PDA

View Full Version : Revising the weapon skills



adom-admin
03-22-2008, 06:54 PM
Right now I'm thinking about how weapon skills will be handled in JADE. There will be some similarities to the skills in ADOM and some differences. Similarities include:

Weapon skills will be improved by practicing with weapons (e.g. fighting with them).
Weapon skills provide bonusses to hit, damage and DV.Additionally there will be some additions:

The most important first: Weapon skills in JADE (like attributes, standard skills, etc.) will be open-ended, e.g. there is no limit of 15 levels like in ADOM. As a consequence effects of weapon skills will need to be smoothed out a bit more to prevent game-breaking effects.
Weapon skills will increase the probability of critical hits: Each weapon in JADE has a critical hit probability (a percentage) that determines if a hit is critical. Critical hits provide a damage multiplier (either constant or variable depending on the weapon) that increases the total damage.
Weapon skills might increase the critical hit multiplier (see above).
Weapon skills might provide "other" special effects (e.g. a chance for stunning opponents when using blunt weapons or a chance for increased bleeding effects for piercing weapons) which become available at high weapon skill levels.What I am right now really interested in is some input on how to model a new weapon skill system without the flaws perceived in the ADOM system. Please note that this mostly refers to the points #1, #2 and #3 above. Right now I'm not really interested in a discussion of "cool" weapon skill special powers - that's something for a later point.

Thanks for any input!

yisk
03-22-2008, 07:32 PM
How about weapon division by the class of damage?

I suggest to allocate:
- Pricking attacks (spears, arrows, daggers)
- Cutting attacks (knifes, swords, scimitars)
- Hit-cutting (i really cannot find a suitable english word) attacks (axes, two-handed swords)
- Blunt attacks (clubs, hummers, rocks)
- Biting attacks (lashes, whips)

It is logical that the sheepskin coat is protected is better from blows by a club, than by a spear. The chain armour, in turn, protects from daggers and swords more reliably. It is besides, obvious, that for evasion or blocking different types of weapon (for example, spears and whips) various skills are necessary. I want to tell, that DV and PV should become complex values.

P.S. This idea is not new. Even in "AgeOfEmpires2" there is a division of the "near" and "arrow" protection.

theotherhiveking
03-22-2008, 08:46 PM
I'd not like to see 53 diferent weapons skills, one for daggers, one for short swords, one for spears, one for shiny axes, one for not that shiny axes...


I mean, are a sword so diferent form a scimitar that there are the need to split it in two weapon skills?

http://i.xanga.com/Dexargile/Dex%20Scimitar.jpg


http://www.medieval-fightclub.com/images/medieval-battle-sword.jpg



or polearms of spears?

http://www.thebladeshop.com/ProductImages/coldsteel/95BOA.jpg

http://www.dragonquestfrontiers.com/sitebuilder/images/polearms1-387x265.jpg


daggers, swords (short or long, it doesnt matter), spears (or polearms) unarmed combat, mazes, axes bows and crossbows are more than enough..


and, of course, shields

adom-admin
03-22-2008, 09:06 PM
How about weapon division by the class of damage?
I suggest to allocate:
- Pricking attacks (spears, arrows, daggers)
- Cutting attacks (knifes, swords, scimitars)
- Hit-cutting (i really cannot find a suitable english word) attacks (axes, two-handed swords)
- Blunt attacks (clubs, hummers, rocks)
- Biting attacks (lashes, whips)


Right now (as of the moment of typing) I have implemented the following weapon classes (1H = one-handed, 2H = two-handed):

unarmed
daggers & knives
blunt (1H)
blunt (2H)
ball & chain (1H)
ball & chain (2H)
swords (1H)
swords (2H)
axes (1H)
axes (2H)
polearms (1H)
polearms (2H)
staves (2H)
whips
bows
slings
crossbows
thrown weapons
It is logical that the sheepskin coat is protected is better from blows by a club, than by a spear. The chain armour, in turn, protects from daggers and swords more reliably. It is besides, obvious, that for evasion or blocking different types of weapon (for example, spears and whips) various skills are necessary. I want to tell, that DV and PV should become complex values.


I have decided against the latter. I would like to keep armor simple enough to be able to display the armor values in a simple expression like ADOMs [+3, +5] to signify a DV bonus of +3 and a PV bonus of +5. If I had to differentiate that even more for the damage types available in JADE it would become too complicated. Right now JADE knows the following damage types (which are subclasses of each other as one would expect):

physical (piercing, slashing, blunt)
energy (fire, electricity, acid, cold)One might consider one value for energy protection and three for physical damage but I know that someone will say "Hey, fur armor protects much better against cold that against..." and so on. So I have decided to keep one protection value for all damage types. Maybe fur armor will provide "cold resistance (5)" as a special ability but that is nothing to directly include in the PV.

theotherhiveking
03-22-2008, 09:45 PM
Right now (as of the moment of typing) I have implemented the following weapon classes (1H = one-handed, 2H = two-handed):
unarmed
daggers & knives
blunt (1H)
blunt (2H)
ball & chain (1H)
ball & chain (2H)
swords (1H)
swords (2H)
axes (1H)
axes (2H)
polearms (1H)
polearms (2H)
staves (2H)
whips
bows
slings
crossbows
thrown weapons
I have decided against the latter. I would like to keep armor simple enough to be able to display the armor values in a simple expression like ADOMs [+3, +5] to signify a DV bonus of +3 and a PV bonus of +5. If I had to differentiate that even more for the damage types available in JADE it would become too complicated. Right now JADE knows the following damage types (which are subclasses of each other as one would expect):
physical (piercing, slashing, blunt)
energy (fire, electricity, acid, cold)One might consider one value for energy protection and three for physical damage but I know that someone will say "Hey, fur armor protects much better against cold that against..." and so on. So I have decided to keep one protection value for all damage types. Maybe fur armor will provide "cold resistance (5)" as a special ability but that is nothing to directly include in the PV.



the skills cross-train? I mean, using a 2H weapon should share around 60% of the exp with the 1H one..

yisk
03-22-2008, 09:53 PM
I would not began to divide the two-handed and one-handed weapon. Instead, for each type of weapon it is possible to create an additional bonus multiplier which is added to skill of possession by a sword or an axe if this sword or an axe are two-handed.

Right now JADE knows the following damage types (which are subclasses of each other as one would expect):

* physical (piercing, slashing, blunt)
Yes, I wished to offer such division of the weapon!

Weapon of any kind cause a damage
1) on a surface (hammers, rocks) or
2) on a line (axes, swords) or
3) in a point (spears, arrows).


I would like to keep armor simple enough to be able to display the armor values in a simple expression like ADOMs [+3, +5] to signify a DV bonus of +3 and a PV bonus of +5. If I had to differentiate that even more for the damage types available in JADE it would become too complicated.
Well, I agree, that the realism is not important, if it harms to game process.

P.S. Hah, if in JADE really there will be snow fields, would be cool to see there a cap "shapka-ushanka" which gives resistance to cold weather, but the hero becomes almost deaf. xDDD http://img1.liveinternet.ru/images/attach/b/3/21/31/21031722_wewwre.gif

Dougy
03-22-2008, 11:18 PM
You'll need to use some kind of logarithmic scale at high levels. I tested this but it's a bit inappropriate at very low levels. So for levels 0 up to 15 I suggest a linear scale, then after 15 revert to the logarithmic.

For example the current to-hit in unarmed combat looks like this


UNARMED FIGHTING 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
To-Hit Modifier 0 +1 +2 +2 +3 +3 +3 +4 +6 +6 +9 +10 +10 +10 +12 +12


So I fit this to MIN(0.88*level,4.17*ln(level)) giving the values (after rounding):


0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
0 +1 +2 +3 +4 +4 +5 +6 +7 +8 +9 +10 +11 +11 +12 +12 +13 +13 +13 +13 +13 +14 +14 +14 +14 +14 +15

The constants 0.88 and 4.17 I used to fit the curve as close as I could to the ADOM data. I wanted to maintain that level 15 gave +12 to-hit with both the linear and logarithmic. These constants can be modified to allow for to-hit, to-damage and DV, and race,class,starsign,weapon type, etc..

Going through the manual the current values seem to be almost linear - although some of them tend to "curve up" that is increase at an increasing rate, which could be a problem if you want to stick closely to these values. If you really want to stick to these closely, a cubic equation is probably necessary at levels 0..15 such that the to-hit at 0 is 0 with gradient 0, and the to-hit at 15 equals the given to-hit value, and "tapers off" slightly at the end. I.e. match the gradient of the cubic function to the gradient of the log function at that point.

Dougy
03-23-2008, 01:30 AM
I tested the idea of a cubic function between 0 and 15 followed by a logarithmic function thereon. I'm not entirely thrilled by it, but I thought I would share the experiment.

Let c be the value you want at level 15. Let x=ln(15). Let k be a parameter - I introduced to vary the slope of the log function.

The cubic function is f(level) = (3cx-c+k)/225x level^2 - (2cx-c+k)/3375x level^3 for 0<=level<=15
The logarithmic function is g(level) = k + (c-k)/x * ln(level) for level>=15.

f(0)=0
f'(0)=0
f(15)=g(15)=c
f'(15)=g'(15)=(c-k)/15x

I've attached a copy of this function when c=12 and k=-20. The main drawback with this type of graph is that it requires time to "get going."

Aerudaer
03-23-2008, 03:11 AM
If the number of points to advance a level stays the same as in ADOM then that looks like a great curve. The ever increasing points to get a level balance out the awards so that it becomes less and less beneficial to slay another monster and keeps the values reasonable. At the same time, it would provide a good incentive to train because the benefits could be substantial.

Sradac
03-23-2008, 03:23 AM
i think if anything it should be harder at the end point adn not to "get going" think about it, if you train with a weapon once you get basic skill in it increasing your skill wont be that hard at first as there is alot you dont know about the weapon. But once you get better it'll become harder to "learn new tricks" as you already know much about how to use that type of weapon. the spell casting in adom did it right, each skill level you had to use the spell more times to, get just a little bit better.

Kalmakka
03-23-2008, 03:25 AM
the skills cross-train? I mean, using a 2H weapon should share around 60% of the exp with the 1H one..

I remember in Morrowind, you could be an expert at wielding a long blade, but if for some reason you had to use a short blade you wouldn't be able to hit a thing. That was awful. :p

Ideally, you could have a matrix determining how much using a weapon trains your other skills. My reasoning is that using any weapon will give you a bit of fighting experience that can be put to use to wield any other weapon.
For instance, using a 2h polearm would give much training in 2h polearm (of course), somewhat less in 1h polearm (perhaps 50% of what you got in 2h polearm), a bit in staves (15%), small amounts in swords and blunt weapon skills (5%), and not a significant amount in the rest (1%)

Dougy
03-23-2008, 03:46 AM
I remember in Morrowind, you could be an expert at wielding a long blade, but if for some reason you had to use a short blade you wouldn't be able to hit a thing. That was awful. :p

Ideally, you could have a matrix determining how much using a weapon trains your other skills. My reasoning is that using any weapon will give you a bit of fighting experience that can be put to use to wield any other weapon.
For instance, using a 2h polearm would give much training in 2h polearm (of course), somewhat less in 1h polearm (perhaps 50% of what you got in 2h polearm), a bit in staves (15%), small amounts in swords and blunt weapon skills (5%), and not a significant amount in the rest (1%)

One problem with this, I foresee, is that usually you get 1 weapon skill point for killing something. So every number would thus need to be scaled up by at least a factor of 100 so something could be increased only by 1%.

Also, I wouldn't want to see someone even theoretically capable of achieving mastery of swordfighting entirely from throwing rocks (or some other silly combination), possibly never even seeing a sword in his/her life.

The matrix idea sounds reasonable, although it introduces some arbitrary proportions. I believe some tricky questions would arise: (a) Would the matrix be symmetric? For example, would using a 1H-sword train your 2H-sword skill just as much as using a 2H-sword would train your 1H-sword skill? (b) Would using a 2H-sword train your 1H-sword skill as much as using a 1H-sword would train your dagger skill?

Andur
03-23-2008, 05:51 AM
Glad to see the ever so mysterious 2-Handed Weapons category from Adom removed, it was never really clear what fell into that category and what didn't.

Splitting the skills into 1-handed and 2-handed is perfectly reasonable, as those are very, very different fighting techniques. Any cross-training between the 2 should be slow and perhaps have a hard-cap limit.

In addition to just hitting things with them to train, some weapon-master/trainer types in the game would be handy. I've never managed to reach grand mastery with a weapon in Adom, even in my ultra-ending games.

The Adom stat bonuses always seemed reasonable enough to me. It was neat that the different categories gave differing bonuses. Adding in critical hit modifiers sounds logical enough. Anything beyond that would be cool too.

Some kind of tie-in with the Tactics system and weapon skills could be neat (different stances).

The shield skill in Adom always seemed a little weird that it was based on the DV of the shield you are using. It seems to me, that the size of the shield, not the size+material (as is the case in Adom), should be the determining factor in how effective you can use it.

Orion
03-23-2008, 11:09 AM
The matrix idea sounds reasonable, ...

I agree with that. If you train one-handed swords, it should rise your two-handed swords skill either (and maybe other skills). It is not realistic, that you know how to use one-handed swords, but you do not know anything about two-handed swords - the fighting technique may be different (for one- or two-handers) but sword is sword...

The second point is, that if you are a fighter and you train one type of weapon, it should teach you something general about weapons... It is not realistic, that you are a master of using swords, but you don't know anything about daggers...

Grey
03-23-2008, 01:18 PM
I don't actually like the separation of two-handed types for each weapon, I must say. What's so different about how to use an axe one-handed or two-handed? It's an extremely similar style of combat - far more similar than say one-handed and dual-wielding two axes. It'd rather see a "two-handed weapon" skill in ADOM, like there is for dual-wielding. Thus anyone trained in swords would have a little trouble handling a big two-hander unless they're trained in the skill. One nice result of this would be that spellcasters would have a hideous time trying to use big weapons since they wouldn't start with that skill (whilst warrior types and trolls would).

To solve some of the problems I perceive in ADOM I think weapons types should give more dramatically different benefits so each one feels different. For instance:
-Daggers have very low DV, low damage, but more crits and bigger energy point reduction
-Swords give more damage, DV and to hit, but less energy point reduction
-Spears give loads of DV, more crits, but less to hit/damage and way less energy point reduction
-Crossbows give more to hit and criticals, but virtually no energy point and damage bonuses
etc.

The individual weapon effects like bleeding/stunning would be nice. Some other ideas might be armour damage, some natural armour penetration (say at level 10 axes always subtract 4 PV from the enemy), disarming, multiple attacks, paralysing (small chance when backstabbing say).

More racial effects might be nice to make the races feel different from each other. In ADOM gnomes get a bonus to crossbows. Other races could have similar bonuses - elves with bows, humans with swords, trolls with clubs, dwarves with axes, etc.

Shield skill bonuses also need to be dramatically reduced in my opinion. And what about armour skills? Some other games use this idea effectively, with separation between armour types. For ADOM we could have the following armour types:
-Unarmoured - naked, clothes, robes, etc
-Light - leather, studded leather, elven chain, furs, wrappings, shell
-Heavy - ring, scale, chain, splint, mail
(This could be separated out further with splint and mail as heavy whilst the others are medium, but overcomplication may be a bad thing.)
Armour skills would affect DV, PV and spell evasion, whilst also reducing negative modifiers to dex/speed/to hit from heavy armours.

Two Weapon Combat could be changed from a regular skill to a weapon skill, with the same sort of upgrade levels. This would mean characters wouldn't be able to just shove points in it to get better - they'd actually have to use it regularly. It would be running alongside the regular weapon skills, so you could technically be getting points in swords, axes and 2WC all in one round. It would give extra to hit bonuses (countering the negatives of dual-wielding) as well as bigger DV bonuses and energy point reductions.

Nezur
03-23-2008, 11:03 PM
-- What's so different about how to use an axe one-handed or two-handed? It's an extremely similar style of combat --

Yes, I agree to a certain extent. Hmm... If you wielded one-handed weapon and no shield, wouldn't that mean you automatically used your both hands to swing the weapon? At least that would be the case for medium-sized and large arms.


-- Two Weapon Combat could be changed from a regular skill to a weapon skill, with the same sort of upgrade levels. This would mean characters wouldn't be able to just shove points in it to get better - they'd actually have to use it regularly. --

Shove points? The JADE page reads:


Experience levels also will play just a minor role. They will give you a very minor hitpoint bonus (+1 hitpoint per experience level) and only serve as a general guideline for your accomplishments... it has no other effects.

In fact I don't if this has changed since it was announced 12/24/1998... Well, anyway. All your skills will apparently only increase if you use them. So in that sense regular skills would be similar to weapon skills. Correct me if I'm wrong, however.

I have to chew on this matter a bit more... Right now I'm too tired to devise anything. :)

bjarketp
03-24-2008, 12:33 PM
Mr. Biskup should be very careful with buffing the weapon skills too hard. First of all, it encourages using only one weapon for the entire game, which is kinda boring. Secondly, without no max cap on weapon skill one might be able to infinitely train his weapon skill on, say, breeders or a grimlin bomb or whatever. Of course each rank should take exponentially longer to achieve, I think that a cap on weapon skill is needed.

The cap in adom was big enough that you could easily complete the game with a melee character, yet not achieve grand mastery in your favourite weapon.

Qui
03-24-2008, 01:07 PM
About cross-training

The matrix idea is a good one, but I propose to handle it differently. Weapon skill points should be added only for the weapon type you're currently using, thus no need to calculate any decimals or anything. A level reached in one weapon would make reaching that level in similar class easier.

Example:
L5->L6 advancement in swords needs 1000 points. But the PC already has L6 in 2-handed swords, correlation level is let's say 40&#37;, so advancing to L6 in swords costs only 600 points.

Bonuses from each correlated classes should sum. It makes sense that if the PC is skilled in two similar skills he should learn even faster.

Example:
L5->L6 advancement in blunt weapons needs 1000 points. The PC already has L6 in axes (30% correlation) and in ball & chain (15% corr.), so L6 in blunt weapons cost is reduced by 45% and is 550 points.

There should be a hard limit on percentage of reduction. Let's say 20%. It allows to enter high correlation in the matrix without silly side effects.

Example:
L5->L6 advancement in 2H polearms needs 1000 points. The PC already has L6 in 1H polearms (70% corr.) and in staves (60% corr.). In total it's 130%, but due to hard limit the cost can't be lower than 20%, so L6 in staves costs 200 points.

I think it is a simple to implement solution which deals with both "he has level 10 in swords but never used one" and "he has level 10 in 2H polearms but stabs himself in the eye when given a 1H polearm" problems.

Tannis
03-24-2008, 02:20 PM
Mr. Biskup should be very careful with buffing the weapon skills too hard. First of all, it encourages using only one weapon for the entire game, which is kinda boring. Secondly, without no max cap on weapon skill one might be able to infinitely train his weapon skill on, say, breeders or a grimlin bomb or whatever. Of course each rank should take exponentially longer to achieve, I think that a cap on weapon skill is needed.

The cap in adom was big enough that you could easily complete the game with a melee character, yet not achieve grand mastery in your favourite weapon.

I disagree and there's an easy solution for this. Unlimited weapon skills won't be game breaking provided that the advantages don't scale linearly; I think the logarithmic approach to weapon skills is a great solution, and while the benefits to things like DV and damage might taper, the advantages of high weapon skills could lie in the already mentioned special abilities that one can unlock.

The problem is, as you mentioned, someone could grind gremlins and get to level 25 in swords, and that seems unfair. I think its imperative that JADE avoid the ADOM solution, where monsters grow harder the more you kill them. This is counterintuitive and leads to the terrible uber-jackal effect.

The solution is simple: split monsters into three (or more classes), such as mundane, skilled, and legendary. The names don't matter. A mundane creature would be a rat, goblin, orc, or giant bat. It's common and not terribly threatening. There's only so much you can learn from them, so they should not get you farther than level 7 or 8 in a given weapon class. A skilled monster could be a gnoll warlord, ogre chieftain, or vampire knight. Mean monsters that pack a punch, and can train skills up to say level 15 or so. After that, increased weapon marks will only be obtainable from legendary creatures: dragons, gorgons, minotaur mages (ugh), gigantic hydras, etc. This way, getting your weapon level past level 20 will be a tremendous feat, since these monsters are rare and deadly. This way, the possibility for increasing your weapon skill exists, but it will be a challenge.

Another advantage of this system is you don't need a ridiculous scale for weapon marks. For example, level 15 could require 100K marks, and level 16 could require 120K; The scale isn't exponentially, but since marks can only come from legendary creatures it will still take ages without having to kill thousands of puny creatures like bats. Thoughts?

Orion
03-24-2008, 04:16 PM
Both, Qui and Tannis, have great ideas - I like them.
Only thing is, that I'm not sure which scaling is the best - linear or logarithmic. Probably the best one is something between these two...

Dougy
03-25-2008, 12:44 AM
Both, Qui and Tannis, have great ideas - I like them.
Only thing is, that I'm not sure which scaling is the best - linear or logarithmic. Probably the best one is something between these two...

Well I initially proposed linear before level 15 and logarithmic after, but a cubic before level 15 seems to fit the ADOM values better.

The other option is the make the points needed to increase in level exponential, although I'm not sure how TB is storing these values and this might be impossible due to the "wrap around" bug.

Personally I prefer the logarithmic scale at the end. When you're a beginner you can pick things up quickly and you learn many new things, but when you're experienced it takes a lot of work for little gain.

Dougy
03-25-2008, 01:23 AM
As for critical hits:

The probability of a critical hit should be given by p(level)=b-(b-a)*e^(-k level^2) where
a and b are determined by the weapon and a few PC traits
a is the probability of a critical hit of a completely unskilled PC
b is the probability of a critical hit of a PC with weapon skill infinity, that is the theoretical maximum
k is a constant that varies the slope of the graph, that is how "quickly" it goes from bad to good. I recommend keeping this at 1/15^2 to compare with ADOM.

I plotted a graph of this function with a=0.05, that is without training the PC scores a critical hit 5% of the time, and b=0.5, that is a fully trained PC would score a critical hit 50% of the time. Constant k=1/15^2. I've attached a copy of the graph.

Tannis
03-25-2008, 01:43 AM
As for critical hits:

The probability of a critical hit should be given by p(level)=b-(b-a)*e^(-k level^2) where
a and b are determined by the weapon and a few PC traits
a is the probability of a critical hit of a completely unskilled PC
b is the probability of a critical hit of a PC with weapon skill infinity, that is the theoretical maximum
k is a constant that varies the slope of the graph, that is how "quickly" it goes from bad to good. I recommend keeping this at 1/15^2 to compare with ADOM.

I plotted a graph of this function with a=0.05, that is without training the PC scores a critical hit 5% of the time, and b=0.5, that is a fully trained PC would score a critical hit 50% of the time. Constant k=1/15^2. I've attached a copy of the graph.

Doesn't every other hit being a critical one strike you as rather high? I feel like 25-33% should be the max for a well trained character, otherwise it might become too unbalanced. Great graph, though.

Dougy
03-25-2008, 02:12 AM
Doesn't every other hit being a critical one strike you as rather high? I feel like 25-33% should be the max for a well trained character, otherwise it might become too unbalanced. Great graph, though.

Well, the constants a and b can be changed to suit without trouble.

Remember that b is a theoretical maximum, nobody can ever achieve it. This isn't a "well trained" character but one that knows absolutely everything there is to know about fighting with that weapon.

At level 15 there's 33% probability, at level 20 there's 42% probability. I'm expecting that JADE will be similar to ADOM in the sense that you would have to play for a _long_ time before achieving level 15 in weapon skills and levels beyond that would only come from scumming.

Dorten
03-25-2008, 04:27 AM
When speaking about logarithmic increase of bonuses with every level of mastery, and exponential increase of their costs, don't you act a bit harsh? With exponential increase in cost, and linear increase of bonuses with levels we already will have logarithmic increase of bonuses counted against weapon marks used. That's because if you have to spent month of real time playing to get 20 level, then you'll have to play for, say, another half of a month to get to 21'th level...
The idea with different classes of enemies is more realistic, however, and, in my opinion, would fit better in game (and it'll enable the possibility not to use exponent as cost function)

moppit
03-25-2008, 04:42 AM
Do you have prerequisites set up for skills yet (and/or will you be doing prereq skills - akin to talents in adom)

Assuming you had or wanted to use a prerequisite skill system, you could do something along the lines of

if Weapon skill:blunts(1h) or Weapon skill:blunts(2h) >= 10 Then allow stunning at (1% + 2d4%)/10 or something

and advance it as a skill or something.. Could even make up fun names for it like Thundering Hammer Strike or Whirling Dervish Technique etc... Pardon the generic unadomesque names..

Then you could perhaps just add a few lines to the weaponawareness class that would case if any techniques were applicable and variable in the modifier% chance for desired/trained effect.

Dougy
03-25-2008, 05:16 AM
When speaking about logarithmic increase of bonuses with every level of mastery, and exponential increase of their costs, don't you act a bit harsh? With exponential increase in cost, and linear increase of bonuses with levels we already will have logarithmic increase of bonuses counted against weapon marks used. That's because if you have to spent month of real time playing to get 20 level, then you'll have to play for, say, another half of a month to get to 21'th level...
The idea with different classes of enemies is more realistic, however, and, in my opinion, would fit better in game (and it'll enable the possibility not to use exponent as cost function)

In ADOM, I believe the required marks are multiplied by the golden ratio each time, that is 1.618.... So yes, that is exponential. See this (http://mathworld.wolfram.com/GoldenRatio.html) for info about the golden ratio. These values are listed in the manual:


Level 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Melee 15 25 40 65 105 170 275 445 720 1165 1885 3050 4935 7985 12920
Missl 7 12 20 32 52 85 137 222 360 582 942 1525 2467 3992 6460

Harsh? In ADOM, you can reach level 15 and that is it. No matter how much you train your char, there will be no increase in your weapon skills. Isn't that a bit harsh? The player knows (or should know) that this is the case in ADOM and similarly in JADE the player will know that there are decreasing returns if he or she continues to scum.

Already in ADOM different enemies give you a different number of weapon skill points per attack.

Dorten
03-25-2008, 05:37 AM
the player will know that there are decreasing returns if he or she continues to scum.
They will be decreasing even if the bonus increase with levels will be linerar... hell, even polynomial!
That's what I meant to say. So, I'd say - the easiest would be to remove the cap, keep marks increase exponential as it was in ADOM, and bonus increase linear... DON'T make it logarythmic.

That's just me, though

Dougy
03-25-2008, 07:41 AM
They will be decreasing even if the bonus increase with levels will be linerar... hell, even polynomial!
That's what I meant to say. So, I'd say - the easiest would be to remove the cap, keep marks increase exponential as it was in ADOM, and bonus increase linear... DON'T make it logarythmic.

That's just me, though

I do see your point, you don't want this "double-whammie" effect. First it's difficult to get to the next level and even when you do, you don't get much and less and less as time goes on.

The reality of the situation is, that nobody is going to be getting over level 30 or so without some form of cheating or some extreme level of scumming. According to the ADOM system, to get to level 30, you'll need a total of 44,791,032 weapon skill points.

For low levels (the ones feasible to reach), the logarithmic and linear systems are very similar. I'll attach a picture of the two graphs along with the points I listed earlier for to-hit bonuses, all with +12 to hit at level 15.

In the ADOM Hall of Fame (http://adom.brinkster.net/hof/) there are flags with chars using the wish spell to get every single weapon skill maxed, i.e. level 15. Without the level cap in JADE the char can continue to wish and wish without limit. We should be careful not to allow this form of scumming to be extremely advantageous.

Orbic
03-25-2008, 09:16 AM
I don't actually like the separation of two-handed types for each weapon, I must say. What's so different about how to use an axe one-handed or two-handed? It's an extremely similar style of combat - far more similar than say one-handed and dual-wielding two axes. It'd rather see a "two-handed weapon" skill in ADOM, like there is for dual-wielding. Thus anyone trained in swords would have a little trouble handling a big two-hander unless they're trained in the skill. One nice result of this would be that spellcasters would have a hideous time trying to use big weapons since they wouldn't start with that skill (whilst warrior types and trolls would).


From experience, there is nothing in common between 1H and 2H weapon fighting.
1H weapon is used in combination with another weapon or a shield for defense. If a 1H light weapon is used by itslef then you rely on dodgeing for defense, or on the weapon itself (late middle ages swords, rapiers - this shows the evolution of weaponry and fighting techniques where the armor is no longer on the body but on the weapon - hilt in this case).
Dual wielding (usually sword + dagger) is on the fencing evolution path after the shield / armour desuetude and good fun, eventually to be replaced with 1H piercing only modern sword - much faster.

2H weapon is used in combination with a heavy armour, the aim being destroying the opponent's armour to hinder movement or cause wounds. So it is in fact much more different from 1H than dual wielding is.

Tannis
03-25-2008, 03:01 PM
I do see your point, you don't want this "double-whammie" effect. First it's difficult to get to the next level and even when you do, you don't get much and less and less as time goes on.

The reality of the situation is, that nobody is going to be getting over level 30 or so without some form of cheating or some extreme level of scumming. According to the ADOM system, to get to level 30, you'll need a total of 44,791,032 weapon skill points.

For low levels (the ones feasible to reach), the logarithmic and linear systems are very similar. I'll attach a picture of the two graphs along with the points I listed earlier for to-hit bonuses, all with +12 to hit at level 15.

In the ADOM Hall of Fame (http://adom.brinkster.net/hof/) there are flags with chars using the wish spell to get every single weapon skill maxed, i.e. level 15. Without the level cap in JADE the char can continue to wish and wish without limit. We should be careful not to allow this form of scumming to be extremely advantageous.

This is an issue for another thread I guess, but I firmly believe wishes should be abolished. At the very least, unscummable: No possibility of wishes from pickpocketing or potions of exchange.

Silfir
03-27-2008, 01:27 AM
What I'd love to see is a system where training in a certain weapon skill would also, to a certain extent, train other weapon skills. I mean, does it really make sense that your training in clubs and hammers has no effect whatsoever on your skill fighting with maces? At least some of the things you learn have got to be applicable for both weapon types!

Dougy
03-27-2008, 04:53 AM
Based on the current list of weapon skill categories, I made a first attempt at making a matrix to allow for cross-training. Over time these values can be adjusted to allow for peoples' preferences. The values are percentages, probably most easily implemented by a probability of increasing the skill by that percentage.

For example: Fighting with 1H-swords trains daggers and knives by a factor of 20%.


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 sum
unarmed 1 100 2 10 8 10 8 3 2 3 2 8 10 10 10 2 5 2 7 202
daggers & knives 2 2 100 1 1 1 1 20 15 10 5 5 2 1 3 2 1 2 7 179
blunt (1H) 3 10 1 100 25 10 2 2 1 2 1 3 2 5 2 1 3 1 5 176
blunt (2H) 4 8 1 25 100 5 10 1 2 1 2 1 2 10 2 1 2 1 2 176
ball & chain (1H) 5 10 1 10 5 100 25 2 1 3 2 3 1 2 5 1 1 1 2 175
ball & chain (2H) 6 8 1 2 10 25 100 1 3 1 3 1 2 5 5 1 1 1 2 172
swords (1H) 7 3 20 2 1 2 1 100 25 10 5 3 1 1 3 1 1 1 2 182
swords (2H) 8 2 15 1 2 1 3 25 100 5 10 2 1 3 3 1 1 1 2 178
axes (1H) 9 3 10 2 1 3 1 10 5 100 25 2 1 3 3 2 1 2 2 176
axes (2H) 10 2 5 1 2 2 3 5 10 25 100 2 3 3 3 2 1 2 2 173
polearms (1H) 11 8 5 3 1 3 1 3 2 2 2 100 25 4 2 3 1 3 5 173
polearms (2H) 12 10 2 2 2 1 2 1 1 1 3 25 100 15 2 2 1 2 2 174
staves (2H) 13 10 1 5 10 2 5 1 3 3 3 4 15 100 2 2 1 2 2 171
whips 14 10 3 2 2 5 5 3 3 3 3 2 2 2 100 2 2 2 2 153
bows 15 2 2 1 1 1 1 1 1 2 2 3 2 2 2 100 15 25 15 178
slings 16 5 1 3 2 1 1 1 1 1 1 1 1 1 2 15 100 13 25 175
crossbows 17 2 2 1 1 1 1 1 1 2 2 3 2 2 2 25 13 100 10 171
thrown weapons 18 7 7 5 2 2 2 2 2 2 2 5 2 2 2 15 25 10 100 194

- The matrix is symmetric, so fighting unarmed trains 1H-axes as much as fighting with 1H-axes trains unarmed. This can be changed if you feel it's unrealistic.
- I attempted to make the sum roughly constant, except for unarmed fighting, whips and thrown weapons, which are special in my opinion.
- I recommend that a secondary weapon skill is trained with the probability given in the matrix only if the skill level is within -30% to 0% of the primary weapon skill used. In this way, the PC will have to maintain his skill in 1H-axes in order to achieve mastery, even if he uses 2H axes all the time. This would require that the total skill points in every category be recorded.
- It shouldn't be too difficult to edit these values to allow for some randomness in each PC, the PC's learning value, skills, luck, etc..
- Everything has a possibility of training everything else. I don't see this as a problem of realism because the PC could learn the whereabouts of a weak spot on a being's body and how to exploit it.