Page 3 of 3 FirstFirst
1
2
3
  1. #41

    Re: why are people gemming agil still?

    Quote Originally Posted by Arel
    Which is why hard-capping ArP is the best possible situation, granted you can get there.
    GG, I always seem to have a wall of text in the end, but I'd rather be thorough. :P

    It seems that the general mentality since Mjolnir's Runestone came into existence and through ToC/ToGC was that you either go Arpen softcap + Arpen proc trinket or hardcap Arpen, anything in between is sub-optimal DPS. Just from a theoretical standpoint, that is ridiculous. I do believe this mentality developed because it was impossible for druids to reach arpen hardcap or get much beyond softcap when the MR came out, and there wasn't anything much better than a MR. Perhaps the individual math is a bit harder to comprehend, and if I get a little free time I'll attempt to crunch the numbers more exactly. However, I can at least provide a simplistic argument for the time being.

    First, I'll supply a personal example since it will provide the basis for the argument. Since kitty dps is one of my offspecs, I've been maintaining one throughout the content. Since I don't always get the best and greatest gear for not being a kitty mainspec, my gear tends to lag based on the content. That being said, going into ICC, I had constructed my gear around Mjolnir Runestone, ~64% crit static, ~12.5k AP static, hit/exp capped, arpen soft-capped. Throughout progression, I noticed that even though I was getting several gear upgrades and gemming agi since "going over arpen softcap is bad!" mentality was around, my DPS was not increasing and severely lagging behind the rest of the raid. Being fed up with sub-par DPS, I decided to overhaul my gear to attempt arpen hardcap. This entailed actually downgrading several pieces... in the end, I achievement 1401 arpen at the cost of approximately 400ap, 10% crit, 1% haste, and 1.5% below expertise softcap. After several parses on fights such as Saurfang, the average DPS increase from my soft cap to hardcap gear was 2.5k DPS. For the general idea of what my gear is/was, my DPS gear falls into the 5.5k-5.6k GS range for kitty.

    Alright, so what's the point? Even at the expense of a lot of other DPS stats, dropping Mjolnir (which provided limited arpen hardcap moments) and going for full-time arpen yielded a ridiculous gain in DPS, 2.5k DPS in my case. I just can't swallow that there isn't a point where you can stack arpen, be below the arpen hardcap, and not see a DPS increase over sticking with softcap + trinket. Now, I propose that there is an effective arpen (eArpen) provided by a Mjonlir/NES. In layman's terms, one should ideally be able to unequip their MR/NeS and add a static amount of arpen to their gear and see no change in their DPS output.

    The tricky part is that the value of arpen and the DPS increase it provides gets better the more you have (just because of how it scales), so we can assume that the eArpen is not an empirical value but dependent on the current state of your gear, aka your current passive arpen. Actually, most people already know this already, because the proponents of softcap + trinket clearly state that "more arpen beyond the softcap devalues the arpen proc of the trinket," meaning the DPS gain of the trinket proc goes down with more arpen. Ignoring the crit portion on MS/NeS contributing to DPS (crit trinkets are a dime a dozen anyways), the eArpen of the trinket is a function your current passive arpen, the amount of arpen the trinket provides, and the uptime of said trinket. Now, I know there are going to be some people screaming "I line up my cooldowns to blow when the trinket procs," and that's all fine and dandy if you could actually control when the trinket procs (especially NES, I've heard the horror stories). What you can do is throw a fudge factor to sooth your spirits, when I did napkin math I tacked on an extra 10% to the estimated eArpen and it worked out fine.

    I did napkin math and actually tested this on my gear since I was going to overhaul it anyways:
    With MS, I saw approximately a proc per minute. 665 arpen for 10 seconds, 50 seconds w/o it => ~111 arpen over 60 seconds => ~111 arpen static. Throw in a fudge factor of 10% and you get 122 static arpen. Using my current arpen as starting point (basically, I was too lazy to calculate it around anything else), so using an eArpen = 122 arpen, I replaced my MS with a Banner of Victory (84 arpen) and replaced two 20agi gems with two 20 arpen gems (I was going to replace them anyways). Granted it isn't a perfect scenario, since I dropped 102 crit rating, gained ~1k ap proc with Banner of Victory, and lost 40 agi, but this is napkin math and I'm not made of gold! Well, long story short, sitting on a target dummy yielded a slight DPS increase (we're talking averaged out to be a 25 DPS increase). However, the result hinted that there is an eArpen value where you can just drop your arpen proc trinket, replace it with static arpen equivalent to its eArpen, and DPS will not be adversely affected.

    If someone can beat me to the empirical math accounting for the trinket's proc rate, the arpen proc amount, and the current stat of your passive arpen to yield an effective arpen value, please do so! Until then, I'll work on the formulas and post later.
    “Society is endangered not by the great profligacy of a few, but by the laxity of morals amongst all.”
    “It's not an endlessly expanding list of rights — the 'right' to education, the 'right' to health care, the 'right' to food and housing. That's not freedom, that's dependency. Those aren't rights, those are the rations of slavery — hay and a barn for human cattle.”
    ― Alexis de Tocqueville

  2. #42
    Herald of the Titans arel00's Avatar
    15+ Year Old Account
    Join Date
    May 2008
    Location
    Italy
    Posts
    2,852

    Re: why are people gemming agil still?

    It's true, not everything in between is a DPS loss.

    However, it's merely a talk about optimization. Generally speaking, by the time you're grabbing a good chunk of ICC gear, you'll be sitting at around 800-900 passive ArP, without counting any gem.

    That is a very delicate threshold. If you get to 1k passive, then gemming full ArP is a HUGE dps gain. Let's say, without math but with a bit of good sense, than if you can shot to 1.2k ArP, gemming included, you can safely drop your trinket.

    If you're juggling around 900ArP, like me for example, then enhancing critical chance gives slightly more DPS. Keyword: slightly. I tested that thing about going deep into Haste because it struck me that with the current stat weights, it may have been worth it, and it looked like fun, chain-gunning white attacks and proccing loads of OoC.
    It does seem good, after some testing, but the general situation remains the same: when you're caught in between the soft and the hard caps, every little shift in your gear can lead to a completely different way. It's hard to decide.
    Quote Originally Posted by Qieth
    I don't do math, blind assumptions work so much better for me.

  3. #43

    Re: why are people gemming agil still?

    lol agi = dodge for pres.
    Quote Originally Posted by Easyplay
    I have an idea for a new ability that Warlock could get in Cataclysm.

    Upgrade/change Banish
    As you cast Banish on your enemy, the enemy will be bansihed to Twisting Nether where he/she would burn for eternity (just for few seconds).

  4. #44

    Re: why are people gemming agil still?

    I did napkin math and actually tested this on my gear since I was going to overhaul it anyways:
    With MR, I saw approximately a proc per minute. 665 arpen for 10 seconds, 50 seconds w/o it => ~111 arpen over 60 seconds => ~111 arpen static. Throw in a fudge factor of 10% and you get 122 static arpen. Using my current arpen as starting point (basically, I was too lazy to calculate it around anything else), so using an eArpen = 122 arpen, I replaced my MR with a Banner of Victory (84 arpen) and replaced two 20agi gems with two 20 arpen gems (I was going to replace them anyways). Granted it isn't a perfect scenario, since I dropped 102 crit rating, gained ~1k ap proc with Banner of Victory, and lost 40 agi, but this is napkin math and I'm not made of gold! Well, long story short, sitting on a target dummy yielded a slight DPS increase (we're talking averaged out to be a 25 DPS increase).
    First, I am making the educated guess that your MS was meant to be MR for the runestone largely based on the 665 number you've specifically mentioned. I've edited this in the quote above to reflect MS = MR.

    Second, I don't think you'd have a typical down time of ~50 seconds on MR (10secs of up and 50 of down). It has an Internal Cooldown of 45seconds that starts the moment it procs (That means by the time the effect ends, you've only got 35secs before the icd is done with) and it typically procs within one to two seconds of that 45sec mark as feral. Putting up time to down time ratio at 10secs to 36/37secs. Which, over a long fight, would average to about 26 to 28% up time. Shorter fights can be subject to some RNG and the possibility of ending, say for example, 4secs before the proc would have happened again (which can throw a wrench into the idea of hitting ~27% up time). I think safe to say 20 to 27% up time for most any given fight all things considered. Whether that up time is spent on the boss or not is perhaps another story, though the way the proc works, it tends to put higher up time on target in scenarios that have you on/off targets consistently throughout the encounter despite what I think most people think, where most emphasize the cases where it procs and immediately thereafter you're off target and it's wasted.

    Next, you can't do this.

    Back when FBN was still being updated this kind of thought process came up about trying to place a DPS value on Grim Toll and Mjolnir Runestone. People wanted to average the proc out and just use that arpen number for the whole proc's effec; people felt they had to have a DPS value to assign to the trinket, but the best you could really do is simulate an exact gear set to get an idea where it stood in DPS value for that gear set. The problem with averaging the 665 10sec up time with a 0 arpen effect on the ~35sec down time is that Arpen in higher numbers is more effective because of the way arpen works. That means if you were to try to take some kind of average and use that flat number, the proc would actually be better than that because it comes in a large chunk all at once which makes it worth more while it is up than what an average would be.

    735 arpen + 10secs of an additional 665 (1400) with a ~28% up time is better than 900 (735+165) arpen all the time. And this is me giving free points to your argument's concept as 665 Arpen for 10secs out of 47... 665proc x 10secs = 6,650 / 47secs cycle time = 141 arpen average.

    Not only this, but a training dummy is not an ideal test. For one, that Attack Power proc on the trinket you traded out is going to heavily be favored on a training dummy test because bleeds will make up more of your damage percent compared to if you had raid buffs...

    Solo training dummy tests will almost always favor AGI and large AP trinket procs over arpen or arpen procs. You have no sunder, you have no windfury, etc. A larger percent of your damage on a Dummy is Rip/Rake relative to what it would be in raid scenario. In raid scenario, White damage and Shred damage will move up relative to Rip/Rake on a dummy because of heavy influences of stuff like sunder/windfury.

    When people say:
    1) Gem all agility as new feral
    2) Gem arpen only to soft cap if you have arpen trinket, and rest gem agility
    3) Gem to hard cap if you can hit 1k+ arpen numbers (kind of a loose rule)

    They say all this under the idea that you're not working on a training dummy and instead you're in a raid with raid buffs and raid debuffs all in effect.

  5. #45
    High Overlord Mortimer's Avatar
    10+ Year Old Account
    Join Date
    Jan 2010
    Location
    Thunder Bluff
    Posts
    128

    Re: why are people gemming agil still?

    Quote Originally Posted by Arel
    A small note now, based on that infamous research. If you have your ArP trinket, you have around 850-900 passive ArP from gear alone, and your raidbuffed crit is over 65% at least, you can try a very funny thing.
    Fill your yellow sockets with Haste, and your red sockets with Agi+Haste gems.

    It nets a VERY high DPS on a theoretical side, close to hard-capping ArP.
    I actually did try that thing, and gained an average of 500-600dps. I don't have logs at the present time 'cause I have little time to raid, but I'll try to record and post some when possible.
    It takes at least 850 passive ArP, gems not included, and you must push Haste over 500, this time including gems, while you're keeping crit over 65%. It's hard to juggle, it's been a bitch for me to set up the gear and try. But it's good. Lots of OoC procs, so an extremely fast rotation and a good deal of Bites.

    It's a bit shitty on balancing the stats because the more you get Haste, the more you Bite, the more you need ArP to strengthen it. Iirc, the best results are at roughly 1k ArP and over 600 Haste. Good luck getting there tho :P
    I'd been hoping you would post about your little experiment soon. Seems like a very interesting thing to try, white damage must sky rocket. :O Any chance you could post the exact gear you used (chardev/wowhead profile)? :D

    Also, from what gearing point did you gain "an average of 500-600dps"?

  6. #46
    Herald of the Titans arel00's Avatar
    15+ Year Old Account
    Join Date
    May 2008
    Location
    Italy
    Posts
    2,852

    Re: why are people gemming agil still?

    Quote Originally Posted by Mortimer
    I'd been hoping you would post about your little experiment soon. Seems like a very interesting thing to try, white damage must sky rocket. :O Any chance you could post the exact gear you used (chardev/wowhead profile)?

    Also, from what gearing point did you gain "an average of 500-600dps"?
    I'll try to up a profile on Chardev to give you an idea.

    As for the second question: by re-gemming mostly, I didn't have any huge gear jumps like getting 4xT10 suddenly or something like that :P I dropped some ArP and some agility and went into Haste. I'll try to post 2 profiles.

    Disclaimer: remember please that I'm mostly a tank. It may very well be that using all that Haste simplified my rotation so I just "played it better". However, it is interesting enough and quite funny when you start a "chain-Clarity proc streak" :P
    Quote Originally Posted by Qieth
    I don't do math, blind assumptions work so much better for me.

  7. #47

    Re: why are people gemming agil still?

    some good posts in this thread ;D


    Quote Originally Posted by Ekna
    The problem these days about gemming agi is that at a certain gear level you get over the crit cap also. with the crit cap being 72% i find myself in certain circumstances over that cap (i've seen 78% )
    In case you look over this thread again, the crit cap is 77.8% on your character sheet assuming all buffs (ret pally/ToW). You're exactly right about the gear level and crit cap however, i've been over the crit cap ever since i got a few pieces of ICC gear, and especially when you start the ArP hardcap grind (and if you're like me you get to next to no hit)

  8. #48

    Re: why are people gemming agil still?

    Thanks for all the replies. Far better posts than you see on the DK forum thank god..
    Yeh my AP is pretty low, this is my 1st of 5 alts so still gearing. Below is my armory link.
    http://www.wowarmory.com/character-s...l&cn=Wraithaur

    So basically what i am thinking is that now that i am at the arp soft cap with NES to start gemming agil, until i have the opportunity to his the arp hard cap.

    One question i would like answered is the distribution of dmg for people
    For me since i am stacking arp its Shred>melee>rip
    Is Rip>melee>shred a byproduct of having so much crit IE 70%??

    PS how crap is it that every class bar Feral Druids can easily obtain a weapon. Our easiest upgrade to get is the staff off 25 Fester, which is hardly easy to get if your on an alt and have to roll against a million hunters.

  9. #49
    High Overlord Mortimer's Avatar
    10+ Year Old Account
    Join Date
    Jan 2010
    Location
    Thunder Bluff
    Posts
    128

    Re: why are people gemming agil still?

    Quote Originally Posted by Wraith
    One question i would like answered is the distribution of dmg for people
    For me since i am stacking arp its Shred>melee>rip
    Is Rip>melee>shred a byproduct of having so much crit IE 70%??

    PS how crap is it that every class bar Feral Druids can easily obtain a weapon. Our easiest upgrade to get is the staff off 25 Fester, which is hardly easy to get if your on an alt and have to roll against a million hunters.
    Not sure if this is an accurate representation of my normal distribution, but looking at the two latest fights still on my Recount it's Melee 33%, Shred 29% and Rip 24% on Koralon 25 (PUG) and Melee 32%, Shred 31%, Rip 20% on Toravon (same PUG). I had a mangle bot, and didn't DPS adds on Toravon (I did however out-DPS a full 251/264 geared cat with a DBW -_-).

    I've always been of the opinion that Ferals should be given priority over Hunters on staves/polearms. To them it's just a stat stick, while to us, it is the biggest deciding factor of our DPS/threat. If Hunters give you shit for thinking you deserve it more than them, ask if they mind Rogues/DPS warriors rolling on their guns on equal basis with them. Same thing, but most hunters throw a shit fit whenever I suggest it.. =P

    Edit:
    Quote Originally Posted by Arel
    I'll try to up a profile on Chardev to give you an idea.

    As for the second question: by re-gemming mostly, I didn't have any huge gear jumps like getting 4xT10 suddenly or something like that :P I dropped some ArP and some agility and went into Haste. I'll try to post 2 profiles.

    Disclaimer: remember please that I'm mostly a tank. It may very well be that using all that Haste simplified my rotation so I just "played it better". However, it is interesting enough and quite funny when you start a "chain-Clarity proc streak" :P
    Cool. Mostly t9 233/245 and offspec roll won gear then? What's the swing speed? Must be in the sub-.5's.

  10. #50

    Re: why are people gemming agil still?

    Quote Originally Posted by Wraith
    PS how crap is it that every class bar Feral Druids can easily obtain a weapon. Our easiest upgrade to get is the staff off 25 Fester, which is hardly easy to get if your on an alt and have to roll against a million hunters.
    if you're on an alt the easiest upgrade to get would be Hersir's Greatspear from 10 Princes, I doubt most pugs can kill festergut 25 (at least on my server) yet, but princes and 10man in general are pretty easy

    some people like BQL 25 caster staff more then the other weapons, so technically we don't have it the hardest but it could be argued

  11. #51
    Herald of the Titans arel00's Avatar
    15+ Year Old Account
    Join Date
    May 2008
    Location
    Italy
    Posts
    2,852

    Re: why are people gemming agil still?

    Quote Originally Posted by Mortimer
    Edit:Cool. Mostly t9 233/245 and offspec roll won gear then? What's the swing speed? Must be in the sub-.5's.
    Err, no.
    It's actually 4xT10 and various other stuff (yeah, I have 8 T10 total, VoA knows something about it :P).

    And THIS is the character profile.

    It doesn't show on Chardev since I cba setting all buffs there, but the swing speed is somewhere around 0.61

    Going full agility on gemming nets a very close result. I think full agility still is a little higher, but the different is like ~60dps on a math model, so quite negligible in reality.
    Quote Originally Posted by Qieth
    I don't do math, blind assumptions work so much better for me.

  12. #52
    High Overlord Mortimer's Avatar
    10+ Year Old Account
    Join Date
    Jan 2010
    Location
    Thunder Bluff
    Posts
    128

    Re: why are people gemming agil still?

    Quote Originally Posted by Arel
    Err, no.
    It's actually 4xT10 and various other stuff (yeah, I have 8 T10 total, VoA knows something about it :P).

    And THIS is the character profile.

    It doesn't show on Chardev since I cba setting all buffs there, but the swing speed is somewhere around 0.61

    Going full agility on gemming nets a very close result. I think full agility still is a little higher, but the different is like ~60dps on a math model, so quite negligible in reality.
    Oh, right. That explains it. You wrote something like "I didn't get 4t10 suddenly", but I thought it was strange that you would able to do this in t9...

    I notice that you're using agility Dragon Eyes. Why not 34 haste ones and full agil/haste in reds? It's only 78 agility traded for 72 haste (0.2sec faster swing), so it might not be a major factor. As you say, the difference between agility and haste is barely noticeable.

    I might try this for the lulz when I get a few upgrades (namely 4t10. Toravon is not my friend...). I only raid 10 man, and not very often, so hard capping ArP is out of the question for me.

    Also, do you have a DBW for your fucking offspec? D:

    Edit: I just realised how likely it is that we're going to be stacking haste in Cataclysm if haste boosts energy generation. Fun times ahead.

  13. #53
    Herald of the Titans arel00's Avatar
    15+ Year Old Account
    Join Date
    May 2008
    Location
    Italy
    Posts
    2,852

    Re: why are people gemming agil still?

    Quote Originally Posted by Mortimer
    Oh, right. That explains it. You wrote something like "I didn't get 4t10 suddenly", but I thought it was strange that you would able to do this in t9...
    Yeah, meaning I had the 4 parts bonus both before and after the change in gems. At T9 level you simply lack the passive ArP to do it, nothing really more to say.

    I notice that you're using agility Dragon Eyes. Why not 34 haste ones and full agil/haste in reds? It's only 78 agility traded for 72 haste (0.2sec faster swing), so it might not be a major factor. As you say, the difference between agility and haste is barely noticeable.
    I tried not to push it too far for now. I probably will change the Dragon's Eyes in the near future if I keep getting decent results.

    Also, do you have a DBW for your fucking offspec? D:
    It's called "groundbreaking luck of doom +5 vorpal" :P
    I went away for 5 days and missed the guild main raid, so I joined an alt run to collect Frosties and random stuff. Got the DBW in that run. We have a fair number of tanks and sometimes I'm going pewpew even with guild raids, so they were merciful enough to let me grab it. I was using Banner and one Fractured Dragon's Eye before I got it.

    Edit: I just realised how likely it is that we're going to be stacking haste in Cataclysm if haste boosts energy generation. Fun times ahead.
    Indeed. I'm extremely curious to see what will happen. They also said all the current ArP will most likely be turned into Crit and Haste, so expect funny numbers when 4.0 hits 80-85% crit and 750-800 Haste could be a realistic situation. Man, I wanna see that.
    Quote Originally Posted by Qieth
    I don't do math, blind assumptions work so much better for me.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •