[Crossfire-wiki] [Crossfire DokuWiki] page changed: user:mwedel:skills

[no-reply_wiki at metalforge.org]

A page in your DokuWiki was added or changed. Here are the details:

Date        : 2011/01/31 00:49
User        : mwedel
Edit Summary: add notes about skills

@@ -38,9 +38,9 @@
  **Wizard Spellcasting** Just as it is now, but maybe skills get redone or merged some, but it basically represents the pyromancy, evocation, etc, skills.
  
  **Cleric Spellcasting** Basically they praying skill as it is now.
  
- == Advancement ==
+ === Advancement ===
  
  Whenever a characters overall level goes up, they gain 10 AP (Note that the value 10 here is arbitrary, and in fact, would probably be a settings file variable - some servers could have a higher value to make things easier, and as it stands now, 10 may just be too low in any case).
  
  At any point after gaining that level, the player chooses how to spend those skills.  Ideally, there is some nice interface for this.  Unspent points can be carried over.
@@ -50,15 +50,15 @@
  The skill levels range from 1 to 100.  Once a skill level reaches 100, it can not be advanced further, but the character likely has other skills they may want to advance.  A cap on skill levels is necessary because as designers, it must be known the parameters one works within.  For example, as damage goes up as weapon skill increases, a character that is level 1000 in a weapon skill may just do too much damage.  If effects of the skill are spread out over the maximum level, one may find that there are large gaps with no bonuses.  For example, if it is decided that at maximum weapon level, the character gets +20 damage, that means that if maximum level is 100, every 5 levels their damage increases - often enough to be a bonus the character notices.  But if maximum level is 1000, then every 50 levels the characters damage increases by 1, which would be pretty insignificant, especially if that level 1000 maximum was put in place just to satisfy a few players who work to max out characters.
  
  However, I do not see any reason why a characters overall level can not be unlimited.  Any form of bonuses (hit points, item power usage, etc) may also cap at level 100, but the character could continue to gain levels if they so desired to increase various skills.  Given the costs and number of skills, it may be a character would have to be very high level before they would ever max out all their skills.
  
- == New Characters ==
+ === New Characters ===
  
  I have not extensively thought about new character creation.  It may be new characters get 20 or 30 AP points to populate some skills (exact value should be a setting).  So the fighter could decide to increase the level of their weapon skills, etc.  Note that a level 1 character would still be constrained by the 2*level cap for skills.
  
  It is conceivable that custom class builder could be designed that lets the character choose a couple skills at cost 3, a couple at cost 4, etc.  Balancing that properly could be difficult - the player could properly decide that they will only use swords, so that is the cost 3 weapon skill, will only use pyromancy as second cost 3 skill, and take armor and praying as cost 4 skills - pretty much getting all benefits with no drawbacks.
  
- == Why is this better? ==
+ === Why is this better? ===
  
  These opinions are purely my own, but I do think it has some advantages:
  
    * Classes now have more significance - choosing to be a fighter vs wizard has actual long term implications.  However, this still does not prevent any class from learning other class skills.  Note also that how much significance the classes has really depends on cost.  If a server admin decides to, they could modify the archetypes and make all skills for all classes just cost 5, so there is no advantage/disadvantage.  Tweaking the costs also would determine difference - instead of the 3/7 mix I have above, a 4/6 mix still has some advantage/disadvantage, but not nearly so much.
@@ -66,9 +66,45 @@
    * This //fixes// problem of experience gains for skills.  Some skills right now do not gain experience easily - while this should be fixed, some skills just do not lend themselves all that easily for a balanced experience advance, either because they are skills that are infrequently used and would need to then give a very big (and seemingly out of proportion) experience reward when used, or the fact they may be considered too easy to use, and one does not want to give easy experience away.  Using skills would still gain experience, but it would got to the characters overall experience total.
  
    * This adds more choices during advancement and can give some goals for players.  This is purely personal opinion, but I prefer having choices to make about character advancement - I don't want to do it every minutes, but when gaining a level is fine.  And if more effects are put in for skills (at level 20, I get a chance to stun opponents with the hammer), it gives more incentive for characters to get to those points.  Within current skill system, there are things you gain, but in most cases, a character is not going to really do much about them - if a character is killing things with their pyromancy skill, it is unlikely they are going to switch to using another skill because of a bonus they may get 4 levels later.  I will note that the choice here is little like dragon characters get to make choices on focus and slowly increase resistances, which may be one reason that is a more popular character.
  
- == Loose Ends ==
+ === Specific Skill Ideas ===
+ 
+ These ideas are actually independent of the revised advancement system, but since this page is **skills**, seems like a good place to put them.
+ 
+ **Item Creation/Identification Skills** These are things like jewelery, smithery, bowyer, etc.  The item creation aspect should get expanded so there are more easy things to create give minor amounts of experience (for example, smithing a dagger should not be a hard task).  Identification of items should change - instead of there being a chance, it a straight level comparison - to identify a level 5 item, you need to be level 5 or greater in the appropriate skill.  Non magical items would be level 1, each plus of an item might add one level, each 20% resistance adds one level, etc.  Some formula could be determined for objects that do not have an item power setting.  There are a few advantages of this:
+ 
+   * The server no longer needs to track if an identification attempt was made on the object or not - either the character succeeded or not, and retrying will not work better since there is no random chance.
+   * Related to the above, a character can retry and perhaps succeed if they have increased the level in the skill.
+   * Items can be traded in hopes that the other character may be able to identify them.
+   * Right now, with current formula, even at low levels, characters can identify the vast majority of items.  This would make that harder.
+   * Greater level difficult for items could give much greater rewards - completely reasonable that if you are able to identify a level 20 item, you get lots of experience - there are not going to be many such items about.
+ 
+ In many ways, this would just have it work the same as literacy and reading books - either you are sufficient level to read it, or you are not - there is no chance involved.
+ 
+ **Weapon Skills**
+ 
+ One thought is to break down the weapon skills some - instead of the basic breakdown of one handed, 2 handed, and missile weapon (ignore the various hand to hand skills), do it more functionally - axes, clubs/maces/hammers, swords, spears, bows/crossbows.  From the basic table above, it is likely that different classes may have a different mix on cost here.
+ 
+ For example, fighters may be able to get all of those at cost 3.  But clerics, following AD&D tradition, might be clubs/maces/hammers at cost 4 (they are still not as good as fighters), and the other weapon skills at cost 5, and perhaps some at cost 6.  Paladins might get swordsmanship at cost 3, but higher cost for all the other skills - perhaps bow being cost 6, since paladins should really meet their foes in face to face combat.
+ 
+ This revised system overall is based on AC/WC going away - that is too difficult to balance - otherwise, at some point, that level 20 (in weapon skill, but maybe level 40 overall) will just never hit something that the level 40 fighter (level 50 in weapon) is hitting.  Now that level 20 character should be less effective in weapon, but that would be controlled by lower damage (30 level difference would be 6 damage) and weapon speed.  Thus, it takes longer for that first character to kill the same creature as the fighter, but unless that creature has good regeneration, is still quite possible.
+ 
+ In order to balance the system, one must think about what the advancement of the weapon skill is.  Off the top of my head, this is what I've thought of - at first level, a character does damage around 10 @ speed 0.75.  Weapon speed will peak at 2.0 - exactly when that happens could vary on a lot of factors - if the character is not wearing armor, they may get that speed much sooner, but there is clearly a disadvantage there.  Weapon damage should peak around 25 for the weapon itself (note, there are many above that) - at first level, weapons the character is likely to find should be less than 10.  A characters skill, at level 100, might give another 25 damage points, and strength 25 more.  So that is a damage value of 75 @ 2.0 speed.  That is a very big swing, but one also has to figure that at first level, the characters they are fighting have 0 armor, and at level 100, everything should have 75% armor.  So that first level character 3.75 damage/tick on the actual creature, at level 100, the character would average 18.75 DPT to the actual creature - 5 times as much - but that opponent is also going to have a lot more hitpoints (probably closer to 10 times as many), so would take longer.
+ 
+ Unfortunately, doing the math, a 1000 hp creature would be brought down in 53 ticks, or a bit under 7 creatures.  IMO, such a creature should be epic (not a huge number of level 100 creatures about.  Increasing armor or hit points would increase the time, but one should not have super well defended creatures.
+ 
+ However, a lot of this is also based on character hit points - a high level character could have ~750 hit points if they really tried.  I'd personally like to even out the con bonus so it is for all levels, not just the first 10 (a level 10 character with 30 con could conceivably have 500+ hp - how can one balance that when a more typical character is probably at ~120?).
+ 
+  So if hit points is changed so that each level (all the way up to 100), they get a die roll of 1-10.  So with lots of improvement potions, character now has 1000 hp from that.  With maximum con, give them another 500 (5/level), for 1500 total.  If one says it is reasonable for really tough (eg, epic boss creatures) to double character hit points, that give that creature 3000 hit points - now that is 160 ticks, or about 20 seconds, assuming creature has no regeneration, that character can constantly keep on the creature - not bad.
+ 
+  One effect of evening out the hit points for characters is that it improves balance.  The level 10 creature is likely to always have 130 hp or so (characters start with 30) - a high con will give them a few more vs low con, but not nearly so drastic as it is now.
+ 
+ This is starting to get off topic from skills - main point was to try to determine a maximum damage rate for that maximum character with melee weapons - from there, one can then start to figure out what reasonable value for other spells are - for example, with an 18.75 DPT for melee weapon, one can also figure that a spell should do 18.75 DPT also at maximum level.
+ 
+ A note on balance here (again) - if we assume the character has 1500 hit points, and that the DPT for unarmored is 75 (you go into an arena, or other person just attacks you), that is still 20 ticks (2.5 seconds) before you are dead - not a lot of time, but perhaps enough to do something.
+ 
+ === Loose Ends ===
  
  The examples here really just focused on the 3 main character archetypes (fighter, cleric, mage).  Crossfire has many more than that, and I think this also helps them out.
  
  For example, with such a system, thieves may be better able to improve their thief like skills - they don't care how they get their experience, so if they are able to sneak around and get a surprise attack for extra damage bonus on a tough monster, it doesn't matter where the exp for that kill goes.
@@ -76,12 +112,12 @@
  Dragon characters probably get a different skill mix - for example, clawing and dragon armor may become skills they need to improve instead of armor being tied to overall level.
  
  These changes do not directly impact balance, except so far as the characters skill level relative to overall level may differ.  But if the skills are properly defined for effects, eg, that at level 100 weapon skill, you get +20 damage bonus, one can work out appropriate power for other skills.  For example, one could assume that a person with 100 weapon skill would need to be at least level 50, and so may have a weapon that does damage 20, have another 10 damage from strength, and attack speed could be defined as at best being 2.0.  So that means that the character has damage 50 at speed 2.0, so would average 50 damage a tick (since damage is rolled as a random number from 1 to damage, and they get 2 attacks).  So one could then figure a spell of similar level should also do 50 damage/tick, but one might lower that average for large area of effect spells.
  
- == Wrapping Up ==
+ === Wrapping Up ===
  
  This page was mostly just to record some thoughts that have been bouncing around my head for a while.  I don't know if/when it would ever get implemented.
  
  Mark Wedel
  January 2011
  
  


IP-Address  : 173.228.106.112
Old Revision: http://wiki.metalforge.net/doku.php/user:mwedel:skills?rev=1296282251
New Revision: http://wiki.metalforge.net/doku.php/user:mwedel:skills

-- 
This mail was generated by DokuWiki at
http://wiki.metalforge.net/