Thread: 4set on SimC?

Page 1 of 2
1
2
LastLast
  1. #1

    4set on SimC?

    Dunno if there's another thread, but I went down slightly on Crit and abit of Strength for 4pc. Now SimC is saying I only went up aprox 900dps. For a fight that lasts 4 - 4.5min I cannot imagine the DPS gain from gaining a 4set is only 900DPS seeing if you didn't have that set bonus you would only be able to use it either at the start or at the end of the fight.

    So, is SimC 4set not calculating correctly?

  2. #2
    Stood in the Fire
    Join Date
    Jan 2011
    Location
    Finland
    Posts
    486
    Well, the overall problem on WoW's "combat table" has always been the crit %. It doesn't matter how often you can get close to 100% crit with recklessness, if you still have most of the uptime time without the buff. It means it will average your crits to reflect the stable buffed crit % of your char (plus that extra -3% or something from the boss level).

    So if you've just critted with BT/MS many times because of either using cooldowns or luck, it will still try to average all those skills back to that crit % by not critting that much on the next few minutes. So unless the parse is very short, it means it will even out without stable crit % modifiers. Meaning you will trade most of the damage from somewhere to that where you demand it. Most classes that don't rely on crit obviously don't have that "problem". That's why crit proc trinkets have always "sucked" outside the buffs range, because on low uptimes it will always average itself out not giving any "extra at all" like strength procs do, unless if you can align to do most damage inside them.

    It will give you more demand burst at times, but if you lose like 2,5-4% crit because you need to choose badly itemized gear for the tier, it's going to hurt so much that despite the extra burst. So it just evens it out as overall damage output most of the times only giving "slight increase".

  3. #3
    both set bonuses give about 2k on a 14N BiS level. If you need to trade in str and crit for this, 900 dps doesn't sound too wrong.
    Last edited by Berthold; 2012-12-17 at 10:35 AM.

  4. #4
    Herald of the Titans Darkfriend's Avatar
    Join Date
    Mar 2011
    Location
    Wisconsin
    Posts
    2,831
    Quote Originally Posted by Berthold View Post
    both set bonuses give about 2k. If you need to trade in str and crit for this, 900 dps doesn't sound too wrong.
    For the gear I had when I got two set, it was between a 1.7% and 4% DPS increase depending on fight, and it should scale the exact same with gear due to being a % increase to MS. (This was as arms).

  5. #5
    Quote Originally Posted by Kankipappa View Post
    Well, the overall problem on WoW's "combat table" has always been the crit %. It doesn't matter how often you can get close to 100% crit with recklessness, if you still have most of the uptime time without the buff. It means it will average your crits to reflect the stable buffed crit % of your char (plus that extra -3% or something from the boss level).

    So if you've just critted with BT/MS many times because of either using cooldowns or luck, it will still try to average all those skills back to that crit % by not critting that much on the next few minutes. So unless the parse is very short, it means it will even out without stable crit % modifiers. Meaning you will trade most of the damage from somewhere to that where you demand it. Most classes that don't rely on crit obviously don't have that "problem". That's why crit proc trinkets have always "sucked" outside the buffs range, because on low uptimes it will always average itself out not giving any "extra at all" like strength procs do, unless if you can align to do most damage inside them.

    It will give you more demand burst at times, but if you lose like 2,5-4% crit because you need to choose badly itemized gear for the tier, it's going to hurt so much that despite the extra burst. So it just evens it out as overall damage output most of the times only giving "slight increase".
    Everything you just said is completely wrong and I have no idea where you gotsuch ridiculous ideas from. Each and every moves crit chance is unaffected by the result of previous moves. If you have 10% crit chance and crit 10 times in a row, your next hit still has 10% cit chance.

  6. #6
    Lost some crit from shoulders seeing how tier warrior shoulders is by far the worst piece, but loss was not higher than 1% so it's not a huge loss for 4pc. I recall having somewhere 22.90ish crit now I have 22.04% crit.

  7. #7
    Quote Originally Posted by Kalmah View Post
    Everything you just said is completely wrong and I have no idea where you gotsuch ridiculous ideas from. Each and every moves crit chance is unaffected by the result of previous moves. If you have 10% crit chance and crit 10 times in a row, your next hit still has 10% cit chance.
    Well simcraft has an option for deterministic rng but thats just now how it works in wow

  8. #8
    Stood in the Fire
    Join Date
    Jan 2011
    Location
    Finland
    Posts
    486
    Quote Originally Posted by Kalmah View Post
    Everything you just said is completely wrong and I have no idea where you gotsuch ridiculous ideas from. Each and every moves crit chance is unaffected by the result of previous moves. If you have 10% crit chance and crit 10 times in a row, your next hit still has 10% cit chance.
    Actually, I've throughly tested this on Cataclysm many times. 2x crit (proc & use) trinkets and recklessness, pooling them as long as possible to gain long uptime with buffed crit, enough so my overpower could sustain around 95-100% crit chance for nearly a minute(15s+20s+12s). My next 5 overpowers (with 85% crit) plus using MS on cooldown, none of the ablities never really critted after that. I could recreate this issue nearly every time I tried. Haven't got double and/or good use crit trinkets yet on MoP so can't test it if it's any better in here, but yeah there has always been some chance on crit "diminishing" with temporary buffs like it or not.

    Using recklessness many times on the fight should put you clearly over the average crit % on all ablities, but it never happens for real. Or maybe I just seem to have the worst luck ever every single time I play on my warrior? =P
    Last edited by Kankipappa; 2012-12-17 at 03:18 PM.

  9. #9
    Quote Originally Posted by Kankipappa View Post
    Actually, I've throughly tested this on Cataclysm many times. 2x crit (proc & use) trinkets and recklessness, pooling them as long as possible to gain long uptime with buffed crit, enough so my overpower could sustain around 95-100% crit chance for nearly a minute(15s+20s+12s). My next 5 overpowers (with 85% crit) plus using MS on cooldown, none of the ablities never really critted after that. I could recreate this issue nearly every time I tried. Haven't got double and/or good use crit trinkets yet on MoP so can't test it if it's any better in here, but yeah there has always been some chance on crit "diminishing" with temporary buffs like it or not.

    Using recklessness many times on the fight should put you clearly over the average crit % on all ablities, but it never happens for real. Or maybe I just seem to have the worst luck ever every single time I play on my warrior? =P
    Your problem is a combination of small sample sizes (short fights where you only popped CDs 2-3 times) and biased sampling (mainly remembering the times things didn't happen). Popping reck does increase your crit% average above the tooltip overall. It's easy to calculate how much based on the uptime, and overall it does average out to increasing it by that much. WoW does not compensate negatively so that your crits will average out to the tooltip after using a crit bonus, or nobody would ever use crit proc trinkets and CDs.

  10. #10
    Stood in the Fire
    Join Date
    Jan 2011
    Location
    Finland
    Posts
    486
    I guess it comes down to bad luck and too short parses if it really isn't supposed to affect it at all.

    However nobody in reality ever uses (or have used) two 15% crit trinkets, because there has always been so few good proc/use combinations. It's just so weird that if you twingle with those things too much, even overpower seems to hit more than it's supposed to in a row, so that the crit % won't even remain over 90% for the next minute, what was supposed to be the "unbuffed value" on my tier 13 on bis gear. So on my personal experience it does some skewing around to get closer to the tooltip value in a short duration when the buffs run out and that gets averaged out better when the sample gets longer and longer.

    It made even ultraxion pain miserable at times to get into the "dry season" after good deadly calm + trinket + recklessness combo afterwards because nothing would crit for the next 30 seconds. Although I Didn't know that checking values on skada is being biased! But at least the effect with only 1 crit proc trinket feels way smaller to even make visible on most cases.


    But thinking more about recklessness and checking wol parses would let me to believe, that the recklessness crit buff uptime alone is so low, that on overall it can't surpass or even get close to the crit reduction what the boss level alone gives (3%?). It would mean that when averaged without any math, recklessness surely gives only max of +1% average crit for the fight, because even on the best crit-nrg parses my overall crit is always ~2% less than the tooltip value, when counting all the abilities with no crit modifiers... Meaning only abilities that I would also use inside recklessness, aka counting out dragon roar and overpower.
    Last edited by Kankipappa; 2012-12-18 at 12:07 AM.

  11. #11
    The Patient
    Join Date
    Dec 2012
    Location
    Sweden
    Posts
    280
    Quote Originally Posted by Kankipappa View Post
    .
    You can probably explain why this guy had 70%+ crits with his bl then, obvious bug in this system that is suppose to track all your earlier abilitys outcome to then "balance it out". http://worldoflogs.com/reports/rt-g5...?s=6598&e=6997

    Your dmgmeter doesnt even do that type of tracking and you still need an addon to accually track your dmg, but blizzard has some sort of hidden dmgmeter which then performs some sort of job after those numbers? Nah...

    This is one of the most idiotic things iv ever heard of. The answer for the answer above is simply plain luck and not that he has 35% crit, 38, counting the boss is 3 lvls above.

  12. #12
    Stood in the Fire
    Join Date
    Jan 2011
    Location
    Finland
    Posts
    486
    Quote Originally Posted by Kadde111 View Post
    You can probably explain why this guy had 70%+ crits with his bl then, obvious bug in this system that is suppose to track all your earlier abilitys outcome to then "balance it out".
    No. I'm sorry, but I don't understand what you mean by a single ability. Sure you can always burn recklessness on execute phase and that should skew the execute values up the same, no? That's not what I was after, I was talking about average and how it's skewed. What happens again when you count all hits/parrys/blocks/misses into one and crits separated, then add them up, and divide the crit count from all abilities? How much does the average crit precent look now? Suddently close or under the tooltip value like usual?

    But of course, it's always tricky trying to count the average from abilities with fixed increased crit chance by default, like dragon roar, bloodthirst or overpower. You would have to know how to count the additional increased average of crit chance from tooltip when you're using those abilities on cooldown. So it's better to count all others separate first and look how the average looks with and without those fixed abilities and it might give some insight already.

    Well, I checked that log. So, by looking Kramix's armory it reveals unbuffed crit 23.67, raidbuffed +5% means 28.67% and maybe add extra 0.03 from 5% allstats buff or something rounding it to ~28.70%.

    Now, counting the reported wol numbers with misses/hits/blocks/misses/glancing/deflect (if any) etc. combined and crits as seperate.
    Ability name - hits / crits - crit %
    Melee: 250 / 99 = 28.36%
    Execute: 17 / 9 = 34.61%
    Raging Blow: 36 / 21 = 36.84%
    Bloodthirst: 24 / 51 = 68.00%
    Bloodbath: 100 / - = 0.00%
    Raging Blow Off-Hand: 41 / 14 = 25.45%
    Heroic Strike: 32 / 14 = 30.43%
    Wild Strike: 28 / 15 = 34,88%
    Deep Wounds: 97 / 37 = 27.61%
    Dragon Roar: 6 / 6 = 50.00% (half missed?)
    Lightning Prison: 16 / - = 0.00%
    Colossus Smash: 14 / 4 = 22.22%
    Stormlash: 25 / 3 = 10,71%
    Heroic Throw: 3 / 2 = 40.00%

    All combined: 689 / 238 = 25.67% - close to 28.70% when you count in the -3% from boss level.
    without 100% fixed hit/crit: 583 / 269 = 31.57% (no bloodpath hit ticks or dragon roar crits)
    without fixed and BT: 559 / 218 = 28.05% (no bloodpath, dragon roar or bloodthirst)

    Suddently BT's 68% crit chance wasn't that groundbreaking by overall, or was it? Sadly I don't know how to add or count 3x recklessness (12.7% uptime) there so I won't even try to count or check on it. Or even think how to adjust the BT's double crit chance in the calculations for trying to figure out actual average crit rate.

    So let's assume wow's combat table will never check the averages on crit, or even on anything... Then I'm just amazed how accurately it works to stay within or under the crit's tooltip value no matter the buffs.
    Last edited by Kankipappa; 2012-12-19 at 10:09 PM.

  13. #13
    The knockback from dragon roar is what generated the immune messages.

    You're including 16 ticks of lightning prison without even understanding what it is. You're including stormlash without considering it using spell crit rate rather than melee crit rate.

    Excluding those, 518 hits, 215 crits = 29.3% crit chance on abilities that follow melee crit chance. If crit buffs made no difference overall like you claimed, he would have had around 25.7%, so your theory's looking pretty flimsy.
    Last edited by Rhia; 2012-12-20 at 01:18 AM.

  14. #14
    The Patient
    Join Date
    Dec 2012
    Location
    Sweden
    Posts
    280
    As an addition to Rhia's answer, reck only increases crit from abilities, not melee swings.

    About the BT, given that he has 28,67% crit his BT will have 57,54% crit as you drop reck his crit for abilities will rise to 58,67% and his BT will be 117,54% <---- This is what i think you should avoid by having reckglyph and instead getting an increased duration with guaranteed BT crits and enrage time.

    (If there was some mechinic fixing his crit he wouldnt get close to 70% crits on BT that fight but 58%~)

    My point with pointing that guys log out was that its RNG, not some fixed mechanic that makes you not crit. The reason you see so even numbers is that the more counts of abilities/longer fight the more precise the number will be to the very crit that is shown in your stats due to probability.

    Thats how logics works, if you flip a coin 10 times you might get heads 7 times but if you flip it 1000 times there is a big chance that you'll end up with about 500 heads and tails. This doesnt mean "God" has put a system into life which would autofix your cointossing to end up 50/50 "just becuase".

    Do you see my point? - This is just silly.
    Last edited by Kadde111; 2012-12-20 at 09:49 AM.

  15. #15
    Quote Originally Posted by Kankipappa View Post
    No. I'm sorry, but I don't understand what you mean by a single ability. Sure you can always burn recklessness on execute phase and that should skew the execute values up the same, no? That's not what I was after, I was talking about average and how it's skewed. What happens again when you count all hits/parrys/blocks/misses into one and crits separated, then add them up, and divide the crit count from all abilities? How much does the average crit precent look now? Suddently close or under the tooltip value like usual?

    But of course, it's always tricky trying to count the average from abilities with fixed increased crit chance by default, like dragon roar, bloodthirst or overpower. You would have to know how to count the additional increased average of crit chance from tooltip when you're using those abilities on cooldown. So it's better to count all others separate first and look how the average looks with and without those fixed abilities and it might give some insight already.

    Well, I checked that log. So, by looking Kramix's armory it reveals unbuffed crit 23.67, raidbuffed +5% means 28.67% and maybe add extra 0.03 from 5% allstats buff or something rounding it to ~28.70%.

    Now, counting the reported wol numbers with misses/hits/blocks/misses/glancing/deflect (if any) etc. combined and crits as seperate.
    Ability name - hits / crits - crit %
    Melee: 250 / 99 = 28.36%
    Execute: 17 / 9 = 34.61%
    Raging Blow: 36 / 21 = 36.84%
    Bloodthirst: 24 / 51 = 68.00%
    Bloodbath: 100 / - = 0.00%
    Raging Blow Off-Hand: 41 / 14 = 25.45%
    Heroic Strike: 32 / 14 = 30.43%
    Wild Strike: 28 / 15 = 34,88%
    Deep Wounds: 97 / 37 = 27.61%
    Dragon Roar: 6 / 6 = 50.00% (half missed?)
    Lightning Prison: 16 / - = 0.00%
    Colossus Smash: 14 / 4 = 22.22%
    Stormlash: 25 / 3 = 10,71%
    Heroic Throw: 3 / 2 = 40.00%

    All combined: 689 / 238 = 25.67% - close to 28.70% when you count in the -3% from boss level.
    without 100% fixed hit/crit: 583 / 269 = 31.57% (no bloodpath hit ticks or dragon roar crits)
    without fixed and BT: 559 / 218 = 28.05% (no bloodpath, dragon roar or bloodthirst)

    Suddently BT's 68% crit chance wasn't that groundbreaking by overall, or was it? Sadly I don't know how to add or count 3x recklessness (12.7% uptime) there so I won't even try to count or check on it. Or even think how to adjust the BT's double crit chance in the calculations for trying to figure out actual average crit rate.

    So let's assume wow's combat table will never check the averages on crit, or even on anything... Then I'm just amazed how accurately it works to stay within or under the crit's tooltip value no matter the buffs.
    That would be probability, my friend.

  16. #16
    Stood in the Fire
    Join Date
    Jan 2011
    Location
    Finland
    Posts
    486
    True that, I overlooked stormlash's spell crit chance. For warrior that only means -5% less though, so it easily goes inside the error margin, so you shouldn't just completely count it out. For fury that would still mean crit buffed around at least 23% in this case. That would still make it overall 28.75% if you count it in.

    Since recklessness does affect only "your special attacks" I'm sure it doesn't have affect stormlash either though. Also forgot to list out lightning prison from there from the list, good catch - isn't it that the boss ability anyway? I guess staying up late makes up for that, heh.

    However since BT is just normal special attack as others, the double crit chance does have some part in it. However since I don't know how to exactly count it, so just counting it out does turn the numbers off a little. How much? Honestly I don't know.

    On my personal logs Arms logs counting overpower (+60% crit) increased the actual crit average rate to around 35% (and without around 28%), where as my buffed crit was at around 29.5% for Arms, at least before the current wrists I got. In current values for Fury when at ~30% crit, BT seems to give around extra 3% overall - by just looking the numbers that is, no idea on actual values.

    So when you look it around the numbers and my first post, it still looks close to being true, unlike Kalmah said. Also keep in mind that players do (and should heh) use Recklessness on sub 20% on boss, so it also gets the value up on the end as the boss dies soon after. It would have to be used on the start of the match and only there so the combat table would have time to average it out "mostly". But for me it still looks it tries to average it, because honestly on "pure random" 33% crit chance you could already have some funky values on some abilities. Like on starcraft I got zerg in a row for 20 times when playing as random.
    Last edited by Kankipappa; 2012-12-20 at 10:50 AM.

  17. #17
    It's like you're picking and choosing when to remember crit suppression. 23.7% melee crit means about 18.7% spell crit chance, +5% from crit buff, -3% crit suppression is 20.7% crit chance on stormlash.

    29.5% buffed crit is 26.5% crit chance vs boss level, etc. If you keep forgetting crit suppression of course it will seem you're falling short of your projected crit rate.

    Looking at a log from last week, only melee specials, excluding bloodthirst and execute (as recklessness uptime for execute is disproportionately high), almost entirely against boss level mobs, with 24.8% melee crit before buffs, 26.8% vs boss level mobs with 5% crit buff:

    Ability: non-crit / crit

    Raging Blow: 350 / 143
    Raging Blow OH: 349 / 142
    Wild Strike: 319 / 196
    Heroic Strike: 255 /90
    Colossus Smash: 132 / 55
    Heroic Leap: 62 / 32
    WW: 115 / 49
    WW OH: 120 / 44
    Heroic Throw: 17 / 4

    Total: 1719 / 755

    755 / (1719 + 755) = 30.5% crit rate compared to the 26.8% your hypothesis predicts

    Edit: forgot to include parries etc, adding those in now, which will lower the crit rate a bit but nowhere near enough to bring it down by 4%

    Edit: added in parries now.

    At most 29% of the damage in that log was vs non-boss level mobs.
    Last edited by Rhia; 2012-12-20 at 03:35 PM.

  18. #18
    Stood in the Fire
    Join Date
    Jan 2011
    Location
    Finland
    Posts
    486
    So you dismissed melee attacks all the way on the list, what can also miss and get parried, blocked etc. and do the most damage for as an ability for fury on overall? The numbers already get skewed when picking out abilities that you do, so to get "more correct" list you would have to dismiss doing that ability at all on the combat. I think it would be more accurate to count all those you do in the fight and the subtracts those abilities what have increased/decreased crit % and see how it affects the total value on anyway. There are also tank ablities on the list but since there is only one or two of them I don't think it matters "too much".

    If just counting all the melees on from that list i got melees value: 2358 / 638 = 27.05%. Probably would be best to check that if I counted the numbers right and add it in too.

    And on the log you pasted, you just included the whole log? So, a bold move would be to somehow also count that 29% (if it's accurate) "uptime" on +3% crit (or dunno what level the creatures are, 90? 91? 92?). With dumb math that would give extra 0,87% with they all were just lvl 90. I can't remember the mob levels tbh because I never check them really.

    Anyway, even counting logs into one where combat log stops and starts many times and fights are short probably won't be best accurate way to calculate overall stats. I'm pretty sure that it won't "remember" different logs and try to average them. It just checks the current combat and tries to keep itself on your current crit % by all abilities you do (plus the extra modifiers like BT ofc). I'd prefer counting the boss fights separate, because on some bosses like the spirit binder what's in the list you also get crit buffs if you visit the spirit zone and get healed - what would be ideal to check alone if it gives any increase of crit on overall at all, right?

    Still, my original opinion stands on the idea how it works. I've never said that it's 100% accurate on everything and that's the reason I used quotation marks when I wrote it, because sure - it might give some benefit, but in my opinion it's still greatly diminished how the system works vs. a system where it would be purely pure crit % extra. For example 15% crit trinket could mean on 33% uptime (what's sounds kinda realistic?) should still give extra 5% crit on overall on all abilities, but still seems to greatly fall under that, because it has only 33% uptime. For the 66% of the time then it tries to get inline with the current crit value and starts to DR your next crits for the next minute to get there.
    Last edited by Kankipappa; 2012-12-21 at 03:26 PM.

  19. #19
    What? I thought your whole theory was that abilities that raise your crit chance get negated out over time so that your average crit rate matches your character sheet crit rate -3% boss level suppression. As recklessness doesn't affect melee attacks, melee swings should match character sheet rate anyway, making them pointless to include. Though you can take the fact that the crit rate of melee attacks is pretty much always lower than the crit rate of specials as an indication that the impact of recklessness is not negated by the game like you're trying to claim.
    Last edited by Rhia; 2012-12-21 at 07:15 PM.

  20. #20
    Stood in the Fire
    Join Date
    Jan 2011
    Location
    Finland
    Posts
    486
    But you don't seem to understand the fact that melee swings are in the same combat table too, so that makes it tricky! If you filter them out what are like the biggest portion of your damage, it will make it look either favourable or unfavourable depending how the crit table has spreaded out. That's the fun thing, because it spreads the crits randomly on ALL the abilities you do, while trying to keep it close to your crit %.

    On the 3rd post I did say that for overall, recklessness might bump up the overall crit for about 1% and I was never talking it being completely negated, but trying to explain why it only shows 900 dps increase for Kirinya as simulated because it's value diminishes by how the stuff works. Reck is probably giving max about ~5% damage benefit on fight anyways when you can align all the best stuff in there. The examples I wrote here was for the dudes especially who thought that the crit is purely random and not limited, which imho it is not, considering how the stuff works.
    Last edited by Kankipappa; 2012-12-21 at 08:27 PM.

Posting Permissions

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