1. FFXIV Reset Timers
    Last daily reset was 4 hours, 24 minutes ago / Next daily reset is in 19 hours, 35 minutes
    Last weekly reset was 2 days, 21 hours, 24 minutes ago / Next weekly reset is in 3 days, 12 hours, 35 minutes
+ Reply to Thread
Results 1 to 13 of 13
  1. #1
    Banned.

    Join Date
    Jul 2007
    Posts
    10,392
    BG Level
    9

    Critical Hit Rate | 3.0 | Bardshrekt

    If you didn't see Spell Speed / Skill Speed infos, you can find it here and discussion here.

    tl:dr, increased level, increased stats, decreased potency per point. This somewhat curbs the power creep and due to the nature of some of the skills that were added, that classes have/use, keeps certain important stats (crit rate, spell speed/GCD reduction) in line with 2.0 launch.

    Important to note! The first few values are not in line with the rest.
    Short formula for SSPD: 2.5 - (0.01 x (YourSSPD - 354) / 26.5) )

    ---

    Critical Hit Rate!

    Now between the SSPD information and people telling me that Critical Hit no longer affects Critical Hit Rate (lol) because muhpatchnotes, I just had to go gather data to see what kind of nerf we were looking at.

    I used Deadgye's data from 2.0, found here, to compare against Lv60 data.

    Since the only important information is what the critical hit stat does to critical hit rate and critical hit damage, the other stats (should be) non-relevant provided sufficient sample sizes within a specific individuals parse data/encounter.

    The information I have so far can be found here

    I have some more data at different Critical Hit rates, but the tests weren't long enough (Warrior auto attack is not super fast) so I've revised the spreadsheet to include only the larger samples.

    Without a ton of accuracy, you can say that roughly 44 Critical Hit will raise your Critical Hit Rate by 1%. I am currently collecting data for 715 Critical Hit, and will collect 754 Critical Hit after that (and the subsequent Critical Hit Damage information).

    I wanted to throw what I had up early in case someone else had some data collected and for discussion.

    P.S.
    For all of the Bards out there wondering why you're not seeing a rough 100% increase in DPS when playing identically to your 2.5 rotation, your main stat being so heavily nerfed is a good start as to why. I'm sure as we get more gear with better itemization this will start to shift in your favor. I won't pretend to speak for WM or why it's good or bad.

  2. #2
    Strider/Doom/Cyclops
    Join Date
    Jun 2007
    Posts
    5,647
    BG Level
    8
    FFXIV Character
    Viper Beam
    FFXIV Server
    Balmung
    FFXI Server
    Quetzalcoatl

    So is it also true that crits are no longer a straight x1.5 modifier, and are scaled based on +CRT?

  3. #3
    Banned.

    Join Date
    Jul 2007
    Posts
    10,392
    BG Level
    9

    Edit: Updated with what information I have. Would need a lot higher Critical Hit stat to see anything different.

  4. #4
    Banned.

    Join Date
    Jul 2007
    Posts
    10,392
    BG Level
    9

    Did baseline (354 crit) crit rate and damage, updated in the googledoc. Currently doing 757 crit.

    Crit Damage looks interesting so far and really opens this game back up to some interesting possibilities once we get some better itemization.

  5. #5
    a p. sweet dude
    Pens win! Pens Win!!! PENS WIN!!!!!

    Join Date
    Jul 2010
    Posts
    17,155
    BG Level
    9

    Nothing to add, but wanted to voice appreciation of the effort.

  6. #6

    Your crit% is sent to the client whenever you apply a DoT. You do not need to test thousands of hits to find an approximate crit% anymore. ACT can tell you your exact crit chance rounded to the nearest tenth.

    Started up a doc, would love people to help test to find the exact crit formula.

    https://docs.google.com/spreadsheets...t#gid=94249285

  7. #7
    Smells like Onions
    Join Date
    Jul 2015
    Posts
    3
    BG Level
    0

    (CRT*0.0002330) - 0.0329793 looks like the formula.

    Alternatively, ((CRT-354)*0.0002330) + 0.0495080 gives you a value closer to your "Theoretical" Crit Data.

    It's up in the air as to exactly which one it is, as ACT rounds it. The more that Doc is updated, the more accurate my CRT solver will become.

  8. #8
    Banned.

    Join Date
    Jul 2007
    Posts
    10,392
    BG Level
    9

    It would make sense for the value to be in the 0.023xx range as well as the bonus to be 4.95xx.

    I only base that on the fact that it seems like critical hit damage is 0.0023xx; 10^-1 as potent seems like a logical step and easily future proofed.

  9. #9
    Smells like Onions
    Join Date
    Jul 2015
    Posts
    3
    BG Level
    0

    If you have any Critical Damage Min/Max Values, could you try this formula and see how well it checks in with your data? BaseDMG* ((CRT-354)*0.0002717)+1.4370748)

    If there's any Inconsistencies, tell me or just send me all your Data Points and I'll add a few modifications here and there.

  10. #10
    Banned.

    Join Date
    Jul 2007
    Posts
    10,392
    BG Level
    9

    What I did for Crit damage is all in the googledoc in the OP

    Code:
    Critical hit	Crit High/Low	Normal High/Low	# of Crits	Multiplier
    @757 Crit 	511 / 462	331 / 300	967	M - 1.5419968304278922345483359746434
    @754 Crit	517 / 468	336 / 304	509	M - 1.5390625
    @738 Crit	510 / 461	331 / 300	1190	M - 1.5388272583201267828843106180666
    @715 Crit	505 / 456	329 / 298	841	M - 1.5326953748006379585326953748006
    @649 Crit	617 / 558	406 / 367	3179	M - 1.5200517464424320827943078913325
    @540 Crit 	497 / 437	340 / 293	1369	M - 1.4755134281200631911532385466035
    @491 Crit	612 / 551	429 / 372	488	M - 1.4519350811485642946317103620474
    @354 Crit	538 / 487	371 / 336	732	M - 1.44978783592644
    Using this: ((CRT-354)*0.0002717)+1.4370748) to compare:

    757 - 1.54657
    754 - 1.54575
    738 - 1.54141
    715 - 1.53516
    649 - 1.51723
    540 - 1.48761
    491 - 1.4743
    354 - 1.4370748

    ---

    There are some variances that don't necessarily make sense in my own data; this could be due to an insufficient number of critical hits to gather all damage points, it could even be due to some other stats forcing something to be higher or lower than it should be. (eg, rounding)

    The first thing that immediately jumped at me when I did this a week ago was how closely related the potency of each point of Critical Hit seemed to be to Critical Hit Rate and Critical Hit Damage.

    What I failed to do was resolve the difference in apparent potency at lower critical hit values, though I primarily did that so I could have a number to work with at more realistic (future) values.

  11. #11
    Smells like Onions
    Join Date
    Jul 2015
    Posts
    3
    BG Level
    0

    You never achieved a complete Min/Max on 540/491.

    612 / 1.451935081 = 421.5064488, 8 damage below your non-crit Min/Max.

    The other data-sets you've sent could possibly work. I'll add them to my solver and see what I get.

    EDIT:

    Try this ((CRT-354)*0.000233495)+1.4482336) I've added the data points which I've checked to most probably be absolute min/max values. This puts my scaler VERY close to the Critical Hit Chance scaler. Not bad at all.

  12. #12
    Banned.

    Join Date
    Jul 2007
    Posts
    10,392
    BG Level
    9

    Yeah that's a lot closer.

  13. #13
    Smells like Onions
    Join Date
    Jul 2015
    Posts
    1
    BG Level
    0

    Quote Originally Posted by BrokenWings View Post
    Your crit% is sent to the client whenever you apply a DoT. You do not need to test thousands of hits to find an approximate crit% anymore. ACT can tell you your exact crit chance rounded to the nearest tenth.

    Started up a doc, would love people to help test to find the exact crit formula.
    Nice spreadsheet. I tried seeing how well the data fit with a JP-made formula from 2.x. It turns out they were probably dead right, so I added a sheet.

    The trouble is that you'd need to believe that there's a hidden base value for crit past level 50, but since it fits the data pretty much perfectly, I'm willing to accept that.

Quick Reply Quick Reply

  • Decrease Size
    Increase Size
  • Remove Text Formatting
  • Insert Link Insert Image Insert Video
  • Wrap [QUOTE] tags around selected text
  • Insert NSFW Tag
  • Insert Spoiler Tag