How important is being at the 1017 crit (crit cap raid buffed) in 1.9.. vs AP stackin

[gnomeater]

So, I have heard and read alot of different theories on the status of crit cap and AP. However, I have found nothing definitive.

Many people have stated that while AP's scaling has been improved, it is still imparitive that you stay at crit cap while raiding....

On the other hand, others say that you should stack crit and maintain selfbuffed levels of 900-950...

So, now i have two different loadequips one at 1498ap/1016 crit and the other at 1573ap/915 crit.

What do you guys think is the best way to go?

[Sir Thrace]

Yeah there seems to be no definitive answer. I think the idea is it all depends on how much AP you have to sacrifice to stay crit capped. Go with the general guideline of 1 AP = 2 Crit and see how much AP it costs you when you are at 1017 crit. If it costs more that 1:2 AP than it probably isn't worth it.

I guess the only other way to get a good answer is to do Maklamos in both your load equips a few different times on each and average your DPS (assuming you don't die) and see which is higher.

[Primalthirst]

Focusing on the crit cap is bad and people need to stop doing it.

Pre 1.8 it was important because AP was terrible.
Post 1.8 crit is unimportant because AP is fantastic.

You should NOT aim to reach the crit cap. You should be aiming to reach optimal stat distribution, which is along the lines of 1 AP = 0.45 Crit rating. The soft cap only comes into play if somehow you still reach it when stacking AP, at that point crit becomes only 1/5th of its pre-cap value and thus is almost useless.

Yes crit is still useful, and yes its value is dynamic based upon AP levels.
But AP is more useful and its value is dynamic based on crit levels.
Both improve as you gain more of the other.

[Souldancer]

Thanks for posting the helpful information. Is there a place where I can find what we *should* be focusing on - the soft caps and hard caps for our dps stats is what I mean.

[Sir Thrace]

Just check out Rain's BiS list in the Warrior Guides section. Start with getting your sigil BiS then pick up the proper gear as it drops.

[Souldancer]

Right, I'm looking at his list which is really, really great, but what I was hoping for was the hard numbers on what is the crit soft cap, what is the AP soft cap, etc. Can I find that information in the sheet? I'm looking more to gauge where I am at currently with what I have geared, so that I can adjust runes and etc.

Does that make sense?

[Sharogy]

"repost"

The correct Weights for AP vs Crit is about 1:1.7 - 1:1.8 in ID gear unless u run some 44/22, which means most available item from the same Ilvl will benefit u greater if they are leather.

this can be easily verified using ferboten's DPS calculator, simply increase Ap by 25, see raid dps increase, vs crit by 50, see raid dps increase, and u see crit win always under softcap.


For more numerical example:

On a dummy, with no consumables just selfbuff, im Criting PS for 3135, BF for 6410, (no GI buff, after stacking to 5 stack of GI, my ps Crits for 3236, bf for 6793), atm i have 36.25% crit.
Backward calculation for avg non crit dmg is the following:

NON GI:
PS hits for 3135/1.5 = 2090
BF hits for 6410/1.5 = 4273
WITH GI:
ps hits for 3236/1.5 = 2157
BF hits for 6793/1.5 = 4529

Note: 150 ap is in itemisation value equal to 300 crit, that means if i switch out 150 ap for 300 crit instead i can obtain a 11.4% extra crit (assume all under softcap values).

Note: (all numbers i typed in this post are rounded up for easy typing, real calculation is done in excel with exact values).
In a core champion rotation of PS PS BF, without GI, expected avg dmg is the following:
(2090*0.6375+3135*0.3625) = 2469 on PS
(4273*0.6375+6410*0.3625) = 5048 on BF
this means the avg dmg of this simple mini rotation would yield: 9985 dmg.

Now Stacking up to 5 stacks of GI buff, expected avg dmg is the following:
(2157*0.6375+3236*0.3625) = 2548 on PS
(4529*0.6375+6793*0.3625) = 5349 on BF
this means the avg dmg of this mini rotation would yield: 10446 dmg.

Now assume this theoretical Gear swap is possible for extra 300 crit instead of getting this 150 AP:
my new crit would be 47.65%, using non GI stack numbers the expected avg dmg is the following:
(2090*0.5235+3135*0.4765) = 2588 on PS
(4273*0.5235+6410*0.4765) = 5291 on BF
this means the avg dmg of this mini rotation would yield: 10466 dmg.

Conclusion: in this given situation, 1 ap for finishers clearly scales better than 2 crit for finishers, but the opposite is true for all builder attacks, and combined in this mini rotation which is fairly representative for the champion rotation in a whole, Crit still scales better in the 1 ap vs 2 crit Ratio.

What does this mean?
it means it is better to stack crit up to Softcap over AP given the same itemisation values. aka, get them leathers.

---------------------------------------------------------------------------------------------------------------------
EXTRA:
To further test the scalling, does this ratio change in any way when i have more buffs? in order to test this i simply apply all the consumables i can afford, and do the same test, if the result shows the gap becomes "smaller" between AP and crit, then it is possible that AP will outscale crit in a raid buffed environment, however if the opposite is true then there is no reason for me to test this in raid buffs as it will likely to only put crit further ahead.

Consumable: Oilstone+blue flask+f00d+tablet
new crit %: 37.75

Current crit dmg values: 3359 on ps and 6769 on bf (no GI stack)
Expected avg dmg:
3359/1.5*0.6225+3359*0.3775 = 2662 on PS
6769/1.5*0.6225+6769*0.3775 = 5364 on BF
Rotation dmg: 10688

Now stack up to 5 stack of GI: 3461 on ps and 7152 on bf
Expected avg dmg:
3461/1.5*0.6225+3461*0.3775 = 2743 on PS
7152/1.5*0.6225+7152*0.3775 = 5668 on BF
Rotation dmg: 11152

with Imaginary crit gear instead of the 150 AP:
Expected avg dmg:
3359/1.5*(0.6225-0.114)+3359*(0.3775+0.114) = 2789 on PS
6769/1.5*(0.6225-0.114)+6769*(0.3775+0.114) = 5621 on BF
rotation dmg: 11199.9

--------------------------------------------

Conclusion: There is no indication that gap would get closer when having more "buffs" on. hence the earlier drawn conclusion is likely to be correct when extrapolated to raid buff environment.

that 2 crit > 1 AP for ID gear lvls all across the board except for 44/22.

--------------------------------------------
Highly recommanded: Ferboten's dps calculator
it maps out all ability across rotations in all different raid buff settings, the conclusion is the same, 2 crit > 1 ap for everything except 44/22 in ID gear.

[PFalcon]

I have not seen a str/AP item in HK or ID, after reitemization and 1.9, that is not worth the tradeoff of crit yet. To be honest the decision has been taken away from us by the devs.

They are never gonna give you a STR/AP item that gets beaten by a dex/crit, STR/crit, or dex/AP item.

About your only choices these days are sane tier sidegrades. Then just apply the 2:1 rule...


...or don't and just max AP. You will do fine.

[Souldancer]

I have not seen a str/AP item in HK or ID, after reitemization and 1.9, that is not worth the tradeoff of crit yet. To be honest the decision has been taken away from us by the devs.

They are never gonna give you a STR/AP item that gets beaten by a dex/crit, STR/crit, or dex/AP item.

About your only choices these days are sane tier sidegrades. Then just apply the 2:1 rule...


...or don't and just max AP. You will do fine.

So it sounds like you are in the camp of Crit is no longer *king*, but the guy before you posted a lot of math (that I could swear I saw somewhere before a couple months back) is of the mind that crit (and therefore DEX) is still superior.

There's too much conflicting information, and I still can't find the answer to what *are* the soft and hard caps of stats. Really, really confusing guys.

[Primalthirst]

Crit softcap is 1187 rating.
Hardcap is over 3000 rating.
Raid buffs and consumables contribute 140 crit.
AP has no caps.

Just realised my original reply had the AP : crit ratio backward, woops.
But trust Sharogy over me, he's done more work on it.

[Galibier]

So it sounds like you are in the camp of Crit is no longer *king*, but the guy before you posted a lot of math (that I could swear I saw somewhere before a couple months back) is of the mind that crit (and therefore DEX) is still superior.

There's too much conflicting information, and I still can't find the answer to what *are* the soft and hard caps of stats. Really, really confusing guys.

The soft cap is 45% or 1187 crit. So 1187-52-52-40=1043. (Bard and archon AP buffs + pot) So once here calculate the benefit of resonance.. 1043-(dex*.05)= equals the crit you should have pre-raid buffs for soft cap.

The hard cap is 60%. Problem is diminishing returns make it so it's over 130 points of crit to gain 1% after you hit 45%. So while it is 1187 crit rating to = 45% you need 3163 crit rating to reach 60%.

Now keep in mind that soul abilities and buffs that say "adds 5% to crit chance" do not apply to your crit rating but the actual crit percentage. So if based on pure crit rating you have 45% crit raid buffed, a % bonus like mark of extermination gives you 50% crit chance, not an 1187+59 (59 .35 being 5% of 1187.)

[Havors]

"repost"

The correct Weights for AP vs Crit is about 1:1.7 - 1:1.8 in ID gear unless u run some 44/22, which means most available item from the same Ilvl will benefit u greater if they are leather.

this can be easily verified using ferboten's DPS calculator, simply increase Ap by 25, see raid dps increase, vs crit by 50, see raid dps increase, and u see crit win always under softcap.


For more numerical example:

On a dummy, with no consumables just selfbuff, im Criting PS for 3135, BF for 6410, (no GI buff, after stacking to 5 stack of GI, my ps Crits for 3236, bf for 6793), atm i have 36.25% crit.
Backward calculation for avg non crit dmg is the following:

NON GI:
PS hits for 3135/1.5 = 2090
BF hits for 6410/1.5 = 4273
WITH GI:
ps hits for 3236/1.5 = 2157
BF hits for 6793/1.5 = 4529

Note: 150 ap is in itemisation value equal to 300 crit, that means if i switch out 150 ap for 300 crit instead i can obtain a 11.4% extra crit (assume all under softcap values).

Note: (all numbers i typed in this post are rounded up for easy typing, real calculation is done in excel with exact values).
In a core champion rotation of PS PS BF, without GI, expected avg dmg is the following:
(2090*0.6375+3135*0.3625) = 2469 on PS
(4273*0.6375+6410*0.3625) = 5048 on BF
this means the avg dmg of this simple mini rotation would yield: 9985 dmg.

Now Stacking up to 5 stacks of GI buff, expected avg dmg is the following:
(2157*0.6375+3236*0.3625) = 2548 on PS
(4529*0.6375+6793*0.3625) = 5349 on BF
this means the avg dmg of this mini rotation would yield: 10446 dmg.

Now assume this theoretical Gear swap is possible for extra 300 crit instead of getting this 150 AP:
my new crit would be 47.65%, using non GI stack numbers the expected avg dmg is the following:
(2090*0.5235+3135*0.4765) = 2588 on PS
(4273*0.5235+6410*0.4765) = 5291 on BF
this means the avg dmg of this mini rotation would yield: 10466 dmg.

Conclusion: in this given situation, 1 ap for finishers clearly scales better than 2 crit for finishers, but the opposite is true for all builder attacks, and combined in this mini rotation which is fairly representative for the champion rotation in a whole, Crit still scales better in the 1 ap vs 2 crit Ratio.

What does this mean?
it means it is better to stack crit up to Softcap over AP given the same itemisation values. aka, get them leathers.

---------------------------------------------------------------------------------------------------------------------
EXTRA:
To further test the scalling, does this ratio change in any way when i have more buffs? in order to test this i simply apply all the consumables i can afford, and do the same test, if the result shows the gap becomes "smaller" between AP and crit, then it is possible that AP will outscale crit in a raid buffed environment, however if the opposite is true then there is no reason for me to test this in raid buffs as it will likely to only put crit further ahead.

Consumable: Oilstone+blue flask+f00d+tablet
new crit %: 37.75

Current crit dmg values: 3359 on ps and 6769 on bf (no GI stack)
Expected avg dmg:
3359/1.5*0.6225+3359*0.3775 = 2662 on PS
6769/1.5*0.6225+6769*0.3775 = 5364 on BF
Rotation dmg: 10688

Now stack up to 5 stack of GI: 3461 on ps and 7152 on bf
Expected avg dmg:
3461/1.5*0.6225+3461*0.3775 = 2743 on PS
7152/1.5*0.6225+7152*0.3775 = 5668 on BF
Rotation dmg: 11152

with Imaginary crit gear instead of the 150 AP:
Expected avg dmg:
3359/1.5*(0.6225-0.114)+3359*(0.3775+0.114) = 2789 on PS
6769/1.5*(0.6225-0.114)+6769*(0.3775+0.114) = 5621 on BF
rotation dmg: 11199.9

--------------------------------------------

Conclusion: There is no indication that gap would get closer when having more "buffs" on. hence the earlier drawn conclusion is likely to be correct when extrapolated to raid buff environment.

that 2 crit > 1 AP for ID gear lvls all across the board except for 44/22.

--------------------------------------------
Highly recommanded: Ferboten's dps calculator
it maps out all ability across rotations in all different raid buff settings, the conclusion is the same, 2 crit > 1 ap for everything except 44/22 in ID gear.

so 2 crit is better than 1 ap? until you hit the soft cap (1043 unbuffed) i presume?

Im debating how best to get 1043 without sacrificing too much AP if that is the case... would making your sigil carry a few crit essences instead of AP help? seeing as most of the ID is all AP heavy?

With full warrior ID gear, AP sigil and Scythe, if i remember correctly, unbuffed crit sits around 900 when i was on PTS...

[Havors]

I would like to see some parses instead of just theory.

A raid parse stacking AP and leaving yourself crit around the 900 mark... and another raid parse getting crit to 1043 pre raid.

[Sharogy]

parses are prone to RNG, calculations are not.

[Havors]

parses are prone to RNG, calculations are not.

well if parses are prone to RNG then calculations are almost pointless as RNG will get its meddlesome stick out and poke them

a good aggregate of parses is the only way to go imho