[WIP] Gameplay design and balancing - CR1 - Freeyorp

[Bertram]

Hi,

I think I've got one equivalent for HP!

I replaced both log and square root part with a bit more linear ones:

Code: Select all

max HP: 150 - Level + Vitality x (Level + 15) / 10

The given HP value as very alike the current vit x 8 + lvl x 3 formula.

Better?

Regards.

[Rotonen]

Minus level? Why?

[Bertram]

To simulate the log(x) curve, ... in less curvy .

So you've got an HP boost at first levels which fades away while you're growing.

[Jaxad0127]

Remember, level is a derived attribute, based on the combined levels of all your skills (or so). Do you really want increased weapon skill leading to more HP? ("You're now more adept with the hammer! Here's more HP for you.")

[Bertram]

The very first hp formula was already based on level. I have no problem with this.

("You're now more adept with the hammer! Here's more HP for you.")
I have no problem with that either. Counter-proposal?

Regards.

[Crush]

"You're more adept with hammers, here are five attribute points you can spend on intelligence"

Sorry but scaling the hp with level isn't less plausible than the rest of the game system and I think that it makes sense.

Or do you want a high level magic user without vitality to have just as few hit points as a level 1 char?

[Bertram]

2 vs 1. For now, Crush and me win

Any other comments about HP or something else?

Best regards.

[Bertram]

Hi,

Ok, I've upgraded the file with the latest formulas, here:

ManaServ - Statistics System - 0.1.0.zip

(69.45 KiB) Downloaded 100 times

I also added in it the mana experience table from Crush, taken here:
http://forums.themanaworld.org/viewtopi ... =24&t=8731

And I added a manasource WIP wiki page upon all this, here:
http://doc.manasource.org/manaserv_statistics_system

Feel free to comment!

Cheers.

[Bertram]

Hi,

The current hit accuracy function was still based on a squareRoot:

Code: Select all

(Dexterity + Weapon Skill + 50) / 2 x ( (6 - SquareRoot(RangeInTiles)) / 5)

To avoid performance problem when using the SquareRoot function:
I translated the part using the root with a more linear function:

Code: Select all

(Dexterity + Weapon Skill + 50) / 2 x (-0.031 x (2 + 2 * RangeInTiles) + 1.12)

This function gives very similar results than the former one for the first ranges,
and bit less for ranges near to ten tiles and above.

Best regards.

[Freeyorp101]

Bertram, and anyone else interested:

I've begun work on a rather different set of systems ([Link]) in light of the result of [this poll]; RHS attributes don't scale with level, but rather only scale with the LHS attributes as defined by the user - the upshot of which is that stat points have more effect per point, as they are no longer being implicitly used toward max hp gain with each level.
Still somewhat provisional, but have generally gone through some testing before I've posted them there. I'm currently working through a partial rewrite of attribute and attribute modifiers in manaserv to allow for the flexibility I have in mind.

Comments and suggestions welcome.


---Freeyorp

[Crush]

What is the +50 for? Remember: hit accuracy is not a percentual chance.

[Bertram]

Hi Crush,

Remember: hit accuracy is not a percentual chance.

Actually, you already said that in that very same thread: http://forums.themanaworld.org/viewtopi ... 916#p86916

Hit accuracy is not a percent value. Please see http://wiki.themanaworld.org/index.php/ ... nd_dodging for how it works. I also think that it should be based on dexterity instead of agility and that there should be a range penalty.

What is the +50 for?

The +50 is added to avoid the first level low points problem. The current formula was getting too quickly too high as said here:
http://forums.themanaworld.org/viewtopi ... 953#p86953

Even if it's not a percentage, I still like this new hit accuracy formula, which is less higher than the former one for the upper levels, IMHO.

The new formula aims at being much more balanced than the current one.

Best regards.

[Bertram]

Hi Freeyorp,

I carefully read your proposal, and I already thank you for the strength put in it.

I have several points I hope you'll fill to help mine and others comprehension:

- Is this proposal intended to be primary implemented in a TmwAthena (eAthena) server?
(Just as the pseudo-code made me feel that way.)
- Would it be possible to see value for an average character, at least on the first 20 levels to get an idea
about how things are balanced and get any gap along the road? (The better way for me is an ODS/XLS sheet, so we can actually see the links between the data and the formulas.)
- Maybe it's just me, but I think some example for stackable/non-stackabe attr/bonus could greatly help to get what you mean.

As I'm not willing to double work on this part, or step on anyone's toes, and as my work is based on the current ManaServ status, to minimize the development time cost, I wonder if we were trying to implement two new systems, one on eAthena, and one on ManaServ. If it's the case, it's fine for me, even if it would be much better to merge efforts about the concept.

If you're willing to port this to ManaServ, we definitely should see how to merge our both ideas, with necessary comments from Crush, Rotonen, and the other devs - by devs, I don't mean only coders, just to be clear.

Best Regards.

[Freeyorp101]

- Is this proposal intended to be primary implemented in a TmwAthena (eAthena) server?
(Just as the pseudo-code made me feel that way.)

It's primary designed for manaserv - you'd probably see a lot of C isms in my psuedo-code since they're mostly taken from a primitive C simulater I made during testing.

Backporting simplified versions of some of these systems to tmwAthena would still be nice, since it would allow some early testing/data with a fairly substantial playerbase (and hopefully fix tmwAthena gameplay at the same time)

- Would it be possible to see value for an average character, at least on the first 20 levels to get an idea
about how things are balanced and get any gap along the road? (The better way for me is an ODS/XLS sheet, so we can actually see the links between the data and the formulas.)

Would .gnumetric also be an acceptable format?

Also, I don't think it's entirely right to refer to an "Average" character, given the flexibility it seems we're going to give characters, even early on - I did take your suggestion of having 100 starting attribute points, and worked from there (I must say, it was rather nice to have some reasonably fine granularity as to distribution of points )
Incidentally, any thoughts as to the starting number of correction points? Given the fairly steep curve that [this poll] seemed to indicate most people were in favor of, this should also be a fairly important point.

As I'm not willing to double work on this part, or step on anyone's toes, and as my work is based on the current ManaServ status, to minimize the development time cost, I wonder if we were trying to implement two new systems, one on eAthena, and one on ManaServ. If it's the case, it's fine for me, even if it would be much better to merge efforts about the concept.

I've tried to avoid directly going over places that you've already started work on; though I did made an exception in the case of the max-hp / hp-regeneration formula, which I did feel needed some reworking after the results of the poll (though I do think your natural regeneration formula gave some rather high values after I had previously mentioned I'd intended items to do most of the work there )
I don't suppose you'd consider dropping by on IRC a bit more? I personally find it a easier medium to work on when many people are working on closely related things (plus it would be nice to see you around some more )


---Freeyorp

[Bertram]

Hi Freeyorp,

Ahhh, it's nice to get such positive and productive feedback

It's primary designed for manaserv

Ok, let's team up!

Backporting simplified versions of some of these systems to tmwAthena would still be nice, since it would allow some early testing/data with a fairly substantial playerbase (and hopefully fix tmwAthena gameplay at the same time)

The eA side, read tmwAthena server code upgrade, or better current stable eAthena take-over, needs a leader upon strategy and balancing. Jaxad, MadCamel, and others I'm not aware of, have done a lot of fixing already on the old-2005 based tmwAthena, and this server needs to be remade with the current eAthena server + specific security fixes, I guess. It was the role I first thought you wanted to take.
I won't go too far in that point as I want to focus on ManaServ only, if possible. But anyway, once we agree on the concept, the application can be reviewed in both.

Would .gnumetric also be an acceptable format?

No problem

Also, I don't think it's entirely right to refer to an "Average" character, given the flexibility it seems we're going to give characters, even early on

Indeed, that's what I called character profiling, which can be also done in the statistics point of view. I wanted you to start with the "average character" test because it'll help at least me to get some balancing first gaps. Then, with this as a base, we can start some profiling in many ways. Anyway it's just a proposal and you can start a profiled character as long as it helps us and the others involved to see clearly the link between the data and their formulas.

Incidentally, any thoughts as to the starting number of correction points? Given the fairly steep curve that [this poll] seemed to indicate most people were in favor of, this should also be a fairly important point.

Indeed. Mea culpa, what is the current curve, rule or formula already? I'll make some test upon it once I'll have it back.

I've tried to avoid directly going over places that you've already started work on; though I did made an exception in the case of the max-hp / hp-regeneration formula, which I did feel needed some reworking after the results of the poll (though I do think your natural regeneration formula gave some rather high values after I had previously mentioned I'd intended items to do most of the work there )

the max-hp/regen touches directly the percentage of damage, time to kill, and battle time stats values. So, some other formulas will have to be changed also if we lower the max-hp one. Yet, I still didn't see the formula you proposed or I misunderstood it with the one Kage gave earlier?
Once max-hp is "set in stone", it will be much easier for both of us to get a precise idea of what regen speed we want, I guess.

I don't suppose you'd consider dropping by on IRC a bit more? I personally find it a easier medium to work on when many people are working on closely related things (plus it would be nice to see you around some more )

It can be hard for me to get the time, the access or the right to activate IRC on my computer depending on the fact that I'm not forcefully at home when writing posts
Plus the fact, that posts help me to read carefully the given points of view and make a better analysis upon written things, sometimes. We should still "write" important things in this thread IMHO, to let the opportunity to the others to react.
Also my family will always be placed before such an occupation, I hope you understand it.
Anyway, I'll try to catch you next time we both meet on irc.

My best regards.