Item Search
     
BG-Wiki Search
+ Reply to Thread
Page 1 of 2 1 2 LastLast
Results 1 to 20 of 29

Thread: Fast Cast Testing     submit to reddit submit to twitter

  1. #1
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    Fast Cast Testing

    90 RDM for 30% Fast Cast + 30% from Revelations/MM/Ambition + 22% from Warlock's Chapeau +1/Duelist's Tabard/Loq Earring = -82% Fast Cast, but only -40% Recast.

    Weakened Reraise was 72 seconds (-40%) instead of 70 (-41%).

    Without Loquacious (80% Fast Cast), Weakened Reraise was still 72 seconds (-40%).

    Looks like there's either an 80% Fast Cast cap, or a 40% "Fast Cast's Contribution to Recast" cap.

  2. #2
    Groinlonger
    Join Date
    Oct 2006
    Posts
    2,964
    BG Level
    7
    FFXI Server
    Fenrir

    I'd throw on some Nebula gear to perhaps further confirm any sort of Fast Cast cap given the mostly ambiguous values posted for Fast Cast gear/traits.

  3. #3
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    Urk D: Don't really want to buy Nebula Cuffs +1 (1% Fast Cast) again to sort through odd/even values. Dedication to exact values . . . waning . . . . _(x . x)_

    I confirmed that RDM trait alone was -15% recast - 51 sec - (+Loq = -16% Recast - 50 sec), and that Trait + Hat/body was -25% Recast - 45 sec. Above, on WAR/WHM, I confirmed that the Atma together were -15% recast. If anyone wants to push it further, everything is there for them.

  4. #4
    Groinlonger
    Join Date
    Oct 2006
    Posts
    2,964
    BG Level
    7
    FFXI Server
    Fenrir

    Well it's certainly more complicated than perhaps odd/even values, at least it seemed that way. Kirschy and I found values of like.... 9/1024 and 11/1024 for different pieces and the like. fdasdf rly

  5. #5
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    I'm not sure. I think we may have overcomplicated it in the past. Do you have a link to some of that testing? I'd like to see if it can be explained by a simple odd/even rule.

    Some things in the game that are allegedly percentages are are definitely handled in /256 or /1024. Haste and -%DT for sure. Other things that are allegedly percentages are definitely only applied as whole percentages. Killer effect PDT/+Damage from Ferine Gausape +2 (980 for 1000 needles), Critical Hit Rate, Hit Rate (both proven in our DEX vs. crit rate thread), and possibly Fast Cast's effect on recast.

    It's also entirely possible that they changed how Fast Cast works in the last patch when they changed the cap

  6. #6
    Groinlonger
    Join Date
    Oct 2006
    Posts
    2,964
    BG Level
    7
    FFXI Server
    Fenrir

    There is quite a bit of information on the first page of this thread that I posted. Feel free to look over it. There is certainly some kind of truncation going on, whatever the case actually as, as some of my results would simply be impossible otherwise.

  7. #7
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    Actually, it all works out with the theory I proposed:


    Column F is:
    =FLOOR(120*(100-FLOOR( (B3~17*B$1+C3~17*C$1+D3~17*D$1+E3~17*E$1)/2, 1))/100, 1)

    I made it pretty, but the effort has left me e-winded. Imagine the same thing for column M, but shifted over and with 80 instead of 120. I reversed one or two of the values relative to what you had, just because that happened to be how I set it up in the table.

    Edit: I can't follow your shorter duration tests (Bludgeon, Waterga, etc.), but remember when you're doing the math that they don't have flat recasts. Waterga is 12.25 seconds, for instance.

  8. #8
    Groinlonger
    Join Date
    Oct 2006
    Posts
    2,964
    BG Level
    7
    FFXI Server
    Fenrir

    Yes I had also reached this conclusion that my results would work if I used those numbers. I'm now currently trying to get a Vivid Strap to continue testing as there are reports of it having only 1% Fast Cast.

  9. #9
    Groinlonger
    Join Date
    Oct 2006
    Posts
    2,964
    BG Level
    7
    FFXI Server
    Fenrir

    Also all that my Bludgeon/Waterga tests showed was that there was certainly a difference between Incantor Stone and Gaiters/Loq Earring.

  10. #10
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    I think Nebula Cuffs +1 also only have 1% Fast Cast. They didn't change my recast alone, and I had no other odd Fast Cast things to test with so I could never get them to do anything.

  11. #11
    Groinlonger
    Join Date
    Oct 2006
    Posts
    2,964
    BG Level
    7
    FFXI Server
    Fenrir

    I'm gaining faith in this model for two reasons.

    1. It is most certainly being truncated due to Nebula Cuffs +1 and also Vivid Strap. Both of these items will not reduce the recast of any spell by any amount unless stacked with another piece. It also implies that truncation is done to Fast Cast before it is applied to recast.

    2. Given the definite knowledge of pre-applied truncation, I don't understand how it would truncate binary numbers in any way that made sense.

    I tested the following pieces today and also job traits.

    Code:
    Trait/Gear		FC R	FC C(Assumed)
    Fast Cast I		5.0	10.0
    Fast Cast II		7.5	15.0
    Fast Cast V		15.0	30.0
    Warlock's Chapeau +1	5.0	10.0
    Duelist's Tabard	5.0	10.0
    Orison Locket		2.5	5.0
    Incantor Stone		1.0	2.0
    Vivid Strap		0.5	1.0
    Nebula Slops +1		1.0	2.0
    Nebula Cuffs +1		0.5	1.0
    Loquacious Earring	1.0	2.0
    Augur's Gaiters		1.5	3.0
    You might find this useful.

    Spoiler: show
    Code:
    6 WHM	5
    27 WHM	5.25
    11 WHM	5.5
    63 WHM	5.75
    21 WHM	6
    52 WHM	6.25
    1 BLM	6.5
    75 RDM	7
    4 BLU	7.25
    15 WHM	8
    9 BLM	9
    1 BLU	9.75
    14 WHM	10
    31 WHM	10.25
    51 WHM	10.5
    71 WHM	10.75
    15 BLM	11
    16 BLU	11.25
    17 BLM	11.5
    18 BLU	11.75
    21 WHM	12
    19 BLM	12.25
    21 BLM	13
    23 BLM	13.5
    26 BLU	14
    30 BLU	14.25
    26 BLM	14.5
    15 WHM	15
    28 BLM	15.25
    30 BLM	15.75
    27 WHM	16
    32 BLM	16.5
    47 WHM	17
    36 BLM	17.75
    63 WHM	18
    38 BLM	18.5
    40 BLM	19
    44 BLU	19.25
    18 BLU	19.5
    42 BLM	19.75
    13 WHM	20
    44 BLM	20.25
    46 BLM	21
    48 BLM	21.75
    51 BLM	22.5
    38 BLU	23
    53 BLM	23.25
    60 BLU	23.75
    1 BRD	24
    57 BLM	24.5
    59 BLM	25.25
    75 BLM	26
    62 BLM	26.5
    63 BLM	27.25
    64 BLM	27.75
    65 BLM	28.5
    66 BLM	29.25
    30 WHM	30
    68 BLM	30.75
    69 BLM	31.5
    70 BLM	32
    71 BLM	32.75
    72 BLM	33.25
    50 BLU	33.75
    73 BLM	34
    58 BLU	34.5
    73 BLM	34.75
    74 BLM	36
    75 BLM	37.25
    61 BLU	39
    7 BLM	40
    50 BLM	41.5
    52 BLM	42.25
    54 BLM	42.75
    58 BLM	43.75
    62 DRK	44
    60 BLM	44.5
    37 NIN	45
    71 BLU	49
    61 BLM	50
    70 DRK	52
    61 BLU	56
    25 WHM	60
    71 DRK	180


    It's a list of recast times of various spells and then jobs that can use them. Put this into Excel and then make a few columns that apply the supposed formula to a few different hypothetical values of Fast Cast. Then make another column that checks to see they're all the same and then to flag if they aren't. Makes coming up with tests to determine the amount of Fast Cast on a given piece very simple.

    Also, there may be other jobs that can use the spell listed (look on wiki for the job > spell > jobs on spell), and there might also be other spells with the same recast (in case you find a recast that will work but the listed spell isn't on the right job or whatever, pebbles works best for this in the string list thing w/e)

    I have yet to encounter a spell that refutes this using Fast Cast gear only. The next step is to perhaps examine further more pieces of gear and also how haste interacts with it.

  12. #12
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    Awesome!

    Not a laundry list for you necessarily, but other values that aren't entirely hammered down.
    Spoiler: show
    Hochomasamune (10%?)
    Warlock's Mantle
    Winged Wand
    Vivid Strap +1


    As far as how Haste interacts with it, Haste is applied before -Recast and there's a floor step in between them I think. Otherwise you'd get goofy results with weakened reraise (100% Slow, so same step as Haste)

  13. #13
    Groinlonger
    Join Date
    Oct 2006
    Posts
    2,964
    BG Level
    7
    FFXI Server
    Fenrir

    Haste (150/1024) + Fast Cast V + Loquacious Earring produced a reraise recast of 43 seconds. So Kirschy was probably right about the retention of the 10ths decimal place before Haste is applied. I think more confirmation would be good on that, and then also maybe use march tiers to determine if it's truly as simple as n/100 for Fast Cast.

  14. #14
    Groinlonger
    Join Date
    Oct 2006
    Posts
    2,964
    BG Level
    7
    FFXI Server
    Fenrir

    I don't think that weakened is 100% slow, I think that it's recast x2. This stacks with Hasso (which is 1.5) and also probably composure (1.25, although I haven't tested weakness stacking with this one.)

  15. #15
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    No, it's 100% Slow. That's why 25% Haste with Weakness gives you 1.75x recast

  16. #16
    Groinlonger
    Join Date
    Oct 2006
    Posts
    2,964
    BG Level
    7
    FFXI Server
    Fenrir

    Well, a decimal is conserved either way because the recast would have been 42 seconds if not (doesn't matter which went first.) I didn't use any haste in my tests either, so I'll double check them to see if holding the decimal over during the weakened tests would have produced the same results as multiplying the base recast by 2.

  17. #17
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    Veela Cape - 2:00 Recast (1%)
    Rostrum Pumps - 1:58 Recast (2 or 3%)
    Veela Cape and Rostrum Pumps - 1:57 Recast (4%, so Rostrums are 3%)

    I can go kill my Mule to test Homam tomorrow night.

  18. #18
    Groinlonger
    Join Date
    Oct 2006
    Posts
    2,964
    BG Level
    7
    FFXI Server
    Fenrir

    Well I went back over my data, it wouldn't have mattered in which way weakness was applied although the decimal did have to be carried over.

  19. #19
    BG Content
    Join Date
    Jul 2007
    Posts
    21,105
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    Homam pants are 5% Fast Cast.

    Homam + Rescuer + 25% Haste = 99 Recast
    Homam + 25% Haste = 102 Recast

    Floor(105*.98 ) = 102
    Floor(105*.95) = 99

    That was higher than I expected.


    Teal set bonus is 10% Fast Cast, making it better pre-cast gear than Rostrum's + Nebula stuff for mage jobs, unless Nebula body +1 is 5% Fast Cast or more. Not incredibly relevant, considering most jobs got "Casting time -%" for most of their important spell types anyway. Need the whole set or you get no bonus. Rostrum Pumps gave me 118 seconds recast with 4/5 Teal.

  20. #20
    Groinlonger
    Join Date
    Oct 2006
    Posts
    2,964
    BG Level
    7
    FFXI Server
    Fenrir

    What spell were you using that had a recast of 140 seconds?