Skill trees

Im not saying make level 30 skills and keep the trees the same. They would have to be changed for that very purpose of not maxing all routes. But think back to d2. Every level that you got was a big deal because it was another skill point in your skills. Ven level 99 to 100 you would be like yes!!! Skill point now where to put that. That’s the feeling I want from this game. Right now you hit level 20 skill by level 40-50 and then its just blah try something new. I’m not saying that giving skills 30 levels or giving a skill point for skills each character level is the right way butt I want that feeling that every level matters

And d2 made you make very very hard decisions.

Seems we have very different memories of D2 :thinking:

All skills were locked behind a level requirement. The highest skills needed level 30 to unlock. And those weren’t the mightiest skills in the game. For example a nova sorc got her main skill with level 12. So you skilled 20 points (I think this was max cap) in nova and are out skilled with level 32. And all the skillpoints did was increasing the main stat (damage / duration /…), nothing changed mechanically. So the rest of your progression you leveled synergy skills that did the same like leveling the main skill but with lower impact.

Your damage was maxed out with around lvl 70-80 when all skills and synergies were maxed. Till max lvl 99 you only could improve your passive stats.

I never leveled a character above lvl 90 because it was not necessary.

So there was this one moment early to midgame when you got your favoured skill. Rest was just constant damage increase with no climax at all.

What made decisions matter was the fact that you couldn’t respec.

When you think about D2 you probably see things very biased because of the good time you had. But if you would use the same mechanics like back then people would not like it today. It was a different time with different expectations.

well its been about 10 years so yea idk, i remember playing necro and there were several skills and each point i put in could give them damage or raise 1 more maximum. I barely remember but i do remember that each level was like… Yes!!! now where to put it!! How often between skill level 15-20 do you say oiii boyyyy here we go!! Never cause you already got the skill the way you want it build now its just a twinkle of added filler.

I could image a second tree in the tree that you unlock after hitting level 20. With this tree you could increase all the passive bonuses of your skill.

But afaik Sarno stated that they are going to increase skill level progression time. Additional skillnodes would make respecing much harder I think.

1 Like

Thats even more of the problem… so level 15-20 dont really hold any wow factor and now were going to make those levels slower and harder to get and respec :frowning:

That could be a really cool addition. Each skill could maybe to go level 25, but 21-25 points were specifically for an unlocked “elite” skill tree that doesn’t appear until level 20. Those levels past 20 could take a lot longer to get to as well. Similar to leveling “Awakened” Gems in Path of Exile.

1 Like

Sounds good, though Disintegrate needs to get the “Extra Wet Noodle” & “Can’t aim up stairs 'cause I’m a Dalek” nodes. And Shield/Hammer Throw need the “Redheaded Stepchild” node.

Yeah! But Desintegrate would stay as week as it is compared to the other skills. It should be viable before hitting the 20+ extra perks because other skills are right now. This skill built towards damage should melt the stones that prevent you falling from the map. Worst skill right now for me. The tree and the perks sound nice. Just performance is u underwhelming.

I’ve been thinking about this while I’ve been making my tempest strike build and am conflicted about skill extra points. When I look at the other 4 skills I’m using right now, it doesn’t feel like they need an additional 5 points to make them any better. Avalanche, ice thorns, thorn totem and storm totem feel fine with 20. 5 extra points in tempest strike, however, would be really wild. I could pick more baseline proc chance and flat damage or you could really justify going all in on buffing the individual element if you choose. Again I’m conflicted on it because you can look at the Tempest Strike tree and say “wow, this is a good tree because I’m forced to make a decision and go a certain path”, unlike the others I’m using that seem to only have one way to go. Is the middle ground on this being able to select a “Main” skill and get a few bonus points to spec into it or are we being too early to judge and there is still significant tree rework coming for most of the existing skills?

1 Like

I wouldn’t say significant reworks for most skills. I think most skills are in a pretty good place and just need some tweeking to get them in to that sweet spot. Several high profile skills are in desperate need of reworks but we have to prioritize between reworking old trees and making new ones for new skills. I’m sure skill trees are something that will continue to get tweeked and fully rebuilt after 1.0 we don’t like to set things in stone in general.

1 Like

Is the general intent to keep the # of nodes in a skill tree roughly at current levels, or to possibly add to it in the long term to maybe a 50% increase from current levels, or even more?

Currently the plan is to have about 25 to 30 nodes per skill tree. Some trees are significantly under this number. Those ones will get more nodes. We find that’s generally the sweet spot where we can get a tree to not have too many dud nodes and still have interesting choices that pull you in multiple directions while still being digestible. We tried making a tree with 42 nodes (summon skeleton), it didn’t work out too well and we split it in to two trees that each are much better I think.

2 Likes

What would be your thoughts on say once you maxed a skill out, it gained “paragon levels” and you could get say another 1-100 levels (doesnt really matter but there would have to be a limit) and each level you gained (would take same exp as level 19-20 did) gave the spell a specialality of your choice i.e.
you maxed out avalanche
now you get to chose a mastery for it (and level it another 20? levels)

  1. mana efficiency (Each level gave 1% Mana efficiency)
    2)Damage increase(Each level gave 1% more damage)
    3)cold/physical penetration(Each level added 5 cold/physical penetration)
1 Like

Flat or % penetration, since I suspect 1% increased damage is likely to be more effective than 5 penetration.

idk im not the balance guy lol. Just throwing out some examples. Maybe down the road penetrations are more effective :slight_smile:

Well, I can say some more about all this but that’s on my own taste of course.

  1. Player should have an opportunity for early building (good @TriKster’s example with hammerdin). But it feels as some kind of a loss when you reach skill’s level cap.
  2. It’s always nice to have more levels for… everything. But it’s always boring to increase same little things in no matter how many ways, like +10% damage from equipment, some buffs and 3 “different” 10-level skill-nodes. And it is a real problem of LE for me.
  3. It is really nice to have a feasibility to improve skills AFTER reaching their level cap. Right now it’s implemented via Legendary items and Idols. Don’t know is it possible for every skill.
  4. Player’s choices shouldn’t get “older” - skill’s modifiers should not loose their effectiveness. For example, Fire shield grants you 100 elemental protection. But when you get +500 from your equipment and passives it feels really weak. At the same time it must be balanced on every stage of your playthrough, because it’s pretty frustrating when your skill becomes useless at the end or overpowered / mandatory at the beginning (like Flame Ward).
  5. Every option should be viable and compatible with any other options. Ideally: every option should be equally effective to provide an opportunity to build that you like rather than that is just effective.

So, how can we achieve all this? In my opinion …

  1. Skill level 100: Ideally, skills should have exactly the same level cap as a character. This allows them to progress with character and to improve overall progression itself. Also it should help them to stay strong enough even without any side influence like unique items, idols etc.
  2. Main skill and 3/6 core nodes: Although any skill nodes may improve or even significantly change main skill, I believe that we should highlight “core nodes” - those ones with great bonus and significant functionality effect, and separate them from smaller ones with less influential effect and bonuses. And what is really important: it should be possible to CHOOSE only some of them and always the same quantity from same quantity for every skill. I’m sure it’s enough to have 6 core nodes with player’s ability to choose 3 of them.
  3. 240 levels to choose: At last, every core node (and main skill) is surrounded by smaller nodes which summary level for every core node equals 33-37 (~35). That means that with main skill and 3 chosen core nodes we have 4 * 35 = 140 potential levels for small nodes. But we have only 100 points to spend. And globally we even have 240 levels for small nodes. And still 100 points to spend only.
  4. Node bonuses should NEVER duplicate each other’s functionality! I hate PoE with it’s 90% +stat nodes, and some Epoch trees makes me sad with one node granting +10% damage and several others granting +8% or +12% or … . Yes, there are more nodes than attributes even in combinations, I know it. But we should always try.

3 out of 6 core nodes
100 out of 140 surrounding small nodes’ levels
100 out of 240 total small nodes’ levels
… for every skill
… Imagine level of customization :heart_eyes: (God I’d wish to see anything like this at any game!)

  1. Most bonuses should be applied multiplicatively to keep them useful at any level of progression. Exceptions: attributes based on other multiplicative attributes, or attributes that have non linear effect… or because of common sense. For example: damage bonus is always multiplicative (unless otherwise specified); area radius is always additive because each next radius increase produces greater increase to overall area.
  2. And a difficult one to accomplish: If node grants additive bonus than it should also grant functionality to greatly improve bonus effectiveness at some conditions.

So, how it works? I’ve decided to fantasize about "how would mage’s Fireball spell look like with all this?"

Warning 1: it’s NOT finished. But I’m eager to finish it!
Warning 2: I’m designing this using my 4K 55 inch TV, so it may require to use scaling for some of you.
Warning 3: Possibly a lot of spelling problems - English is not my native :blush:

So… meet 145 / 240 level Fireball !

Notes about my doc: all percent damage bonuses are multiplicative. Crit chance / mult, any speed, area radius, distances, effect duration, etc. - additive (unless otherwise specified).

And some more thoughts …

  1. With 100 levels for every skill player should specialize in them very early. I believe it’s enough to grant one specialization slot every 5 character levels starting with 3-d. So the last one opens at character level 28.
  2. First and even second core node points should also be granted early on to provide an opportunity to build your build’s basis. So the core node skill points are granted at skill’s level 5, 20 and 35.
  3. It is hard to lose 100 skill levels, even 50… even 20! So specialization slot should grant leveling ability for a skill - that’s it. Skill progression should be saved even upon removing from specialization slot.
  4. Any skills effects should work only while skill is on the panel, including passive bonuses. But if skill is NOT in specialization slot than it uses its basic functionality. Even if it’s leveled to a MAX.
  5. Small note: you gain your last specialization slot at level 28. Or may even recpec. That means that with 100 levels for every skill you’ll be leveling them even after reached character’s level cap. Progression! :slight_smile:

Enjoy! :slight_smile:

P.S. I don’t know why but my creativeness increases with this in the background. Just listening this all the time now :crazy_face:

1 Like

It’s a nice idea, but I suspect, that it might not be particularly practical (plus every node would need to be unique due to #4). Think of it, GGG took ~3 months to come up with 180 passive nodes for Delirium league. Not 240 nodes per skill…

Edit: I must admit, I did only kinda skim your post, sorry, it’s a good post, lots of space, paragraphs, etc…

First: I’m sure it’s not easy. Even to finish my doc is pretty hard for me :slight_smile:

Second: it obviously would be an absurd to show all this in the same way - at once. My doc is a scheme, logic - not UI design. But we can imagine something more practical, right? Like …

Watching skill specializations shows you first level of detail - Main skill node with shortened description, and Core nodes around, also with shortened description. Small nodes are also shown but darkened, probably without links, with very short description that appears ONLY under cursor.

Click on the Skill or any Core node and you’ll get next level of detail. Interface scales this part of tree, focuses on it, lights up small nodes and their links, some description, shows more detailed description for a Skill / Core node.

Third level of detail: shows details for every node you pointing with cursor, perhaps on the right side of screen?

And the main approach: keep your skill’s core idea in your head and every way to level it should be right. Player may lost in your tree but his every step should be right. How do I follow this rule?

Well …

Focus: Fireball - is a fast casting missile spell with some AoE and good potential to crit and ignite.
Implementation: every node field (around core nodes) has something that improves those qualities. Choose ignition in a main field, than improve it’s damage with shock effects from Blast wave and stack it faster with Flamethrower.

Yeah, I know, nothing is ideal. And some combinations may be more or less effective. But they are possible and viable I believe. This is the most important. In my opinion of course.

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.