1. Updated Resource Submission Rules: All model & skin resource submissions must now include an in-game screenshot. This is to help speed up the moderation process and to show how the model and/or texture looks like from the in-game camera.
    Dismiss Notice
  2. DID YOU KNOW - That you can unlock new rank icons by posting on the forums or winning contests? Click here to customize your rank or read our User Rank Policy to see a list of ranks that you can unlock. Have you won a contest and still havn't received your rank award? Then please contact the administration.
    Dismiss Notice
  3. We have recently started the 16th edition of the Mini Mapping Contest. The theme is mini RPG. Do check it out and have fun.
    Dismiss Notice
  4. Dismiss Notice
  5. The Highway to Hell has been laid open. Come along and participate in the 5th Special Effect Contest.
    Dismiss Notice
  6. Check out the Staff job openings thread.
    Dismiss Notice
Dismiss Notice
60,000 passwords have been reset on July 8, 2019. If you cannot login, read this.

New skill system is boring at low-mid levels. Solution - flat milestone bonuses!

Discussion in 'Gaias Retaliation ORPG' started by ALEXz407, Jan 1, 2015.

  1. ALEXz407

    ALEXz407

    Joined:
    Aug 23, 2014
    Messages:
    206
    Resources:
    2
    Tutorials:
    2
    Resources:
    2
    Release of a new version of the map was the real New Year's gift from Zwiebelchen.
    Many players with high leveled chars immediately began testing new skill combinations and write feedbacks. However...

    The Problem
    Players compare high level changes. Many of them already have top items. In this way difference between for example "Might 20 lvl" and "Might 25 lvl" is very noticeable. But what about low-mid levels where player have low stats? Let's look at a graph:
    PercentDmgIncrease.png
    You can see that the player with 15 base damage (roughly d1 items) will receive +1 damage from Might only with 7 points invested in this skill, or talking about levels - with 13 level character minimum. I want to ask - is it fun?
    Spirit also gives almost nothing on low levels. There are only a few enemies, requiring resistances. And you get an extra stat point only with d2 level items and at least 10 points in this skill, because numbers rounded down.
    If you want be a good tank or improve your survivabilty as a caster you can choose Constitution as a first skill. Well, get 1 HP as a squire, or 0 HP as Magician (because mages must suffer! :). Also it grants 1% evasion every 2 skill levels - nice addition, but it is almost nothing with low HP pool.
    Perception gives linear increase in crit chance and crit damage. Equally useful at 1 level and 50. Also it boosts both damage/healing output. Skill for stacking number one.
    Reflexes is the most profitable skill for stacking. It grants linear stat growth and at high values effectiveness will grow exponentially. Skill for stacking number two.
    As we can see the new skill system repeats mistakes of the old skill system: it is best to stack 2 skills. In addition really useful at low-mid levels only two of the five skills. The situation changes at high level with green-blue items, but still build 25/25 is best.

    Summarizing all written: This map lacks now: 1. fun from skill selection at low-mid levels. 2. reasons for combination more than 2 skills.
    1. This is bad, because this kind of fun guaranteed (1 skill point every level), achievable (you can see experience bar) and customizable (you can select 1 skill from 4-5 available for better fitting your role). I can say this part of game is important, because it demonstates your progress. Other features are too randomized (loot from mobs) / grindy (abilities/crafts for gold) / not soloable (bosses).
    2. I don't want to say that 25/25/0/0/0 builds is bad, but it would be great to see other combinations like 25/11/8/6/0 or 8/8/8/8/18 which will be more versatile.

    Solution
    So we need to accelerate feel of the progress on low-mid levels, and increase the significance of all skills. But we also must keep in mind that at high character levels skills should be highly correlated with equipped items.

    The solution lies on the surface - combine flat and percentage growth at skills. Flat growth must diminish with skill level, so percent growth will be more valuable at high levels.
    For greater significance flat values are combined in milestones. Each milestone has same increase in stats, but you need to invest more into skill to reach next milestone. The idea is illustrated in the attachment:
    FlatBonusSuggestion.png

    As you can see milestones are placed at skill levels: 1, 2, 4, 6, 8, 11, 14, 17, 21, 25.
    In total there are 10 milestones for each skill: 6 available with base class and additional 4 can be obtained with advanced class.
    Each milestone provides bonus which is noticeably boosts character role at low levels:

    Might - +1 Attack Power and +1 Spell Power
    Flat increment will give noticeable damage upgrade at low levels.

    Reflexes - +1 Armor Point
    Armor is always needed for tanks, also this bonus will help new damage dealers survive mobs attack when soloing in starting location.

    Spirit - +10 Mana Points
    MP are always needed for casters. This skill will compensate lack of mana if you use HP/Stats oriented items.

    Constitution - +20 Health Points
    Key feature for begginers tanks. Also this can help damage dealers with low HP items survive AoE/Aggro pulls.

    Perceprion - -2% Aggro generation
    Critical chance and damage are fairly universal stats, therefore another universal bonus connected to them. It also limits criticals for tanks, making place for more stable aggro generating skills.

    You can compare damage increase from Might with milestones (total increase = orange + selected bars):
    FlatAndPercentDmgIncrease.png
    As you can see - even small amount of points into this skill will give you guaranteed damage boost. But better boost you will receive with damage oriented items and high skill values.

    Also, if it's possible, small change in formulas from "Round Down" to "Round Up" will increase the attractivness of skills with percentage boosts (stats from Spirit, damage from Might, evasion from Constitution).

    I hope that these changes will increase the diversity of builds and will make game process more rewarding at low-mid levels. Which will attract more new players.

    Thank you for reading to the end :).
     
  2. kevs926

    kevs926

    Joined:
    Aug 12, 2010
    Messages:
    229
    Resources:
    0
    Resources:
    0
    i dont care on #1

    but i agree with #2

    for #2, a non-linear skill bonus is one way to address it. I dont like 25/25/ to be optimal builds. Also the skills should have much more impact. Meaning, increase their values, but decrease values from other sources (aka items)
     
  3. Ihazdialup

    Ihazdialup

    Joined:
    Jul 11, 2011
    Messages:
    599
    Resources:
    0
    Resources:
    0
    Well the the way this was already countered against for low levels is that now our characters build drastically faster on each level up than before, and that should havebeen extended to the base classes as well
     
  4. Xonok

    Xonok

    Joined:
    Mar 27, 2012
    Messages:
    3,042
    Resources:
    8
    Spells:
    3
    Tutorials:
    5
    Resources:
    8
    Okay, apparently there's one more person besides me that considers combining flat and scaling bonuses an option.
     
  5. Jumbo

    Jumbo

    Joined:
    Jun 22, 2007
    Messages:
    1,292
    Resources:
    0
    Resources:
    0
    flat bonus is a good idea, and I also agree with it (unless Zweib has a better idea). However, adding armor points through a stat is very wrong and armor should be kept solely for items. Keep in mind that armor values equals real damage. Therefore a tank will be able to reach very high armor numbers from a combination of items and this stat.
     
  6. ALEXz407

    ALEXz407

    Joined:
    Aug 23, 2014
    Messages:
    206
    Resources:
    2
    Tutorials:
    2
    Resources:
    2
    If skills will have a big impact, we again have a situation where 25/25 builds are best. It also reduces the fun of getting new items.
    So i tried to balance flat bonuses with this rules:
    1. Noticeable dmg up at low levels (almost 1/1 with start weapons) and low dmg up at high (1/3 or 1/4 with top weapons).
    2. Tanking accessory equal armor up (perhaps stronger at low levels).
    3. Two items equal HP/MP pools bonus (now it looks like x2.5-x3 bonus at high level, but current items isn't top tier) with bigger impact on low levels.
    4. Aggro reduction that overcompensates crit bonus damage at medium levels and equal to damage bonus at high levels.

    Yes, it raises base stats faster now. And it works well with linear increasing secondary stats like: AP, SP, ASPD and Armor Penetration.
    However, such secondary stats like Crit chance, Crit multipliers, Spell haste and Evasion now depreciate faster, because their values have inverse relation with base stats.
    OldNewStats.png
    I can not say whether it is good or bad. In any case, it does not provide such flexibility as flat bonuses from skills.

    New base HP/MP stats looks really good. Not sure about damage increase, but it comes with lowering base ASPD so i think it's good too.

    Let me insert a link on your vision of flat bonuses.
    The first way with ratings + diminishing (didn't play in LoL, but I think it's like in World of Warcraft) sounds good, but it is more difficult to understand and require secondary stats rework.
    The second way is what I suggest too. Milestones add sharpness to this system, because player often meets with the choice "add 1 point to this skill for big advantage in secondary stats now, or invest it into main skill with big advantage later."

    Well, Zwiebelchen wrote that he wanted to "make spirit also affect armor stats of items". So i don't see a problem with adding armor points with skills. I'm just afraid what bonus will be in percents, what greatly affect high level tanks and adds nothing to low levels.
    Talking about numbers. Flat bonus "10 armor" at 25 skill level isn't too high. It's equal to armor of tanking accessory from foothills. My crusader has 79 armor points in foothills items. Additional 10 armor will help with spawns on Brood/Garg, but agains Gazrow hits for 250+ damage it doesn't help a lot. Do not forget that d4 will be much stronger area, so bonus will be even less noticeable.
    At low levels it will greatly help with solo killing mobs at wilderness, but don't make tanks too sturdy for damage in d1 (+3/+4 armor doesn't help much vs boss with 30 dmg and trash with 20 dmg). Also do not forget about Armor Penetration at mobs that raises minimum damage to tank.
     
  7. Shamu

    Shamu

    Joined:
    Jun 24, 2014
    Messages:
    37
    Resources:
    0
    Resources:
    0
    Using bars in your graphs is a lot more confusing than lines would have been imo.

    Crit% and Critdmg% are not the same for lvl 1 and lvl 50. The more crit you have, the stronger Critdmg% gets, and the more Critdmg you have, the stronger Crit% gets. this skill scales with items and talents.

    As long as there are only stat gains on skills, there will always be an optimal build. If it isn't 25/25, then some other build will be optimal, and all others won't be.

    If its not a universal 25/25 optimal build, then at least every class/role combination will have one, and they will probably overlap with one another. I don't see why 25/25 should be bad, and another fixed optimal build that is spread across more skills would be better. They both allow for the same amount of versatility.

    A way to change this would be to introduce hardcaps on stats and being able to reach these. If one class can reach 100% crit somewhere during d4 progress, then the perception skill would lose part of its value. As long as this is not possible for all classes, but only one or two, there would be classes that have to skill according to their itemization, rather than faceroll a haste/crit build. If this cap is reached relatively late, then the skill reset costs are a very effective goldsink for high end content.

    While crit is the most obvious hardcap, others could still be introduced, like having critdmg cap at 200%, haste at whatever% and so on. As long as the hardcaps are communicated, and not just some obscure values that can't be found anywhere, skill distribution would depend heavily on itemization. And yes, I played a lot of world of warcraft. Try coming up with your very own BiS ICC gear (when there were a lot more different stats than now), and you'll see how important hardcaps and softcaps are in their stat system, and how they force you to think about what you're doing.

    Having auras that grant stats would then also mean that you would need multiple item sets to handle stat buffs effectively.

    I would like to mention that the crit cap was a stupid example, and that crit isn't necessarily a stat that should be capped, unless you are at the highest tier that will ever be made. Hastecap and Critdmgcap seem like a better idea.

    This doesn't make the low and mid levels any more interesting though, so it only solves half of the problem. then again, when I play a character that isn't at the levelcap, I don't expect him to be 100% balanced and I don't expect gear and skills to be mindblowing. Levelling up is always somewhat of a chore that has to be done before reaching the "real" content, since anything that you gain while levelling up will be heavily outclassed by maxlvl gear.


    tl;dr: Hardcaps and Softcaps are great, mkay? And endgame content > leveling content
     
  8. ALEXz407

    ALEXz407

    Joined:
    Aug 23, 2014
    Messages:
    206
    Resources:
    2
    Tutorials:
    2
    Resources:
    2
    I used bars to show difference between damage increase at each skill level. The trend line is noticeable still.

    You are right. But I wanted to say that 1 point in Perception will grant you flat 1% crit chance and 2% crit damage both at level 1 and level 50. For comparison, Might will not grant you anything at low damage/skill level so it's useless on low levels.

    I agree.

    Great idea. I spent some time to figure out what stats can be capped, here is a list:
    Flat values, doesn't capping: Hitpoints, HP regen, Manapoints, MP regen, Armor, Armor penetration, Attackpower, Spellpower, STR, AGI, INT, Run speed.
    Percent values, linear increase, can be capped at any value: Crit melee/ranged damage, Crit spell damage.
    Percent values, linear increase, useless over 100%: Crit chance.
    Percent values, linear increase, eliminates game mechanic at 100%: Evasion, Resistances.
    Percent values, exponential increase, infinite power at 100%: Attack speed, Spell haste. The chart below illustrates what it means.
    AspdSpellHasteIncrement.png

    Warning about reaching stat cap can be placed at stats window. Something like that:
    Normal: Critical Strike 35%
    Capped: Critical Strike 70%+15%
    And warning at the top of window right above Spellpower (if there is at least one capped stat): Red stats are overcapped and grants you nothing!
    StatsWindowChanges.png
    Also, they can be grouped:
    >>> Cap warning will appear here <<<
    Damage
    - Critical Chance xx%+xx%
    Melee & Ranged
    - Armor Penetration xxx
    - Attack Haste xx%+xx%
    - Critical Damage xxx%+xx%

    Magic
    - Spellpower xxx
    - Spell Haste xx%+xx%
    - Critical Damage xxx%+xx%


    Defense
    - Evasion xx%+xx%
    Resistances
    - Fire xx%+xx%
    - Water xx%+xx%
    - Lightning xx%+xx%
    - Magic xx%+xx%
    - Shadow xx%+xx%
    - Poison xx%+xx%

    This will make stats window more systematic, but will increase its width and height.

    Now about cap values.

    Crit chance cap: 70%
    70% crit chance makes crit cap achievable. Also high crit gain will benefit STR&INT characters more for AGI collection.
    Formula: 5% + (AGI / [Base AGI] - 1) x 25%
    Skill: Perception (+1%)
    Talents: 5% - Druid | 10% - Bishop, Sorcerer, Assassin, Berserker | 15% - Monk, Sorcerer, Necromancer, Hunter, Crusader
    Selfbuffs: -
    Buffs: -
    Auras: 3% (maybe increase to 5%?) - Pack Leader (Hunter)

    Crit melee/ranged damage cap: 225% (75%)
    Crit damage is highly scaled with STR. This is done to make AGI characters attack faster&weaker, when STR chars can crush enemies with devastating crits. It also makes rangers not so overpowered.
    Formula: 150% + (STR / [Base STR] - 1) x 40%
    Skill: Perception (+1%)
    Talents: +50% (maybe nerf to 30%?) - Assassin | +30% (maybe nerf to 20%?) - Bard
    Selfbuffs: -
    Buffs: -
    Auras: -

    Crit spell damage cap: 225% (75%)
    Magic crit damage scales similar to melee/ranged.
    Formula: 150% + (INT / [Base INT] - 1) x 40%
    Skill: Perception (+1%)
    Talents: -
    Selfbuffs: -
    Buffs: -
    Auras: -

    Attack haste cap: 70%
    70% attack haste means 333% increase in white melee/ranged damage. However, this cap easily reachable only for thieves. Other characters will stay near to 25%-50% where attack speed gives 135%-200% boost (that can be easily upgraded to 200%-333% boost with Bishop's buff).
    Formula: (AGI / [Base AGI] - 1) x 25%
    Skill: Reflexes (+1%)
    Talents: +5% - Monk, Hunter, Assassin, Bard
    Selfbuffs: +10% - Berserker | +150%?? - Ranger
    Buffs: +15% Symbol of Fury (Cleric) | +25% - Symbol of Fury (Bishop)
    Auras: -

    Spell haste cap: 70%
    Similar to attack haste 70% of spell haste give you 333% increase in damage and mana consumption if you can use skills without delays.
    Formula: (AGI / [Base AGI] - 1) x 25%
    Skill: Reflexes (+1%)
    Talents: +10% - Bishop, Sorcerer, Necromancer, Druid
    Selfbuffs: +50%?? - Ranger
    Buffs: +5% - Song of Peace (Bard)
    Auras: -

    Evasion cap: 75%
    Difficult achievable and very usefull stat. That is Why it grow mainly with items. At cap only 25% melee/ranged attacks will hit tank.
    Formula: 5 + (AGI / [Base AGI] - 1) x 5%
    Skill: Constitution (+0.5%)
    Talents: +10% - Berserker
    Selfbuffs: +up to 20% - Monk
    Buffs: -
    Auras: +3% - Confidence (Cleric) | 7% - Pack Leader (Hunter)

    Resistances cap: 75%
    Spell resistance is situational defense stat. That is Why it can easily be stacked to max 75% with items/buffs/skills even without main stat formula. Capped at 75% so magic attacks will do 25% damage to characters.
    Formula: -
    Skill: Spirit (+1%)
    Talents: +5% - Hunter
    Selfbuffs: -
    Buffs: +10% All - Fire Shield (Necromancer) | +10% FWL - Song of Elemenets (Bard)
    Auras: 15% All - Confidence (Cleric) | 25% All - Confidence (Bishop)

    I used the following priorities when writing formulas and skill bonuses (talents/buffs/auras are without modifications):
    For damage dealers and healers: STR/AGI/INT on item > Secondary stat on item = Skill > Talent = Buff/Aura
    For tanks: Secondary stat on item > Skill > STR/AGI/INT on item = Talent = Buff/Aura

    And you can see that now there are softcaps. For example:
    - Attack haste softcap for Monk 55% (+15% from Divine Fury)
    - Evasion softcap 65% (+3% from Confidence and +7% from Pack Leader)
    - All Resistances softcap 40% (+10% from Fire Shield and +25% from Confidence)
    - Attack haste softcap for Berserker with Bishop 35% (+10% from Berserker's Rage and +25% from Divine Fury)
    - e.t.c.

    Creating an ideal build is really endgame fun. During leveling most fun (imo) are: choosing skills and feeling how you character become stronger with your choice, real increasing in stats with new gear, interesting quests and boss battles.
     
  9. Zwiebelchen

    Zwiebelchen

    Joined:
    Sep 17, 2009
    Messages:
    6,804
    Resources:
    12
    Models:
    5
    Maps:
    1
    Spells:
    1
    Tutorials:
    1
    JASS:
    4
    Resources:
    12
    I'm currently thinking about removing the skill system completely and instead move the talent tree from the bag to the main hero instead.

    I feel like the skill system is just an unneccesary element that more or less always leads into mono-statting, no matter how you change it.
    There's no real decisionmaking involved, except for which stats you up. Everyone will more or less always end up with 25/25/0/0/0 ... so I could also just make it 1/1/0/0/0 in the first place.

    That's why World of Warcraft changed talent trees completely to work more like one-shot decisions that affect your skills, always presenting you with choices of three of which you can always select only one.

    I'm currently thinking about possible new ways to make the base classes more interesting on level up aside from skills. Maybe kind of a copy of the talent system, just for base classes with 3 general trees?


    Afaik, WC3 allows for 6 skills in the levelable skill menu. This means I could easily have the 3 talent lines for advanced classes AND 3 talent lines for the base classes in one menu. The advanced lines get "unlocked" once you reach level 30 in the same way as they get unlocked now. The three base lines are always there and you get one point every 4 levels. The advanced lines have the requirement of all "base points" spent in order to be selectable.
     
  10. Ihazdialup

    Ihazdialup

    Joined:
    Jul 11, 2011
    Messages:
    599
    Resources:
    0
    Resources:
    0
    If done right I would be good, though I think my own idea would be you have the advanced trees in the backpack still and are practically kept the same, but have the base talents on the character that you get every 2 levels or something and make the base trees bigger.

    That my own personal thought, but that could end up being very OP...
     
  11. BoxOfFaith

    BoxOfFaith

    Joined:
    Apr 16, 2014
    Messages:
    379
    Resources:
    0
    Resources:
    0
    True diversity I don't think will ever exist within the game. Its unlikely that different people will end up loading assassins built in a different way. A meta build will always pop up, and it will always be widely used. I think worrying about these issues will just lead in a big circle of dissatisfaction.

    I think the best target to aim for is ensuring that every class can fill a role relative to one another. Making sure all the DPS classes are relatively equal, or each have their own unique perks. So even if everyones Assassin is more or less the same, hopefully theres little incentive to take multiple assassins in a team, as if someone is already on one, hopefully a different class could fill a second DPS role in a better way. Much like a Hunter could be considered more valuable for the gargoyle fight due to its ability to range.

    However customization through items could be a thing depending on the situation. If you have an undergeared tank, then more tanky gear on your Assassin could be more valuable than full on DPS gear for example.

    On topic of this thread, I think its unnecessarily convoluted. I think you're aiming way too hard for perfection, something that games like WoW honestly stop caring about and they just ensure that nothing is broken, and continue to focus on pleasing their high level players. The low level gaias game isn't broken, its definitely still very playable. Could it be better? Probably, but as long as its playable and not broken, I don't have much issue with anything.
     
  12. kevs926

    kevs926

    Joined:
    Aug 12, 2010
    Messages:
    229
    Resources:
    0
    Resources:
    0
    +1000
     
  13. Zwiebelchen

    Zwiebelchen

    Joined:
    Sep 17, 2009
    Messages:
    6,804
    Resources:
    12
    Models:
    5
    Maps:
    1
    Spells:
    1
    Tutorials:
    1
    JASS:
    4
    Resources:
    12
    The question for me is: would removing the skill system in general have a negative impact on the map, as in: would players actually care about the loss of "stuff to do when you level up"?

    Obviously, if there never were a skill system in the first place, people wouldn't care, but now it would feel like I'm taking away the candy I just gave.


    I think I'll experiment with some base talent lines on paper for now. Maybe, if I can come up with something interesting, I'll add that.
     
  14. ALEXz407

    ALEXz407

    Joined:
    Aug 23, 2014
    Messages:
    206
    Resources:
    2
    Tutorials:
    2
    Resources:
    2
    If new talents lines will be balanced as available now, nothing really will change, because biggest impact on character power obtained from last skills in talent line. So it will look like: choose 1 talent line from base class, choose 1 talent line from advanced class, spend remained points in stat increasing talents.

    It would be epic If you can implement talents system like currently used in WoW. Each time you choose one from three talents that have approximately equal influence on character's power.

    [​IMG]

    However, this is a whole bunch of work:
    5*5*3 (75) talents for base classes.
    6*5*3*2 (180) talents for advanced classes.
    Even if we replace some talent lines with flat (base class) and percent (advanced class) stat gains, we still have at least 30 jass coded talents for base classes and 90/120 jass coded talents for advanced classes.
    And who knows if it is possible to replace the standard abilities system.

    What do you think?

    P.S. Another possible combination of talents: 45 jass coded talents for base classes and 60 for advanced.

    [​IMG]
     

    Attached Files:

    Last edited: Jan 5, 2015
  15. Zwiebelchen

    Zwiebelchen

    Joined:
    Sep 17, 2009
    Messages:
    6,804
    Resources:
    12
    Models:
    5
    Maps:
    1
    Spells:
    1
    Tutorials:
    1
    JASS:
    4
    Resources:
    12
    @Alex: the "select one of three" instead of "up one by one" is actually possible through engineering upgrades. I used that method to apply the talents to the bag. And it's not like I'd need to recode all the talents, because they are already there. It just changes the way you acquire them (and adds some minor object data ... nothing huge).

    In fact, it would make stuff easier, as then I can go with mutual exclusive talents, so I don't have to care about making said talents work in combination (like for example the "have 2 spirits" talents, which gave me a real headache to code).

    I guess that's the next goal for me after removing the skill system in general. It would be easy to extend it to lower levels in the process aswell.

    I think the "select 1 of 3 every 5 levels" would be the optimal thing (I'd exclude the first 10 levels to make it match the game's learning curve).

    That means you get one of three talents at level:
    10 (tier 1 base choice)
    15 (tier 2 base choice)
    20 (tier 3 base choice)
    25 (tier 0 advanced choices)
    30 (tier 1 advanced choices)*
    35 (tier 2 advanced choices)*
    40 (tier 3 advanced choices)*
    45 (tier 4 advanced choices)*
    50 (tier 5 advanced choices)*

    *those already exist

    This means I'd need to code 3*3*5+3*10=75 new talents. Sounds a lot, but isn't, if you consider that we already have 3*5*10=150 talents. And let's be honest: base class talents will probably be pretty simple adjustments of certain ability or stat values.


    I like that. This is what is going to happen next.

    The only problem would be the save/load compatibility. But yeah, I think I can handle that through -comp imports.

    ---------------------------

    On top of that, I finally did the maths about a possible enchanting system ... and the results look good. In fact, an enchanting system that would allow upgrading each item would have a weaker impact on the save/load length than the current skill system (26^5 > 2^24*). The only problem would be the bloat in object data (two item objects instead of one) ... but I think the game can easily handle that.
    There would not be choices here; the item would basicly only upgrade up from the initial state to the upgraded state.
    Obviously, the material cost for such upgrades depend on the item level and quality grade.
    I'd also add a new stackable material type to the game that is enchanting dust, which you can only get from disenchanting magic items.

    *24 because there's a total of 24 equipment slots that can hold items
     
  16. kevs926

    kevs926

    Joined:
    Aug 12, 2010
    Messages:
    229
    Resources:
    0
    Resources:
    0
    first of all, it won't always lead to mono-statting.

    Decreasing bonuses and more levels per skill
    Make it so the optimal build is the more spread out skills, rather than maxxing one stat only. They would only mono-stat if they really want a pure tank, pure dps, but will come with more sacrifices. The values per level is highest at level 1, and decreases for each level, making it lowest at level 51. 51 levels sounds ridiculous, but just incase someone wants glass cannons or super tanks they got it. Although the values at level 51 should be really small, for people to even think about building like this.

    Different Values for each class
    Due to the restriction you made that all heroes will have the same 5 skills (with same values per level also) to choose from, the values cant be increased/decreased to the extent that will make a certain character too imbalanced. You must adjust the value accordingly

    Useful and Unique
    This one is achieved by 1.2a(9) version somehow. You managed to make skills useful for any class and different from each other. Some values are just too weak. Although Luck, Speed, Power (correct my term usage, i forget them) all boost skill effectivity, it doesnt boost all skills, thats why they are on separate categories and we have to decide which one we go for. For example: Crits only affect dmg dealing skills (altho it affects heal now, but you really should revert it back). Haste doesnt affect instant cast skills like bladefury, crippling curse. Power affects all spells but isnt as effective as haste and crits.
     
  17. Zwiebelchen

    Zwiebelchen

    Joined:
    Sep 17, 2009
    Messages:
    6,804
    Resources:
    12
    Models:
    5
    Maps:
    1
    Spells:
    1
    Tutorials:
    1
    JASS:
    4
    Resources:
    12
    This is a quick sketch I made at work, about possible talents for base classes:

    Talents for advanced classes will have the same effect as they have now, but also get changed so that you have the "1 of 3" choice every 5 levels instead of "up the line" progression. Some values get adjusted to match the power of the talent choices of each tier.
     
  18. Shamu

    Shamu

    Joined:
    Jun 24, 2014
    Messages:
    37
    Resources:
    0
    Resources:
    0
    That looks pretty nice actually, but I do see a problem. As a crusader tank, you will want to be tanky and you will want to keep the aggro. Having 15% more threat and -15% crit dmg from enemies is something you will really have to think about, which is good. So if you have no problems with holding aggro, you'd want to take less damage, right? But what if the group setup changes? All it takes is one overgeared or overleveled dps to make you regret not taking the increased aggro. The current cost for resetting talents is too high to allow for a reasonable variation. Would the cost for resetting base class talents be seperate from the advanced classes? Or you could allow for much simpler resetting by removing costs and making it possible without an npc (for base classes only). This would also make it more forgiving and give reason to experiment for new players. I really feel like this should be something that you can change freely, since the aggro boni/mali seem very strong.

    EDIT: and also, while I don't know what bosses will come with d4, the base talents look like the optimal choice will shift depending on the mechanics and nature of a given encounter. But I don't want to go on about free base respec forever.
     
  19. Zwiebelchen

    Zwiebelchen

    Joined:
    Sep 17, 2009
    Messages:
    6,804
    Resources:
    12
    Models:
    5
    Maps:
    1
    Spells:
    1
    Tutorials:
    1
    JASS:
    4
    Resources:
    12
    There will be no such thing as a seperate base talent and advanced talent reset if this change comes. They will both basicly be part of the same system.

    Respeccing cost is not really a problem, since that can be changed easily. It won't be free, though, as I want people to think about their choices strategically. A certain minimum amount of penalty is neccesary to give the required impact.
     
  20. PachaXD

    PachaXD

    Joined:
    Jan 19, 2012
    Messages:
    124
    Resources:
    0
    Resources:
    0
    changing to a talent trees system seems like the best option for me, it wont only increase the weight of the decisions we take while leveling, it will also make those decicions really affect the way our character works. If we talk about diversity of builds, that WONT happen, like never... always one build will be top one, will provide the best combination of bonuses for a purpose, the only real divversity of builds are simple choices like: i want to build my serk as a tank, or as a dps, my bish as a healer or a dps, my monk as dps or as a tank. In any game, even if we talk about wow, theres a conbinations of skills that will be optimal for a certain thing (mean it pve and pvp at WoW┬┤s example mostly) we just cant avoid that from happening.