Thoughts on character/skill tree design

This is the place for people to research themes and ideas for their mods as well as for others to post their suggestions for mod makers to make use of.

Moderator: Nefarius

Total votes: 1

User avatar
Retired staff
Posts: 2498
Joined: Tue Mar 04, 2003 11:40 pm

Post by Char » Sun Mar 26, 2006 9:02 pm

Laz, I think you should have read the question :P.

However, I suspect that he really meant the Holy Shield skill, as that one is the only one that cannot be used to damage enemies.

User avatar
Posts: 24
Joined: Thu Jul 20, 2006 5:21 am

Re: Thoughts on character/skill tree design

Post by d2modder » Sun Jul 23, 2006 5:04 am

Wow, I must say that you even impress me, i agree on the paladin issue -> which happens to be my favorite character, now one hes supposed to be a buffer cuz well hes a paladin, BUT at least 50% of his skills needs to be non buff, and it would be nice if all characters had at least 1 buff and 1 passive skill on them, depending on the character, the mage should have meditation, the amy have fanticism, barb have thorns, necros have defince, and maybe rename them??? ideas of course, P.S. this helps in my modding progress {M}
Im here to help those who seek assistance, i will help to the best i can and also list alternitives if possible :)

User avatar
Posts: 10
Joined: Tue Oct 31, 2006 10:30 am

Re: Thoughts on character/skill tree design

Post by ArTiSiMo » Wed Nov 01, 2006 2:42 am

im pretty new to modding skills, and i dont know much about it. i was wondering if you could give syngergy to pet type passives or pet type skills that can benefit your character, and/or pet.

User avatar
Junior Member
Posts: 152
Joined: Sat Apr 26, 2014 9:19 am
Korea South

Re: Thoughts on character/skill tree design

Post by mmpx222 » Thu May 01, 2014 7:13 pm

Okay, here are some ideas on skill design, it would be better to put this here since this thread is already chock full of valuable information :)

When designing skills, I often come across this question: how do I want to tweak the numbers? There are 3 ways to go about this, and I'd like to analyze their pros and cons.

Linear Progression

Skills with linear progression have damage/duration/etc. that increase at a fixed value per each skill point. This is used in many LoD skills, and the entirety of the synergy system in LoD works in this fashion.

  • Easy to understand, easy to implement
  • Works for almost everything: mana cost, missile range, spell radius, etc.
  • If the skill provides bonuses that are not modified by items, the skill may not scale well enough at high levels (e.g. psychic hammer damage)
  • If the skill provides bonuses that are multiplied by equipment mods, it may be overpowered at higher levels (e.g. +% life)
Recommended Use:
  • Utility spells, whose effectiveness should be easy to understand.
  • Any form of skill that would otherwise become too complex to understand.

Increasing Returns

Some skills become far better the more skill points you put in them. Ever wondered why nobody recommends using Meteor as a 1-point wonder spell? It's because the damage ramps up sharply at level 17 and beyond. Many damage-dealing spells in LoD use this type of progression.

  • Scales well with game difficulty, which usually increases exponentially in Nightmare and Hell.
  • Prevents skills from becoming a 1-point wonder, even for high level players with +20 or more to all skills.
  • Skills with no conflicting alternatives will become a "mandatory" investment, becoming a blackhole of skill points. Too many of these skills, and the player is going to feel massively limited by the lack of skill points.
  • Any skill that directly improves other skills or stats (e.g. Masteries) may become overpowered.
Recommended Use:
  • Damage-dealing spells that don't depend on weapon damage.
  • Skills that define a certain playstyle and/or should not be mingled with other skills
  • Any skill that should be a significant investment and not be a 1-point wonder.

Diminishing Returns

Skills with diminishing returns are better for their investment at lower levels, but are not worth maximizing because they become less effective at higher levels. Many passives, auras, and curses in LoD use this type of calculations.

  • Easy to balance for high-level characters with massive +skill bonuses.
  • Encourages players to think rather than blindly investing into a skill.
  • Rich players will often put only 1 point in the skill, relying on +skill bonuses to push them to higher slvls.
  • Inflates the number of skills a player has, reducing the number of potentially viable builds.
  • + to all skills/+ to skill tree mods become ridiculously important.
Recommended Use:
  • Passives, auras, and other support spells.
  • Any skill that improves the overall efficiency of the character and does not define a certain playstyle. (e.g. +Magic Find)
  • Any skill that affects a property that can also be provided by stat points or items. A skill should not provide a linear or exponential +Life bonus, since it makes the character overpowered and/or removes the need to look for +Life mods on items.

Personal tip, if you're unsure which one to use, go with the linear progression first.
D2TXT / D2INI - Python scripts for editing TXT files, or converting between TXT ↔ INI files

User avatar
Frain Breeze
Posts: 4
Joined: Sat Nov 26, 2011 2:35 am
Location: New Jersey
United States of America

Re: Thoughts on character/skill tree design

Post by Frain Breeze » Thu Jul 03, 2014 4:47 am

I haven't seriously looked at modding in a couple years now, but mmpx222's post inspired me to toss this out there:

Diablo 2 needs inverted skill trees.

What's that supposed to mean? All of the skill trees in Diablo 2 start with one or two passives or absolutely basic moves, increasing in complexity as you go down the line and branch out. It makes sense in that your character is learning their skills. Problem is, forcing you to start with the basics kills replayability.

The reason is simple: It makes every single character play the exact same way in the early levels. There are about 30 pure combat skills available before you hit level 12. Of them, half are functionally identical to Normal Attack with a chipped gem or speed modifier on your weapon. The remaining half has a lot of overlap in itself, with things like Multishot, Charged Bolt, and Teeth being their own classes' takes on the same idea. Then take out the things that aren't even worth considering as a main attack, like Holy Bolt and Psychic Hammer, and you realize that there really is no variety in the beginning of the game.

This is why I've never been able to get back into Diablo 2 as strongly as I used to. I know that before I get to any exciting bits, I have to play my character up to level 12 or even 18, using the same exact play style as countless other characters before it.

So what I propose, for anyone still modding this game, is to flip the skill trees. Give them the interesting stuff right away, so their new character feels like a new character as soon as possible. Save all the passives for when they need a place to dump spare points, and either replace the dopey single-target spells with something more interesting, or make them into quick "sniping" skills with a high mana cost that you use to take out stuff like shamans and aura bosses.

User avatar
Dark Alliance Beta Test
Champion of the Light
Posts: 256
Joined: Fri Jul 19, 2013 11:13 pm

Re: Thoughts on character/skill tree design

Post by Doub » Thu Jul 03, 2014 12:11 pm

With todays standards D2 Vanilla is a finished game from every aspect. Simply not worth wasting your time. The only thing that i still admire about D2 is how the 7 heroes look like.

User avatar
Forum Regular
Posts: 542
Joined: Sat Aug 31, 2002 8:26 pm
Location: Brazil

Re: Thoughts on character/skill tree design

Post by Desocupado » Thu Jul 03, 2014 12:49 pm

Making weaker versions of similar spell obsolete (example: Cold Bolt -> Freezing Blast -> Glacial Spike -> Frozen Orb)
-Make low skills really obsolete with lower level caps (thus capped damage) and higher damage at level 1.
-Assuming the 100-ish skills point limit, we could make it goes like: Cold Bolt (5) Freezing Blast (5) Glacial Spike (5) -> Frozen Orb (15) and each progression requires max level of the skill before.
This is done by using a formula in each skill point cost calculation field - (stat('previous skill'.blvl)=5)?1:stat('newskills'.accr)+1
Making forced choices (the idea is around for some time)
-Taking the theme above, let's say we want cold sorceress to have either huge aoe (Frost Nova/Frozen orb/Blizzard) or direct damage (Cold Bolt/Freezing Blast/Glacial spike).
Change Cold bolt cost to (stat('Frost Nova'.blvl)!=0)?stat('newskills'.accr)+1:1 and frost nova to (stat('Cold Bolt'.blvl)!=0)?stat('newskills'.accr)+1:1
-You can also make multiple verifications - let's say you can either pick thunderstorm, teleport or energy shield.
Change Energy Shield cost to (stat('Thunderstorm'.blvl)=0)*(stat('Teleport'.blvl)!=0)?1:stat('newskills'.accr)+1
Synergies based on character level
-By using level in some calculation fields (including synergy) we can give an overall progression when a character level. A level 100 Necromancer should be able to control more minions than a level 1.
In this case we add "+(stat(‘level’.accr))/10" to petmax - giving one addition skeleton every 10 levels
You could even make all skills cap at 1, and only increase their power with levels (this idea is good along a different skill point gain rate or a lower level cap)

Post Reply

Return to “Mod Concepts & Research”