Monk Limb Counter Derivation

So I've been working on getting better numbers for a monk limb counter. I've been staring at these numbers til my eyes bled, so I thought I'd crowd source it at this point to try and get more data in a more easy fashion. 

As close as I can determine the equation for unstanced HFP with no knuckles simplifies to

d = 3/50*h + 153

Where h is the max health of the target. This assumes that the target's limb health and max health are equal, which seems sensible to me, but can certainly be debated.

There's about a 2.5% error here (From 500h to 11500h, the possible linear functions result in a different number of punches to break for 289 health values) We could certainly trim that number down by filling in the gaps between differing number of punches to break.

I also started getting data on dragon and scorpion stance, and with and without level 2 knuckles, but its not nearly enough to derive a function from.

If you're interested in helping, the dropbox link for this excel file is https://dl.dropboxusercontent.com/u/102618736/LimbCounter.xlsx

If you're willing to gather data for this project, please post it here, so at least it will be public and peer reviewed-ish. I'm concerned about getting trolled with false data, but this is a shit ton of numbers to gather, so... yeah.

Thanks in advance!
image

Comments

  • Quick update:

    Without considering a stance modifier, the knuckle modifier seems to apply to both the static and scaling part of the equation. Assuming the base equation as stated above is correct,  and the knuckle modifier for level 2 knuckles being 1.25, the equation would resolve to 

    d = (1.25)(3/50)*h + (1.25)(153)

    Also, according to both my old limb counter and the data I've gathered, scorpion stance does not affect punch limb damage.

    *Disclaimer for this post is the data used to derive the information here is far less than the data I used to derive the base equation in the OP. It just looks like a nice, simple equation that fits the data I have. +

    Again, any data that anyone can contribute would be more than helpful. There's still a lot of stances, kicks, and punches out there I have 0 data on.
    image
  • I'll take a look at it using what I know. By the way, sawbones generally has doubled the precision of any particular sample for me, so far. Very worth it for any in-depth limb damage testing.
  • Yeah, your assessment generally fits my model for unstanced.

    The 2986 HP one is weird, though. Not in a way that 10 hits would fit it instead, but it seems to do just slightly more than the expected amount of damage.

    Are you sure about health points 1860 and 9100 not being 7 and 13 instead, respectively?
  • @Nim Just checked my logs, confirmed at 8 and 14. Why do you think they're lower?

    Also, I've got an anomaly, in that WWK differs from SNK on only a single data point I've gathered so far... unstanced, 5582 health. If anyone else could confirm that WWK and SNK are indeed different, I'd feel a lot better about it.
    image
  • They're edge points, and if they were and the base damage were 1.5 higher (dagger testing suggests damage works in intervals of 1.5 HP, though I suppose there's no guarantee that's not just how daggers in particular work), then the 2986 point would have worked as well.

    Otherwise, it sort of doesn't, unless there's some really bizarre rounding going on or something.
  • @Jacen You can do itttttt!!!! All. Night. Long.
    image
  • Excel file has been significantly updated. Excel File

    I'm kind of exhausting the health pools in Hashan. Any blanks that anyone could fill in would be very helpful. 

    I'm wondering if maybe JBP is the base formula, and everything else scales in some manner from that. I'm not sure, but that's why I added it to the spreadsheet.
    image
  • Ok. 

    Base HFP is d = 0.06h + 152.5. Equation has a 1.1% difference in possibilities
    Base JBP is d = 0.04h + 95. Equation has a 3.7% difference in possibilities.

    The long accepted Dragon multiplier of 1.3 appears to be correct.
    Scorpion stance does not modify punch damage.

    Both knuckles and stance appear to multiply the whole equation. Therefore, the full equation for HFP might be 

    d = (knuckle_mult)*(stance_mult)*(0.06h + 152.5)

    I had thought they would relate back to the same base equation, but it doesn't appear so. Not without separate multipliers for the scaling and static portions.

    Shit I wanna change class now.
    image
  • Excel file has been updated. 

    I can't find a linear function to fit unstanced HFP now. The data so closely represents a linear function that I can't imagine it isn't; there's very little to be gained by using a more computationally expensive algorithm like the log function.

    I've tried truncating and ceiling the calculated upper and lower bounds, but neither changes the data enough to allow a line to fit it. I've double checked my logs to confirm the number of punches I used in the cases that are preventing a line from fitting, but I see no errors there.

    If someone else could take a crack at these numbers, I would be most thankful.
    image
  • any updates to this?
  • Not from me, I'm both no longer a monk and not playing anymore. Feel free to pick up where I left off though!
    image
  • The link doesn't work...
  • thanks!
Sign In or Register to comment.