This has been answered many times before, but to recap...
At low end Wrath gear (i.e., lots of blues), AP gems may have been slightly better for MM; however, at higher gear levels, agility gems work out slightly better for DPS since we have the 4% agility bonus, you get both AP and crit, and the extra crit procs useful abilities. These include Piercing Shots, GftT, certain trinkets, etc.
In addition to the DPS benefits, agility gems also provide a little extra armor and dodge for survivability.
Here's a comparison of the modified stats:
32 AP gem => 35.2 AP
16 agility gem -> 16.64 agility -> 16.64 AP -> 18.3 AP (this is over half the AP from the 32 AP gem)
.....-> 0.20% crit
.....-> 32 armor
.....-> 0.25% dodge
For my character, gear, and spec, the hunter spreadsheet shows a loss of about 2.6 DPS when I replace an agility gem with an AP gem. This is not a huge difference, but it is better.
The spreadsheet always favors agility over Attack Power and always has, even prior to Ulduar and all through Burning Crusade, but my curiosity is to how the calculation works. For me it seems odd that if you factor in 5 gems (roughly the number of red slots I have open on my gear right now and a nice even number) you're basically sacrificing 84.5 Attack Power for 1% crit plus armor and dodge. I have never found armor and dodge all that valuable in any encounters since raid wide physical damage is usually rare and taking a blow on a boss fight (which is where it really counts since trash deaths are lawl) will usually kill you if you actually are struck by the blow. Armor and dodge are icing on the cake if the 1% crit is really worth 84.5 Attack Power.
No matter how I cut it though, factoring in how attack power affects almost all your shots, I do not understand how the 1% crit is really more valuable. Sure every 100 shots you have a statistically slightly better chance of having one more trinket proc, go for the throat proc, or Piercing Shots proc, but you have to figure that the 1% has to be on a shot that procs Piercing Shots and that the crit that occurs because of the 1% gain happens outside the hidden trinket cooldown.
To be honest, I have yet to meet a Marksman hunter on my server who can out DPS me (not that that's a surprise, my server seems to have none of them) and I keep up with SV hunters just fine. Attack Power really shines as a buff to your pet, your serpent sting, and your Chimera shots. (Most notably SS which is a fire and forget mechanic that you buff and which will almost always proves to be invaluable due to the fact that even during heavy movement fights there it is, rolling along on the target). The 1% crit won't affect your pet's damage or your Serpent Sting. There's nothing that your AP won't honestly affect that should offset the lost of 1% crit in this situation. So I stand stubbornly against some internal calculation on the spreadsheet that gives more value to agility.
The spreadsheet always favors agility over Attack Power and always has, even prior to Ulduar and all through Burning Crusade, but my curiosity is to how the calculation works. For me it seems odd that if you factor in 5 gems (roughly the number of red slots I have open on my gear right now and a nice even number) you're basically sacrificing 84.5 Attack Power for 1% crit plus armor and dodge. I have never found armor and dodge all that valuable in any encounters since raid wide physical damage is usually rare and taking a blow on a boss fight (which is where it really counts since trash deaths are lawl) will usually kill you if you actually are struck by the blow. Armor and dodge are icing on the cake if the 1% crit is really worth 84.5 Attack Power.
Agility is king for a few reasons.
-First, you get a 4% buff to it in the MM tree on top of the 10% buff already there to the AP portion of that because of TSA.
-Piercing Shots basically acts as a 30% buff to crit damage on the shots it affects. On top of the talents that already buff crit damage, this makes having a high crit rating extremely attractive
-If you don't crit, your pet becomes focus starved, cutting into its DPS significantly
Overall, hunters have very high crit damage returns. Crit is a lot more valuable than you think and the spreadsheets calculations show this. Thought I don't take the spreadsheet as gospel when it comes to my spec and gearing all the time they have done all of the legwork for you when it comes to figuring out exactly how the shot dmg is calculated.
Your spec will rely heavily on how you've chosen to gear your character. If the spreadsheet is saying that a 14/52/5 build does better, by all means, try it out; but be wary. The spreadsheet models patchwerk type situations. ideal circumstances. Ulduar is almost NEVER "ideal circumstances". With lots of switching targets your pet ends up doing a lot of running around (which means less time DPSing). Any talents that buff your instants gains a little more value when you know the fight is going to require you to move. Hunters have tended to theorycraft from within this box of ideal circumstances created by the spreadsheet (which is still an amazing tool btw); but we need to take other things into account.
There are small talent choices that vary with gear, I'm currently trying 1 pt in Focused Aim to get up to the hit cap for instance, but generally I've found using Rawr and the spreadsheet, gemming/gear choices seem to be outside talent choices.
Also, I understand the spreadsheet shows ideal circumstances, but I've always played to minimize movement and set myself up to keep my pet on and burn as quickly as possible. For example, Razorscale, there is a lot of movement in his air phases, but the important part is burning him down on the ground, where you have an ideal sit and dps moment.
Hodir is a similar situation, the first few times it felt like a movement fight, but again, short movements to pick up buffs and move out of fire (ice) and keep the pet rolling on Hodir, so I've found the more I do it, the less I'm moving.
Looking at the best DPS in Shandara's spreadsheet, 7/57/7 outperforms 14/52/5, but thats at about 1k more dps, so I guess all I'm saying is that that the Aspect Mastery talent appears to help most of the way through Ulduar, but as I get into 2 and 3 pieces of 8.5 gear, shifting to 7/57/7 probably outweighs Aspect Mastery.
For me it seems odd that if you factor in 5 gems, you're basically sacrificing 84.5 Attack Power for 1% crit plus armor and dodge. I have never found armor and dodge all that valuable in any encounters since raid wide physical damage is usually rare and taking a blow on a boss fight (which is where it really counts since trash deaths are lawl) will usually kill you if you actually are struck by the blow. Armor and dodge are icing on the cake if the 1% crit is really worth 84.5 Attack Power.
No matter how I cut it though, factoring in how attack power affects almost all your shots, I do not understand how the 1% crit is really more valuable. Sure every 100 shots you have a statistically slightly better chance of having one more trinket proc, go for the throat proc, or Piercing Shots proc, but you have to figure that the 1% has to be on a shot that procs Piercing Shots and that the crit that occurs because of the 1% gain happens outside the hidden trinket cooldown.
To be honest, I have yet to meet a Marksman hunter on my server who can out DPS me (not that that's a surprise, my server seems to have none of them) and I keep up with SV hunters just fine. Attack Power really shines as a buff to your pet, your serpent sting, and your Chimera shots. (Most notably SS which is a fire and forget mechanic that you buff and which will almost always proves to be invaluable due to the fact that even during heavy movement fights there it is, rolling along on the target). The 1% crit won't affect your pet's damage or your Serpent Sting. There's nothing that your AP won't honestly affect that should offset the lost of 1% crit in this situation. So I stand stubbornly against some internal calculation on the spreadsheet that gives more value to agility.
Yes, when I was referring to the armor and dodge benefits of the agility gems, I was only doing so out of completeness and some additional benefits. However, that additional dodge and armor from a few gems with rarely (if ever) make the difference between life and death.
As have been stated by others, crits are very important to MM hunters, so I will not repeat those arguements again.
The nice benefit of agility gems is that it helps to provide a balance between AP and crit. If one gets out of balance with the other, your DPS will suffer. If you have high AP but low crit, your non-crits will hit pretty hard, but since you are not critting as much, your average shot damage may be less. Likewise, if you have high crit and low AP, you may crit a lot but for lower damage, thus lowering your average shot damage. At some AP/crit ratio (I do not know exactly where it is), you maximize your DPS.
Like you, I have a hard time finding other MM hunters that can keep up with me, but a lot of that has to do with other factors than just gem selection.
In the end, the difference between agility and AP gems does not seem to be that much, so whichever works out best for you is fine. I just personally prefer the agility gems since I believe that they are a little better and help provide the balance between AP and crit.
Another added benefit for gemming agi is that it stacks with Kings in a raid situation as well as your true-shot aura and combat experience.
Also, as we reach higher and higher levels of AP with gear improvements, the AP loss (84 in the example above when we're running with 6500+ AP in raids and much higher with procs) becomes a smaller and smaller % of our total AP. But 1% crit will continue to be 1% crit and will be a bigger and bigger dps increase with better gear (higher dmg hits means bigger crits).
Ow and current glyphs are Serpent Sting, Chimera Shot, and Hawk.
Playing with the spreadsheet, Glyphs of Serpent/Steady/Kill Shot or Serpent/Hawk/Kill Shot is the highest dps, however Serpent/Steady/Hawk is only 3 dps behind.
Glyph of Serpent Sting seems to be the must have glyph, with Kill Shot / Steady Shot / Aspect of the Hawk are all relatively close, with Glyph of Kill Shot slightly higher depending on fight length.
Playing with the spreadsheet, Glyphs of Serpent/Steady/Kill Shot or Serpent/Hawk/Kill Shot is the highest dps, however Serpent/Steady/Hawk is only 3 dps behind.
Glyph of Serpent Sting seems to be the must have glyph, with Kill Shot / Steady Shot / Aspect of the Hawk are all relatively close, with Glyph of Kill Shot slightly higher depending on fight length.
The spreadsheet is fairly good on those glyphs you mentioned, because most of the them can have their math applied easily. Aimed and Chimera are difficult for the spreadsheet to handle, since they modify cooldowns by 2 and 1 second respectively. Kill Shot is fairly doable for the spreadsheet, since the modifier is -6 second and thus substantial enough to skip several GCD's. It's difficult then to measure the benefit of aimed and chimera, but I don't think you can entirely ignore them like you just did.
Personally I use talents in Barrage and Imp Barrage, despite people saying they are bad talents and all. Thus the aimed glyph has extra value for me, but with the numbers in Wow Web Stats I get the following.
In that log I have 24 aimed shots, with a 20% reduction in frequency that would amount to 19.2 shots so my glyph gives me 4.8 extra shots in that fight. Ignoring the fact that the rotation is not perfect and that cooldowns do overlap and so forth, I would be able to shoot 4.8 steady shots instead of those aimed. Those 4.8 steady shots translate into 15,196.56 damage. 4.8 aimed shots translate into 24,421.88 damage which means that with the aimed glyph I gain 9,225.32 damage. Which is a 0,74% dps increase. Not really much actually! This calculation is done under better than normal circumstances, so the number is a bit lower than that I guess. Also the sample size is quite small and is not really "proper" statistically speaking, but it does give you an idea of what the glyph does. However, I don't remember seeing math on other glyphs so it's a bit difficult to compare it.
I think Glyph of Chimera is the one to have (behind Serpent Sting of course). It bumped my overall Chimera Shot damage from about 7-8% to between 10-12% of my overall damage. Not bad for a single glyph
I think Glyph of Chimera is the one to have (behind Serpent Sting of course). It bumped my overall Chimera Shot damage from about 7-8% to between 10-12% of my overall damage. Not bad for a single glyph
I use Chimera glyph as well, and while it is true what you said, don't forget that you fire an extra chimera in place of a steady shot every rotation with the glyph. So even though your Chimera damage went up from 7-8 to 10-12 (same thing happened in my case), your Steady Shot damage goes down (i think in my case from 15 to 12% on average). Overall it is a damage increase ofcourse, but do not think it is a clear cut 3-5% damage increase using that glyph.
I realize the spreadsheet adds alot of value to Kill Shot glyph, but i find the concept of the glyph quite bizarre (Apart from YS phase 3 i only get to shoot one or two extra Kill Shots extra with this glyph). The reason i am still using Hawk glyph with 5/5 Imp Aspect of Hawk is because it is a self sustaining glyph/talent. The more haste you have, the more it procs, the more it refreshes itself, etc. Coupled with the fact about 25% of a MM hunter damage is white damage (plus an additional 5% through Wild Quiver) i will not give up this glyph. And well Serpent Sting glyph is a musthave for MM, think everyone agrees on that.
The Chimera Glyph is also useful because the shortened CD timer puts more of your damage in instant casts, great for fights on the run. I've been using Chimera, Serpent, Steady Glyphs but I was thinking of switching our Steady for something else (probably IAotH Glyph for the reasons you stated). But with steady still doing a decent amount of DMG I'm hoping I won't be giving up more than I'm gaining.
After my first week or so of Ulduar I decided against stacking haste (as I realized that it was a rather expensive stat overall, percent wise). I've been picking up random upgrades here and there and, though I have lost a little crit, I 've gained a LOT of AP, ArP, and HASTE (which is now at 13.05%). Is this to much haste, should I be frantically searching for ways to gear out of it an replace it with crit? I really was under the impression that anything over about 10 or 11% was overkill.
I am at 9.48% haste right now, and thanks to Hawk glyph and 5/5 Imp Hawk (standard MM spec ofcourse) i have a high uptime of Quick Shots anyway. I see haste as a lesser of two evils (the other being ArP), i wouldnt go out of my way to avoid haste, but im not gearing/gemming for it either. 100% of our damage range is affected by AP (and thus agility), a large part by crit, a lesser part by haste (directly through auto shot/wild quiver, indirectly through the few steady shots which in turn makes the Chimera Shot glyph worthwile), and an even lesser degree by ArP (doesn't affect pet, chimera shots, piercing shots, arcane shots, serpent sting). Edit; it actually does affect Piercing Shots, and to be clear, i am not saying ArP is a bad statistic per se, just saying i won't be stacking ArP or haste.
Anyway, in your case, 13% haste does sound like alot, although you arent using Hawk glyph at this time ofcourse. Again, i don't see haste as a stat i should avoid per se, and neither the opposite (i.e. full gearing for haste).
I know it's just anecdotal, but I really seem to have been noticing my pet focus starved recently. Pre-3.1 it would always be near full. I know that some streaks of non-crits can quickly focus starve it, but I've kept noticing my pet focus starved which makes me believe that there may be some changes in what procs GftT. Has anyone else noticed focus starvation or is it just me being "Onyxia deep breaths more" ?
And you're correct with your spec Aker, I do seem to get 110 DPS more out of a 15/51/5 spec going from a 7/57/7'ish spec
There was a change on focus regen, it dropped by 1 per second:
"
I just finished a test in game that shows untalented pet focus regen to be 5 focus per second, not the 6 used in the spreadsheet.
Crab on a target dummy, no points in Bestial Discipline or GftT. Started recording at 0 focus and only casting Claw. Casted Claw 36 times in 3 minutes = 5 times per second. 25 focus per cast / 5 seconds = 5 focus per second.
This should deal a small blow to BD and noticeably improve GftT in the spreadsheet.
Edit: Ran a second test to confirm numbers. Same parameters, 60 casts in 5 minutes for a rate of 5 focus/sec.
How significant this really is depends a lot on the pet itself and their abilities. If you have a wolf, 1 point into GfTt should be all you need. Any other pet would probably require 2.
I am at 9.48% haste right now, and thanks to Hawk glyph and 5/5 Imp Hawk (standard MM spec ofcourse) i have a high uptime of Quick Shots anyway. I see haste as a lesser of two evils (the other being ArP)
What? No. Armor pen is worth about twice that of haste at most gear levels and scales better.
What? No. Armor pen is worth about twice that of haste at most gear levels and scales better.
Perhaps, but my point was i am not going out of my way for either of these two stats; if it's there, it is there. I am more into the traditional stats; AP, agility, crit and hit, with some haste/ArP bonuses from gear since it is simply there. Basically for the same reason i wouldn't gem for ArP (or anyone else for that matter).
They really need to let Hunters on that "lol Hybrids get better haste" bandwagon. It's just wasted itemization right now because alternatives to the Chim glyph exist. Even if made 50% more effective, it still falls at the bottom of the DPS stat barrel.
I was really hyped about getting the Chim glyph, but after picking up some Ulduar gear and dropping some Haste, it's kind of stupid to use it. It actually hurts your mana efficiency and there are very few fights where it's actually helpful. Combine that with lag and you're pretty much just wasting itemization. Right now, it should be made a point to avoid haste in favor of anything non-haste.
You're going to pick up around 250 Haste grabbing upgrades, but Armor Pen seems to be Ulduars "new" stat of choice.
but Armor Pen seems to be Ulduars "new" stat of choice.
Really? Looking at the gear available to us from Ulduar, I'd say there's much more haste than there is ArP. Of the usable items there are 27 with ArP and 37 with haste. Looking at our tier set, which we will most likely have 4/5 of, there are 3 pieces with haste and 1 with ArP. I'd say it's the other way around, Ulduar's stat of choice seems to be haste which is unfortunate because the hunter stat of choice is ArP.
Says who? People don't give haste enough credit. Between Chimera Glyph and the buff to WQ, haste is still better than it was in Naxx. Though I don't suggest stacking to soft cap, having enough haste so you can get to soft cap from any haste effect (IAotH, etc.) is really beneficial to MM hunters. People don't seem to remember that Auto-Shot is still our NUMBER 1 DPS ability. Haste helps that (I know ArP does as well)
On the subject of Armor Pen, remember that half our damage still comes from abilities not affected by Armor Pen at all (not the least of which is Chimera Shot). Since our two main instants (Arcane and Chimera) are not affected by ArP both haste and ArP start to lose some of their value as movement increases.
Either way, they're both stats that really aren't that ideal for Hunters; and I think this is intentional. Hunters of almost all specs were very OP at the beginning of Wrath. Though we went through a series of Nerfs and Buffs, I think non-ideal itemization is a way of keeping our DPS in check.
I've actually been looking at a lot of gear with mostly the same stats, except dumping a lot of haste for Arp (a good trade off). With latency, getting 5 special shots in between every glyphed chimera just doesn't appear possible. It seems to me that dropping a steady (collision management will mean sometimes you do have to delay an arcane or aimed instead) between every chimera, while letting you use chimera every 9 seconds exactly, is the way to go. That is, you have a very brief wait time between the fourth GCD after chim before the next chim is available. Haste has no real effect on special shots in this scenario.
Edit:
Originally Posted by Anindor
People don't seem to remember that Auto-Shot is still our NUMBER 1 DPS ability. Haste helps
I will argue that this line of thinking is misleading. You cannot look at auto shot being our single highest damage ability as suggesting that we buff that shot. Here's a rough breakdown of my non-pet DPS on Hodir (not the best, but the only accurate parse I've got right now):
auto 26.5
chim 16.5
arc 12.8
chim serp 7.1
kill shot - 7.0
steady - 6.9
piercing - 6.8
serpent - 5.7
aimed - 4.5
wild quiver - 4.1
silencing shot - 1.3
White attacks (auto+wild quiver) did about 30.6% of my personal DPS. Special shots did about 68.6%. If you had to choose between buffing auto or buffing arcane, ok, I could easily see going with auto. But when you're talking about haste you're really talking about buffing that 30.6% while ignoring that 68.6%.
Similarly, if you look at any affliction lock parses shadow bolt is going to be their number one damage source, but DoTs as a whole far outweigh it, so buffing their sbolt damage would probably not be the best path to take. Hopefully there's some clarity in my ramblings here.
Just want to say something about Haste and Latency here because people are still using it as an argument against haste.
There is a severe misconception over the affect that latency has on gameplay. If you have 200ms latency, it takes that long for a packet to get to the server and back to you. People interpret this as "there is a .2 sec delay between every shot I cast because it adds it to my GCD". This is WRONG. As long as you have a consistent latency, you will be sending and recieving packets in a steady stream. The WoW client triggers your GCD, not the server. The server merely confirms that the shot was cast with the WoW client. The latency will only really affect the time it takes the server to process the damage done to the target. Though it can affect aspects of fights and the game, latency in small to moderate amounts should have little to no affect on your ability to be constantly casting.
If you want proof of this look at WWS reports from Patchwerk for top guilds around the world. If you live near a server in the US, your latency can often sit at around 70ms or less (this also depends on your hardware, ISP, router, etc.). In Australia, a player can't really get their latency under about 250ms at best (usually worse), certain areas in Europe have the same problem depending on what servers they play on. If latency really cut into your GCD wouldn't players in other countries would have a significant handicap and therefore do significantly less DPS. This just isn't the case.
There are many reasons that Haste is not an ideal stat, please stop saying latency is one of them.
I would like to know what MM shot rotations ppl are using. Ideally I am looking for a rotation that (i) maximises DPS and (ii) is simple enough to maintain in a raid environment.
Once Serpent Sting has been applied I use the following rotation: Chimera , Aimed, Arcane, SS, SS (repeat).
I have been using this because we have:
Chimera (10 sec CD + GCD);
Aimed (10 sec CD + GCD);
Arcane (6 sec CD + GCD); and
SS (no CD + GCD).
When the last SS GCD clears Chimera is off its CD and the rotation can be repeated.
Obviously where Kill Shot is available it takes priority.
One observation: The way you laid out the shots above implies that, for example, its actually 11.5 sec after the first click/press for CS before you can do that again. I have always thought the GCD is included in the CD - although I must admit to never having formally tested that - so it's truly 10 sec between clicks/presses. The same is true for Aimed & Arcane. That might seem like semantics but it matters as you start weaving shots with varying CDs together.
One observation: The way you laid out the shots above implies that, for example, its actually 11.5 sec after the first click/press for CS before you can do that again. I have always thought the GCD is included in the CD - although I must admit to never having formally tested that - so it's truly 10 sec between clicks/presses. The same is true for Aimed & Arcane. That might seem like semantics but it matters as you start weaving shots with varying CDs together.
The GCD is indeed included in the CD. In the case of Chimera, you can fire your next Chimera exactly 10 seconds after you fired the first one, and here in lies the issue: one can choose to fire 5 shots between two chimera shots but you would 'waste' time, or fire 6 shots between two chimera shots but delay your chimera shots each rotation (it does mean you will have 100% cast time, but overall you fire one or two less Chimera Shots during the fight).
An alternative would be the Chimera Shot glyph ofcourse, making the time between two Chimera Shots 9 seconds (which is 6 x GCD, one of which is the initial Chimera Shot). With hasted steady shots you would fit in five shots (instants/steady shots) between the two Chimera's perfectly, but thats the issue, can you guarantee hasted steady shots without soft haste cap? No. Not even with 5/5 Imp. Hawk and Hawk glyph, but it does come close, and that's why i have chosen for the Chimera glyph.
The GCD is indeed included in the CD. In the case of Chimera, you can fire your next Chimera exactly 10 seconds after you fired the first one, and here in lies the issue: one can choose to fire 5 shots between two chimera shots but you would 'waste' time, or fire 6 shots between two chimera shots but delay your chimera shots each rotation (it does mean you will have 100% cast time, but overall you fire one or two less Chimera Shots during the fight).
An alternative would be the Chimera Shot glyph ofcourse, making the time between two Chimera Shots 9 seconds (which is 6 x GCD, one of which is the initial Chimera Shot). With hasted steady shots you would fit in five shots (instants/steady shots) between the two Chimera's perfectly, but thats the issue, can you guarantee hasted steady shots without soft haste cap? No. Not even with 5/5 Imp. Hawk and Hawk glyph, but it does come close, and that's why i have chosen for the Chimera glyph.
Thank you both for your helpful responses and for clarifying the GCD point. I will do some testing and revert.
One interesting consideration is the increased chance to proc Imp Steady Shot in the 6 shot cycle (assuming one has picked up 2/3 or 3/3 of this talent).
What do people think the the most useful spec is right now for MM? I am currently using the 7/57/7 spec and i am pondering using 11/53/7. Could someone please post the best setup for 11/53/7 thoughts and comments on this spec are appreciated.Thanks
Im sure this question has been asked but I was not able to find it well reading thru the numerous pages.
An alternative would be the Chimera Shot glyph ofcourse, making the time between two Chimera Shots 9 seconds (which is 6 x GCD, one of which is the initial Chimera Shot). With hasted steady shots you would fit in five shots (instants/steady shots) between the two Chimera's perfectly, but thats the issue, can you guarantee hasted steady shots without soft haste cap? No. Not even with 5/5 Imp. Hawk and Hawk glyph, but it does come close, and that's why i have chosen for the Chimera glyph.
The argument for/against the CS glyph shouldn't be that it allows some number of other shots between CDs in a perfect world, i.e. haste capped & 0 latency, it should be that having CS land every 9 sec, even with dead space is more DPS than having CS land every 10 sec with whatever dead space that CD results in. This is complicated by the fact that the AiS CD is also 10 secs so unglyphed the rotation is a little crisper - CD, AiS, some number of SrS, AS & SS until CS is of CD and repeat.
This post in the survival forum laid out how to calculate how long you could delay casting ES due to an SS before it became a net DPS loss. Presumably similar math exists for CS, I just haven't bothered to figure it out.