the numbers as negligible?
To test this, I performed the same steps as before with the Best MM DPS profile, but instead of a custom trinket, I used a custom gun identical to the [Furious Gladiator's Shotgun]. This time, the crossover occurred when I had added 374 ArP, 129 points less than in my previous test. Clearly the trinket proc had a greater effect on the numbers than I had originally guessed.
End result:
701 total ArP
Agi = 1.521 dps per point
ArP = 1.521 dps per point
That still means that we would need 374 itemization points on top of existing BiS gear to reach the crossover point. Clearly that's not happening any time before the next expansion. However, just with BiS gear ArP is already the third best stat for us, beating out Crit by nearly a tenth of a dps per point. This led me to my second thought:
2) How much ArP would be needed to make it our second best stat, beating out AP?
I did the same test with the custom gun and found the crossover to occur once 160 ArP had been added.
End result:
487 total ArP
AP = 1.315 dps per point
ArP = 1.315 dps per point
That actually would be within the realm of possibility for Icecrown gear, assuming it will have iLevels 13 points higher than Ulduar gear.
I would like to point out something that many may be over looking before they start dropping stats like agi and crit for more ArP. ArP does not effect magic, only physical, and auto shots and steadies are a very large portion of our DPS.
Obviously you want to stand and shoot as much as possible, but the moment you do have to move the value of ArP is going to drop dramatically because most of the shots it effects aren't going off. Agi, and AP on the other hand effect everything, and crit effects everything except serpent sting dot damage.
Keep in mind the spreadsheet assumes a perfect tank and spank like patchwerk, where you spend the entire fight without moving. In contrast, think about how sacrificing something like crit for ArP would work out on a fight like hodir.
I was wondering how the "supposed" up-coming changes to the ArmPen cap (capping it at 100% armor reduction) could affect the numbers that have been floating about in this thread recently. Is this mechanic change likely to affect these numbers, or (and I may have missed it if it has already been mentioned, for which I appologise if it has) are the numbers from Thalen via the spreadsheet running at less than 100% armor reduction, so in essence the cap wouldn't matter anyway? If it does affect the dps increase from stacking ArmPen, to what extent are we looking at? I don't have access to the spreadsheet sadly, and my indepth theory skills aren't the best either, but would be interesting to see how this change could affect the dps yield from ArmPen.
I was wondering how the "supposed" up-coming changes to the ArmPen cap (capping it at 100% armor reduction) could affect the numbers that have been floating about in this thread recently. Is this mechanic change likely to affect these numbers, or (and I may have missed it if it has already been mentioned, for which I appologise if it has) are the numbers from Thalen via the spreadsheet running at less than 100% armor reduction, so in essence the cap wouldn't matter anyway? If it does affect the dps increase from stacking ArmPen, to what extent are we looking at? I don't have access to the spreadsheet sadly, and my indepth theory skills aren't the best either, but would be interesting to see how this change could affect the dps yield from ArmPen.
It is likely they'll limit the effect of ArP so it doesn't take off more than the calculated cap (as it can do now). Hence the comment that even with 100% ArP you still won't take off all the boss' armor.
The stats thread puts the ArP amount for 100% reduction at 1239, so both of the crossover points I found should be unaffected if a cap goes in. The spreadsheet backs that up, showing armor reduced by about 60% when the AP crossover is reached and by about 80% when the Agility crossover is reached.
That said, keep in mind that reaching the point where ArP beats AP hinges upon the way that Icecrown gear is itemized. Going just by iLevel it should be numerically possible, but we won't really know until that gear is released.
For now, I'm taking 2 things away from this:
1) ArP is not the awful stat it once was. We still don't want to stack it like Agility, but there's no need to deliberately avoid it either.
2) If you have a Grim Toll, hold onto it. Just in case.
The stats thread puts the ArP amount for 100% reduction at 1239, so both of the crossover points I found should be unaffected if a cap goes in. The spreadsheet backs that up, showing armor reduced by about 60% when the AP crossover is reached and by about 80% when the Agility crossover is reached.
That said, keep in mind that reaching the point where ArP beats AP hinges upon the way that Icecrown gear is itemized. Going just by iLevel it should be numerically possible, but we won't really know until that gear is released.
For now, I'm taking 2 things away from this:
1) ArP is not the awful stat it once was. We still don't want to stack it like Agility, but there's no need to deliberately avoid it either.
2) If you have a Grim Toll, hold onto it. Just in case.
Bolded part: Not quite. 12.39 armor pen rating = 1% armor pen, so you have 100% armor pen - but not 100% armor reduction. 100% armor reduction takes 8089/7773.8 ~= 104% armor pen with Sunder/FF and 10643/8625.2 ~= 123% armor pen without.
I'd also like to make a point about item "stacking" since it keeps coming up here and there. The way gear is itemized now, stacking doesn't really mean what it did in the past. You cannot find much gear that is, for example, heavily weighted towards agility at the expense of other base stats or AP (exception being trinkets). All of the gear pretty much has a healthy amount of agi, stam, int (on armored pieces), and AP. Then you will have two of the following: hit, crit, pen, haste. The only exceptions appear to be a few ilvl 200 pieces that have only one of these four (Giantmaim, Leggings of the Stone Hall).
Hit's not going to be dethroned with current mechanics, so most of our "stacking" decisions amount to comparing crit, pen, and haste. We're already making these decisions and most MM hunters are probably finding that they are favorite crit>pen>haste before Ulduar and pen>crit>haste by the end. Comparing crit/pen/haste with agi (or AP) is not likely to occur by any significant amount unless you have a ton of pen or the trauma/mangle affect on piercing shots turns out to be way bigger than I'm thinking. Even then, it would be more of a gemming issue than an overall gearing issue.
Saying that one stat is always > than another is a good way to end up with bad gearing. MM stats need to be balanced. If all you ever looked at was Agility, they you would end up sorely lacking the haste that MM needs. Then again, if you go and stack the hell our of haste you're going to be severely short on crit.
Certain stats, such as hit and haste, need a minimum amount. Getting to hit cap (either with gear or FA) is important. Past cap, hit is useless. MM hunters (in particular ones using Glyph of Chimera) need to have enough haste that any haste affect (quick shots, BL\Heroism, Rapid Fire, etc.) will get you to haste soft cap (SS to GCD). Haste still has value past that, but it loses its priority to things like Crit and ArP.
Per point of item budget, Agi is obviously still our best stat. Crit and ArP are interesting because, for MM, the really support each other. ArP affects all your physical shots. Most of these shots are the ones that trigger Piercing Shots (which only triggers when you crit) so you would be doing yourself a disservice to stack either one at the expense of the other.
Obviously you don't want to overlook raw AP numbers but I find that if you watch these stats and keep them in balance, the AP numbers will basically take care of themselves.
Saying that one stat is always > than another is a good way to end up with bad gearing. MM stats need to be balanced. If all you ever looked at was Agility, they you would end up sorely lacking the haste that MM needs. Then again, if you go and stack the hell our of haste you're going to be severely short on crit.
Certain stats, such as hit and haste, need a minimum amount. Getting to hit cap (either with gear or FA) is important. Past cap, hit is useless. MM hunters (in particular ones using Glyph of Chimera) need to have enough haste that any haste affect (quick shots, BL\Heroism, Rapid Fire, etc.) will get you to haste soft cap (SS to GCD). Haste still has value past that, but it loses its priority to things like Crit and ArP.
What is your reason for arguing this other than the aesthetic appeal of having an entirely GCD based rotation? Yes, soft capping haste increases special shot usage and has an ostensible appeal, but what is the actual value of doing this? I will continue to argue that it is relatively little compared to what you can get from other stats and that haste as it stands now is basically a red herring.
I'm getting roughly the same value for haste (around 0.713, with a variance of about .02) in the spreadsheet for all permutations of 0 haste and soft haste cap, 0.2 latency and 0 latency, and Chimera glyph and no glyph. Crit and pen are coming out way ahead in all of those permutations. Do you have some other metric by which you can justify haste as a more powerful stat?
I would also argue that the statement, "Certain stats, such as hit and haste, need a minimum amount", is unfounded. Hit and haste both scale linearly for auto shot and have an added benefit of increasing proc frequency and RNG based buffs (like IAotH). In neither case is there a required minimum or any kind of threshold in which those stats drastically change except for the hit cap and the haste soft cap, which would more properly be considered maximums. Also, getting to hit cap is not important - it's just that the best use of itemization points is to go with hit until cap. If it took 90 hit rating to get 1% hit it would be a weak DPS stat and we wouldn't be gimping ourselves grabbing hit just because capping is "important."
But what I'm saying is the hit-cap, at current level of point to percent ratio, is one of the BEST ways to spend itemization point for point. Given the current way our stats scale, there is absolutely NO reason to not get close to or up to hit cap (especially if you're using FA). If the point to percent ratio were different we would OBVIOUSLY have to re-evaluate how valuable hit cap is.
I was also not arguing that you should get to soft cap with gear. Getting to about 8 or 9% haste will get you to soft cap with haste affects, that's not an unreasonable amount considering its itemization in current gear. With glyph of Chimera Shot, any time over the GCD directly delays the Chimera Shot, reducing the amount of times Chimera, and thereby all the shots fired after it, will be fired. On top of this, haste buffs our Auto-Shot which, in turn, buffs WQ. Haste is not our top stat for itemization by any stretch, but it is far from a red-herring, especially with more armor-pen buffing our auto-shots.
For instances like this, the spreadsheet is a good start but doesn't always measure properly I believe. Having .2 latency does not mean that the GCD takes .2 secs longer (even though the spreadsheet operates under this assumption). For justification of this statement read back to my haste post a couple pages back on this thread. Haste is just plain better than the spreadsheet gives it credit for when using Glyph of Chimera shot. As I said, once you have enough haste to get to soft cap with haste effects, it starts to lose some of that value but is still not terrible, just no longer a priority.
Also, getting to hit cap is not important - it's just that the best use of itemization points is to go with hit until cap.
I'm confused by your logic. How does that not make hit cap important if getting to hit cap is the best use of itemization points? It seems like you're just arguing with the way I presented the information, not the actual concept.
I know I have asked this before, but never got a real solid answer.
How exactly does latency affect shots and glyphs?
I understand how the spreadsheet devalues the chimera glyph due to 0.02 latency, but I was once told once you get thru your first rotation, the time it takes for skills to be back up is the same. In other words, if you do have the chimera glyph, that the shot itself will be up 9 seconds later. Is that the case, or do you always lose time due to latency? If that was the case, don't you still take off a full second of the shot anyways, or is that second "lost"?
I understand your point about latency and about delaying chimera (and thus your rotation), but I don't see where this was actually measured. You've highlighted the benefits of haste in regards to this, which I agree with, but I don't see anywhere that you were measure the extent to which it does this, which is the key.
How does that not make hit cap important if getting to hit cap is the best use of itemization points? It seems like you're arguing this for the sake of finding an argument.
Because saying it is important to be hit capped is very different than saying hit is our best stat. It is important that your tanks are crit immune, it is important that everyone have XX health for Naj'entus, it is important that your interrupters are hit capped - for some examples. It is not important (for DPS) that you are hit capped if your gearing options require you to sacrifice a lot of other stats.
I understand your point in the way I worded the statement about it. I believe it is mainly a disagreement of definition. If the percent/point ratio was lower for hit rating or, for some reaso, the other stats began scaling better than they do now, the statement that hit cap is important would have to be re-evaluated.
On the note of haste, I'll see if I can work out some numbers that accurately represent the value of haste. The post was anecdotal and, as you said, needs support.
Because saying it is important to be hit capped is very different than saying hit is our best stat. It is important that your tanks are crit immune, it is important that everyone have XX health for Naj'entus, it is important that your interrupters are hit capped - for some examples. It is not important (for DPS) that you are hit capped if your gearing options require you to sacrifice a lot of other stats.
I believe being hit capped fits your description of important. Hunters don't tend to get a lot of jobs that involve this type of play, but it does happen(often when something goes wrong). Just this week my guild was on Ignis and the mage that was supposed to be breaking adds went down, and by the time I got to them there was 2 up with a 3rd on the way. The tank got them into the water, and it was a good thing I was hit capped because right then, I absolutely could not miss.
Another much more simple example is missing on an MD, this could mean a raid wipe in certain situations. Point is, there are game/fight mechanic considerations for being hit capped beyond maximizing DPS, even for us lowly hunters.
I thought this was a bit random but I'll be reporting on it a little later:
I was testing on the target dummy as MM (7/57/7) with kill shot, serpent sting, and steady shot glyphs.
My initial test was with the following rotation:
Serpent Sting + Silencing Shot -> Chimera Shot -> Arcane Shot -> Aimed Shot -> Steady Shot (repeat)
I tried this rotation about 5 times.
My second test used the following rotation:
Serpent Sting + Silencing Shot -> Chimera Shot -> Steady Shot (until Chimera was up again).
I tried this rotation about 5 times.
After taking a look at recount I noticed that the second rotation was consistently 400 or more DPS higher than the results recorded on the first.
I was using 3 peice 7.5 1 piece 8.5 on the tests; so I can only image that 4 piece 8/8.5 would make the gap wider. I was using a wasp as well.
I'll see if I can post back later with WWS numbers for comparisons; but does anyone have any idea what could account for such a large difference?
I thought this was a bit random but I'll be reporting on it a little later:
I was testing on the target dummy as MM (7/57/7) with kill shot, serpent sting, and steady shot glyphs.
My initial test was with the following rotation:
Serpent Sting + Silencing Shot -> Chimera Shot -> Arcane Shot -> Aimed Shot -> Steady Shot (repeat)
I tried this rotation about 5 times.
My second test used the following rotation:
Serpent Sting + Silencing Shot -> Chimera Shot -> Steady Shot (until Chimera was up again).
I tried this rotation about 5 times.
After taking a look at recount I noticed that the second rotation was consistently 400 or more DPS higher than the results recorded on the first.
I was using 3 peice 7.5 1 piece 8.5 on the tests; so I can only image that 4 piece 8/8.5 would make the gap wider. I was using a wasp as well.
I'll see if I can post back later with WWS numbers for comparisons; but does anyone have any idea what could account for such a large difference?
I don't think I even have to go try this for myself, because unless you fucked something up what you are saying is that steady shot does more damage than arcane or aimed shot which is definitely not the case. My vote is on RNG having fun with you.
I don't think I even have to go try this for myself, because unless you fucked something up what you are saying is that steady shot does more damage than arcane or aimed shot which is definitely not the case. My vote is on RNG having fun with you.
That could be a possibility. I would assume that it would be crits and Piercing Shots going off, but even then, not having Aimed would be a huge loss in potential damage. I could understand Arcane not being used, since it doesn't proc PS, but Aimed as well?
More info is needed to see how exactly Mandy is getting higher numbers.
I thought this was a bit random but I'll be reporting on it a little later:
I was testing on the target dummy as MM (7/57/7) with kill shot, serpent sting, and steady shot glyphs.
My initial test was with the following rotation:
Serpent Sting + Silencing Shot -> Chimera Shot -> Arcane Shot -> Aimed Shot -> Steady Shot (repeat)
I tried this rotation about 5 times.
My second test used the following rotation:
Serpent Sting + Silencing Shot -> Chimera Shot -> Steady Shot (until Chimera was up again).
I tried this rotation about 5 times.
After taking a look at recount I noticed that the second rotation was consistently 400 or more DPS higher than the results recorded on the first.
I was using 3 peice 7.5 1 piece 8.5 on the tests; so I can only image that 4 piece 8/8.5 would make the gap wider. I was using a wasp as well.
I'll see if I can post back later with WWS numbers for comparisons; but does anyone have any idea what could account for such a large difference?
5 times through a rotation is nowhere near enough times to get accurate numbers for dps. Do 100 repetitions and maybe I would consider this as anything other than RNG.
5 times through a rotation is nowhere near enough times to get accurate numbers for dps. Do 100 repetitions and maybe I would consider this as anything other than RNG.
My mistake; by rotation I mean I used that rotation from full mana to about OOM. I'm not exactly sure what that would equate to.
Like I said, I'll get some WWS parses up as soon as possible. I would have done it by now, but all of the authentication issues have me afraid of logging out to reset my combat logs.
My mistake; by rotation I mean I used that rotation from full mana to about OOM. I'm not exactly sure what that would equate to.
well you have your answer. Aimed shot is much more expensive to use that steady shot, almost 3 times more. and I believe arcane shot is slightly more(my tool tip says SS 189, arcane 252). By using steady over those you lasted longer before going oom, which inflated your DPS. What your seeing is DPS being limited by mana pool and efficiency, not the shots themselves.
Heart of the Pheonix. I personally think every hunter with a fero pet should have this. I find it much more valuable than 3% more pet damage. If you have to rezz your pet more than once in a fight, either your pet management is poor, or the fight is just pet unfriendly and it's becoming more of a liability than an asset to keep it up. HotP also brings your pet back to life with full health/focus.
Another thing to consider - you are basically trading one pet talent for Heart of the Pheonix, as opposed to two talents of your own for improved revive pet.
A third thing to consider - when you are rezzing your pet, it's coming back with 0 buffs. The obvious things to point out here is 1) it's doing less damage than before 2) it's even more susceptible to dying again
As long as you have HotP, I can't think of a single fight where having Improved Revive Pet would actually be a dps gain. For YS, when your pet dies in phase2, leave it dead until phase 3, then use HotP. For Mimiron, you can rezz in between phases, but with good pet management it won't die until maybe phase 4. For the other fights, it's all about pet management. If you slip up and make a mistake, that's what HotP is for.
I was wondering what spec I would use in order to pick up HotP because as stated in this post you only lose 3% pet damage, but i can't seem to find a talent spec that only loses 3% damage. The two options that I have seen thus far are to drop Rabid and 1 point in Avoidance, which is more than 3% damage according to the Spreadsheet, or drop Wild Hunt AND 1 point in Spiked Collar, which is also more than 3% damage.
05/13/09, 8:42 PM
Similar Threads
