Jump to content
The Ninja Forum
Arphee

CFH

Recommended Posts

So-, It's been a long time, SHouldn't have left you but i'm back, with a much more fine tuned system with much broader reaching implementation than originally intended, This post was first started as a mean to push back the current limitations and exploitations of the CFH sweet-spots and has since evolved into a broad reaching and comprehensive solution to the way SF is handled, CFH ranges are balanced, and Counter Call's are executed during MPVP

There are also a number of systematic changes that alter the way calculations, logging, entry, and multiple parties joining are handled.

 

 

 

CURRENT METHODOLOGY AND FORMULAS

Under constant review and balancing, much work to do still, numbers nowhere near final.

 

SF Calculations

 

Now im sure by now most if not all of us are familiar with the way SF is calculated, and the mass SF inflation caused by  the defense and generals (Highest - Lowest + Total) function, i'm sure you've also ran into encounters where despite a person supposed to be as strong, or stronger than you, and the fight has gone easier than expected?

Well, thats because Someone with 10 Tai defense and 1.25M Buk defense is calculated as having both capped, the current way a persons "Strength Factor" is calculated doesn't factor the differences in stat training between users,  a 4000 SF Tai wall, is at a significant disadvantage against a 3500 SF Buk user.

So? How do we fix this?

Simple, eliminate the bloat, remove the Min and max functions, and label it Power, Raw Strength, the label is semantics, the formula is the new universal standard for stat progression.

Raw Power = (HP+OFFENSE+(TOTALDEF/4)+(TOTALGEN/4*10)/1000)

But a more detailed sollution is nessecarry to replace strength factor for inter-user engagements,.

SF =  (HP + Offense + (Total Def+ Opponent OffenseDef - Opponent Offense/4) + (Total Gen + Opponent OffenseGen -Opponents Gen*10/4)/1000)

 

Now, the mathematically inclined among you will instantly realize how different this formula is, and several glaring would-be-problems for instance this means a Capped Jounin fighting a Capped EJ would have 5812.5 and 7687.5 SF respectively, or that very weak users, could wind up with negative or little to no SF.

Fear not, for these issues have been addressed and balanced for via logic detection, Where in any time a players SF would be penalized below 87% it reverts to RawStrength*.87

Put it all together and Bam a much more fair and accurate display of power.

IF)(HP + Offense + (Total Def+ Opponent OffenseDef - Opponent Offense/4) + (Total Gen + Opponent OffenseGen -Opponents Gen*10/4)/1000)<(HP+OFFENSE+(TOTALDEF/4)+(TOTALGEN/4*10)/1000)*.87,(HP+OFFENSE+(TOTALDEF/4)+(TOTALGEN/4*10)/1000)*.87,(HP + Offense + (Total Def+ Opponent OffenseDef - Opponent Offense/4) + (Total Gen + Opponent OffenseGen -Opponents Gen*10/4)/1000)

Speaking of accurate displays of power,  and while this is the way my sheets are calculating things, there is an alternative way that allows for expansion of various factors in the future by moving player SF calculations into a separate field from raw.

Raw Power + ((Opponent Offense-Appropriate Defense / 4)+ (Gen - opponent Gen/4*10)/1000)

This would allow for future additions to the CFH rebalanced project that aren't feasible at the moment, for instance a modifier that increases a players SF by Raw Power * .0X based on equipped armor as an example, or some kind of (+ / -) mechanic based on the average of a players 4 highest tagged jutsu's levels

MPVP

 

We all know the current ranges for calling against people, they're fairly easy to digest, but beyond 1v1 things break down entirely into an unfair situational mess based solely remaining health at time of entry leaving scarce opportunity for counter calls, and often requiring to sacrifice your original kill so the entering party can get both- or how when the person who originally called dies your left completely unable to counter call despite being absolutely wrecked? Yeah well, i think thats all a load of bullshit. Ever notice how when Counter calls DO rarely succeed  and immediately escalate to being raped by an uber? Thats total bullshit too, and i plan on fixing ALL OF IT!


RSF

Gone are the days of an RSF system that forgets the ass beating you just recieved when someone kicks the bucket in a CFH situation,  RSF will be updated upon each new ENTRY into a battle against the user they're countering for their SF calculation, this RSF will remain the same regardless of HP loss or deaths of players  on EITHER side until a new entry on either side, and will not remove past entries.

Simply put the new RSF will always be the SF of each player on a side upon entry added together, devided by their opponents SF when they entered divided by  one another, no HP gimmicks, no tricks


RSF Limits and Entry

1V1 Things don't change much, but MPVP beyond the first call is worlds different than the current system.

Call Limits 1 V 1
.76~66 Round 4
.65~.55 Round 3
.54~.44 Round 2
.44=<    Round 1

MPVP
.76~.66 3 Rounds after call available (Disabled until entry)
.65~55 2 Rounds after call available  (Disabled until entry)
.54~44 Round after call available (Disabled until entry)
.44=< Round of call (Disabled until entry)

You may be asking "What in the world does "Disabled until entry" mean!" well as i'm sure many people are aware there would still be a way to cheese CFH, in the form of calling late, under the present system your CFH entry points are based on ENTRY  if your in a battle with someone that gets you well below half at that .7X boundary, them waiting until the round  before they would die (Round 7 give or take) to call, then battle-holding those thirty seconds would mean more than likely whoever enters would kill you before round 11, or round 10 with my reduced counter times,

Remember how RSF updates upon entry? and retains that information, well, were going to use the same kind of setup to ensure CFH remains balanced and fair, rewarding those who plan ahead and act / react expediently, Rather than being based off of when a player called, its based off of when a player was ABLE to call rather than time of entry,

An example

A Catches B,  B's RSF is .70, thus  B can call on Round 4 B instead calls on round 6.
On round 6 C enters, and player A's RSF becomes .69 Player A is able to call on round 7

The long an short of it is, the system will base call eligibility on call eligibility not utilization. the system remembers when the option was available, and if used will process when counters are available based on when they could call, not when they did.

MAX CALL SF

This is  probably the largest change from the original system, which had a bad habit of severe power escalation, under ideal circumstances when cheesing the system mid-game- when counters were successful, they were often successful to the tune of capped EJ's massacring the nabs who just barely managed to kill another capped EJ, Das not cool, cfh is a battle of attrition and should slowly de-escalate as it progresses, not become a meme

1V1
((2*Stronger Player - Weakerplayer)*Stepdown)*RSF

MPVP
(Enemy SF Total *Stepdown)*RSF\

 

Stepdown

1V1  .9
MPVP .8

 

 



T-T-THATS ALL FOLKS!

For Current data samples and tests feel free to veiw my spreadsheet
Remember, its like a strip club. 
Look, Don't Touch!!

Comment:

 

 

Right now things are a bit tighter than i'd like when it comes to the MPVP formula, and peoples ability to call beyond counters, but is extremely balanced and stable.

​At present its in more than usable condition, but I'm still far from satisfied, with more weeks and months of user test data, and  fiddling with numbers  and changing approach im sure i'll reach my intended desire of 40 round fights with more than 3 calls.

Till then, This is muh progress report, feel free to comment, criticize, and critique 
 

Edited by Arphee
  • Like 1

Share this post


Link to post
Share on other sites

TL;DR

 

Thats what she said, about my D

 

Actually now that i think about it in this context the joke doesn't apply.

 

Uh, CFH on a witty comment? Anyone

Share this post


Link to post
Share on other sites

Could you explain DefType and SubGen?

 

Also, what happens if an opponents offense exceeds your overall defense? Is it possible for weak players to have negative SF with this formula...? xD

 

Without understanding DefType and SubGen I can't really give accurate input. However, it seems like this formula actually results in SF that varies depending on your opponent? That adds a whole new level of complexity to CFH, and I'm not sure I like it.

Share this post


Link to post
Share on other sites

Could you explain DefType and SubGen?

 

Also, what happens if an opponents offense exceeds your overall defense? Is it possible for weak players to have negative SF with this formula...? xD

 

Without understanding DefType and SubGen I can't really give accurate input. However, it seems like this formula actually results in SF that varies depending on your opponent? That adds a whole new level of complexity to CFH, and I'm not sure I like it.

 

Def type is the offense of the opponents specialization Ae

 

Buk = Buk

Tai - tai so on and so forth

 

Sub gen is the offenses general EG

 

Buk = Speed

Tai = STR

Nin = INT

Gen = Will

 

And yeah, xD like i said, not fully fleshed out, i was barely awake when i finished it, im sure those kinds of issues could be fixed with an "or limit" function.

Share this post


Link to post
Share on other sites

Just in terms of mechanics, there are a couple issues.

  • I could just set my specialization to another offense I haven't trained at no penalty to me if my bloodline is not a "highest" bloodline. This would allow me to cheat the system by boosting my opponent's SF by ungodly amounts.
  • If the system is changed to detect the actual highest offense (stat-wise), how will it determine which the player uses in an actual fight to gauge what DefType/SubGen should be (assuming an end game player with 2 capped offenses).
  • You will have no idea what any of the numbers will be until you enter into battle with someone, because some variables are based on your opponents stats. This means there is no longer public profile SF, and you can't do any research ahead of time to see who you can call on/who will call on you/stat goals for CFH.
  • This disregards the fact that players in mid-game typically spam primary generals that can be trained simultaneously with their offense to boost their combat ability by a large amount, and puts a very large emphasis on secondary generals. It's a 70/30 split between primary and secondary generals. This formula might work well at end game, but it's even more broken at mid-game than the current one if you asked me. The current formula does weight the players highest general to provide a more accurate indicator of mid-game potential by adding it twice. This new one totally disregards that and seems like it could result in some really wacky matchups.
  • Think spammed primary general player vs. someone who spammed a defense that doesn't correspond... Neither player experiences a noticeable jump in SF vs. each other, but one of them trained a stat that will have a large impact on the matchup, even though the SF doesn't reflect that. To be specific, Jounin Ninjutsu user spams Speed vs. Jounin Genjutsu user who spammed Buki defense rather than strength. Speed and Buki defense won't be weighted because they don't qualify as SubGen or DefType respectively, and as a result the 2 players will be relatively even in SF because their Total Gens vs. Total Defs will balance out number wise. However, the Nin user will totally crush the Gen user because their stats are actually relevant to the fight, while the Gen users are not.

I feel like it's a pointless exercise to plug in numbers if the base model still needs tweaking. I don't disagree that the current CFH model needs to be adjusted, but moving it in a direction that isn't counterproductive in some way is actually more challenging than it appears.

Share this post


Link to post
Share on other sites

I feel like it's a pointless exercise to plug in numbers if the base model still needs tweaking. I don't disagree that the current CFH model needs to be adjusted, but moving it in a direction that isn't counterproductive in some way is actually more challenging than it appears.

On point 1 it could just auto detect highest offense the way the current SF formula does, maybe even penalize people who have mismatched specialzation and highest offense under the assumption of  dual offense after cap.

 

On point 2, it could use  a split function like the battle system's calculator has the functionality, with some logic functions and detection minimums it'd be easy to set something up, something along the lines of Countif valueA-D > (Highest*Threshold) then compare all trained offenses that have been trained high enough to use competitively

 

On point 3 SF on peoples profile would just display their SF against you personally, still easy to research who you can call against, you just have to do the legwork yourself.

 

On point 4 & 5 in the case of

 

J NIN SPAMS SPEED VS J GEN SPAM BUKI Def

 

Both players under the current system would both be about equal as well, both defense, and Gen spam get counted as double for your highest stat, and the nin user would still have the overwhelming advantage because they trained optimally and raised their damage output rather than tankiness.

 

Because im having a hard time imagining what this would look like in my head, im going to just do the math for 2 months worth of jounin stats + missions and EM's, then make training allotments.

 

So yeah, looking at a microcosm it looks like my system would actually be better, and more accurate under this test case scenario.

 

J NIN SPAM SPEED= 984.93 Under current SF rules 896.47 under Arphee Proposed

J GEN SPAM BUKDEF= 991.65 Under current SF rules 889.75 under Arphee Proposed

 

The current system says that the Speed spammer is at a 6.72 DISADVANTAGE with the way things currently are. My system says the Buk spammer is at 6.72 disadvantage- which is funny, i didn't think it would be the exact opposite given the drastic changes to how things are calculated, In any case im going to do this again, with 6 months worth of stats and see what it pushes out.

 

e.e

 

Okay now its driving me crazy, at 6 months of progress both systems still just put out opposites, mine saying the Gen user is at a disadvantage, which he is, the current system saying "Hue hue nin are weaker with dem gens", i don't understand the mathematical principal behind why my system works the way it does spitting out the opposite despite drastically different criteria

 

 

Point in case, its a small improvement over the current system in this use-case scenario, the major difference being CFH getting addressed via a reduction to .76, something literally impossible under the current systems calculations, .87 is the minimum possible without literally destroying everything, within .87 range just doesn't work, and isnt remotely balanced without the steep drop off between the rank differences, a .87 Jounin will have far less throughput than a .87 EJ against the same player, EJ being able to cfh is a great addition to the system, even if it makes Chuunin and Jounin slightly less desirable, especially given how easy it is to exploit at present.

 

Fact is the current system cannot be fixed with a, or multiple small tweaks given the nature of how it works being related to stat caps and the ratios between them.

 

EDIT: Got rid of the excess quote text, this shit is waaaaay too long as it is, apologies to those that have to read it x-x;

 

Edit Again: Did some more tetsing for fun with 100% pure spam + offense for midgame, and without the more evenly distributed training regiment, they actually produce identical results, just lower, This is fascinating, i feel like a stoner having my mind blown by math x-e

 

OLD SF   2845.58   2845.58           Arphee SF   2448.01   2448.01 Edited by Arphee

Share this post


Link to post
Share on other sites

Well you used the same equation with different data points so the resulting SF won't be drastically different from the original in most cases. I'm not sure what the server load would be like calculating a players SF based on your stats for every individual who views their profile but that might not be ideal (and unless we're gonna start sharing user stats publically there's no other way to know what the SF lineup would be).

 

My main gripe is that your proposed solution doesn't solve any of the current flaws with the system... and I still haven't seen how it interacts with the other equations involved in CFH. Maybe I'll run some numbers later and see what the final results are rather than guesstimating and basing my opinion on that.

Share this post


Link to post
Share on other sites

Well you used the same equation with different data points so the resulting SF won't be drastically different from the original in most cases. I'm not sure what the server load would be like calculating a players SF based on your stats for every individual who views their profile but that might not be ideal (and unless we're gonna start sharing user stats publically there's no other way to know what the SF lineup would be).

 

My main gripe is that your proposed solution doesn't solve any of the current flaws with the system... and I still haven't seen how it interacts with the other equations involved in CFH. Maybe I'll run some numbers later and see what the final results are rather than guesstimating and basing my opinion on that.

 

You can have the machine itself process it, its basic math, the current SF already draws from the exact same statistics, TBH the new formula i propose wouldnt be too different from how thinsg are right now in terms of per capita stats V stats, but it gives more flexibility to adjust the RSF and call statistics, which is my main concern at this point.

 

A lot more work needs to be done beyond Person V Person SF amounts and the adjustment of RSF minimums between ranks to make CFH work as originally intended, but i think my suggestion is firm step in the right direction, Do you know if pana or anyone else is interested or looking at this because i seriously want to help hash the issue out, and make sure it doesnt have unforseen vulnerabilities down the line like the previous 3 implementations have had.

 

Honestly, id appriciate being kept in the loop or knowing which way the staff is leaning towards these suggestions, no point in theory crafting further if the implementation cant or wont happen.

 

As a stop-gap  / alterneative measure at the very least there should be a % Penalty to allowable RSF when using the system out of originally intended ranks.

 

each rank above has higher caps and are capable of more throughput making a 4800 SF chuunin 4800 SF jounin and 4800 SF EJ have drastically different combat potential, despite stat "equivelency", furthermore EJ VS Ej CFHs shoudl haev a drastically lower SF RSF threshold since there is no throughput disadvantage.

 

Issue with this is its just as, if not even more convoluded, forces people to do real mental gymnastics to determine who the ycan call against and who they can call in, and is overall far more impractical netting 1 SF but forcing the use of 7 RSF's

 

It really is a tough issue to tackle, especially when your trying to be as fair as possible while restoring the utility of the lower ranks, there is no easy fix like there was for making CFH not the most broken PoS in the world (PS Thanks for that Ven-chan u did a good job by telling them to add the *rsf portion of the formula (o-o)b

Share this post


Link to post
Share on other sites

Damn, SF was never meant to be this complicated...

Neither was CFH, but by intrensicly tying the two together to create the current system we have you get backed into a corner.

 

CFH was always designed with the intention of lower ranked players calling other lower ranked players in against higher ranked players, and under this use-case scenerio CFH works as intended, with a couple of problems showing up with the countercall mechanics.

 

Now the simplest sollution would just be to  kek loudly and disable EJ vs EJ cfh entirely, and to limit J vs EJ and C vs J to only calling in jounin / chuunin within current SF ranges, an only allowing calls if they meet current SF requirements.

 

Thats the easiest thing to implement in honesty, but its also a heavy handed sollution that doesnt feel fair, would make my life a million times easier, but a 3000-4000 SF EJ getting rekt by a capped jounin deserves the right to call, the same way a 5000 SF ej's deserve the right to call on a capped EJ and bring in a friend about the same strength.

 

:/  necessity is the mother of innovation, and simplicity is elegance of design, my suggestion really is the best universal concept i can think of to provide a more flexible sliding clase that restores the importance of rank versitility while providing fair play for midgame EJ,  only real issue with my system that bugs me is the effects and possibility of SF dropping into the negatives or extremely low when the stat gap between players is extreme.

 

Best sollution i can come up with for it would be detection for when the SF penalty would be larger than a threshold EG X% of raw SF (HP+OFF+(DEF/4)+(GENS*10/4))/1000 and substituting the negative or small value with Raw SF*X 

 

I was planning on testing this method and implementing it into the previous sheet i made but something came up, sort of having a really shitty day, in any case, thats the theorycrafting behind it, not too hard to imagine how it works in actual application.

Share this post


Link to post
Share on other sites

bleep

 

However, I guess it could be cool to have an event day where one of the days in the week is to disable cfh. 

Would be much better than half priced item shop day. xD

Edited by KENSHlN

Share this post


Link to post
Share on other sites

In my heart I was more of a fan of just going back to single combat

 

The entire premise behind tying SF to CFH was to balance the scales regardless of rank, but there is no perfect solution, there never will be.

 

I think we came closest to something workable in dev, not sure what happened since then...

  • Like 1

Share this post


Link to post
Share on other sites

The originally released CFH feature with SF was used by players with low RSF (think .2-.5) to call in players that actually exceeded their opponents by a large amount. Wake up and see a weak Jounin on your village, attack, find yourself in battle with a top 10 EJ on round 2. It actually got to a point that there was a player who ran around baiting it constantly, though I can't remember who it was now. Was pretty easy to abuse.

Share this post


Link to post
Share on other sites
Guest Faithy

I dont understand cfh anymore... I fight a jounin who outdamages me every single round and he can call in another jounin who is equal his strength x.x 

 

there is a conspiracy against chuu chuus 

 

 

131 Combat Band Lost

130 Combat Ryukon Lost

 

I couldn't even kill 1. >:C

 

rebellion!!!!!!

 

I bet they got 20pvp exp each. Where is my AoE and Summon? 

Edited by Faithy
  • Like 3

Share this post


Link to post
Share on other sites

Sounds like a hitch in implementation, the design was pretty solid, CFH could only ever balance SP to within some defined range.

Share this post


Link to post
Share on other sites

20? My dear, this was on 100% increased gains day, I got 39.4 for that little skirmish~

I outdamaged you, but you had more health, defense and a home field advantage! I could hear the people in Silence cheering for you and throwing whips into the fight to aid you. Our damage really wasn't that far apart and can probably be explained simply by J armor > Chuu armor and or jutsu levels. You were about 700 Sf higher than me, if you can't call on someone that far ahead of you and bring in someone near your strength, Cfh becomes pretty useless.

 

P.S. Ryukon got more than 20 cause he's a bawss

Share this post


Link to post
Share on other sites

It is so rare you see a CFH scenario that is close to resembling "balanced" because if the 2vs1 is close they can counter call and live easily. If the 2v1 is one sided counter calls are not even an option. I agree with Cedzen, scrap CFH and make everyone raid solo.

  • Like 3

Share this post


Link to post
Share on other sites

CFH can work both for and agaisnt you, that's how it makes sense afterall. 

I say pick your fights better. 

Share this post


Link to post
Share on other sites

Pick your fights better? So attack people who are weaker than you, but not too much weaker. Good point, well made. I'm pretty sure all popular PvP games work like that.... Right? D:

Share this post


Link to post
Share on other sites
Guest Faithy

I thought life was gonna be easy as endgame chuu chuu when I decided it. At the start of core3 nobody could cfh on chuu chuu because only the lower rank can cfh on the higher rank. 

 

But it all changed when SarahMcGigolo and Kenshin-obaachan locked down all villages. Say hello to SF based cfh... 

 

now juujuu who are 1 month away from surpassing my stats can cfh in another juujuu of same strength against me and endgame juujuu lower my hp to 20% before I can call on them >:C bad pana!

 

I am too lazy to rank up and spend 2 years on EM quests doe... so I will just look at slayze and cheer him on before he burns out as well. 

  • Like 2

Share this post


Link to post
Share on other sites

The original formula worked like this: (Stronger Team SF x 2) - (Weaker Team SF)

 

It was sound in theory when you practiced in optimal CFH settings. It prevented strong players of lower ranks from abusing CFH. It allowed different ranges for different strengths. What it didn't do was account for players who fell outside the optimal range of .54-.87 RSF. Here's what it looked like in one of those situations. Assuming a 4000 J attacks a 1500J : (4000x2) - (1500) = 7500... Suddenly the number 1 player in the game can join that call and absolutely murder that poor Jounin... and they can join on Round 2 provided the 1500 J doesn't get one shot. That was absolutely 100% broken and ruining PvP for the duration that it worked like that.

 

The fix that was implemented was this : [(Stronger Team SF x 2) - (Weaker Team SF)] x Weaker Team RSF
 

This fixed some of the inherent flaws of the CFH system by restricting the SF of a player who could join based on the strength difference between the two original fighters. In other words, the player who joins a call can never be stronger than the player they are joining against. That's a very basic, and very important concept for balancing the CFH system. If the player who joins can kill the player they are joining against in a 1v1, then there really is no basis for fairness. This new formula works similar to the old one, in that the player who joins a call can be stronger depending on the RSF between the original fighters.

 

Something else the original formula failed to do was account for multiple CFH, in which a 1v1 becomes a 1v2, then a 2v2, then a 2v1, and then a 1v1 again (with the potential to loop again before reaching the limit of 3 players per side). The issue lies in the fact that when a team is fighting, their SF is added together as if they were one person. This results in the system assuming a player is against 1 really strong player, when in fact they are against 2. This is a problem for obvious reasons, as 1 player with high SF can wipe the floor with 2 people who add up to the same SF. This means that 90% of the time in a 2v1 situation, if the solo player can call and make it a 2v2 he will win.

 

Currently 2v2 is nearly impossible due to a few variables that *also* went unaccounted for when a 1v2 situation arises, namely

1) SF recalculations when a new player joins the fight

2) SF recalculations when a player dies

 

Due to the way SF was calculated during these two instances, the player who fought against 2 enemies died nearly every time without the ability to call back. And as I said before, the current formula allows for a player who survives to call back and get a guaranteed win for their teammate once the 2 issues above are fixed.

 

So no, I don't think it was solid. It worked in a very specific range and fell apart outside that range very quickly. It didn't account for a variety of situations and attempt to provide balance. It still needs work. I'm not attacking anyone or claiming that it could have been done better, it takes a lot of foresight to predict every possible outcome. This is what patches are for :D

  • Like 1

Share this post


Link to post
Share on other sites

I wish I was a month from capping!

But very well said, Ven. I too remember the time when raiding Samui meant certain death by cfh (Or Titus...) and I like the new formula in comparison. However, I do think it needs to be tweaked juuust a bit. Really I just feel you should have the option to call in backup as soon as another person joins the fray... As long as it's not a capped Sf Jounin baiting a 2k Sf whatever to call just so they can call in a friend and get more points. Cause that's just hilariously cruel.

Share this post


Link to post
Share on other sites

Well the old system let rank holders chain CFH a higher ranked player until they won. It would be hard to develop a system worse than that. To account for variables at mid-game you would need to over-complicate the formula like Arphee did and at that point nobody will care and you will create a new set of issues (such as rolling SF values)

You can't have a balanced CFH system, you can just make it slightly less appalling every update.

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.

×
×
  • Create New...