Feedback šŸ”‹ Skill leveling is a nice system, but also underdeveloped in D4 - what can be done about it?

PC General Discussion
1
Have you ever wondered whether it is worth to invest more points in a skill? click on it

I like that skills can be leveled up. Adding points with equipment is a nice system. However, after the introduction of Tempering affixes it feels a bit underdeveloped, with mostly using either cooldown reduction or adding some percents to damage. While cooldown is literally cool to have and clear to read, the majority increasing only damage is just boring.

Damage varies a lot due to buffs and upgrades from many additional sources (Equipment, Paragon, Aspects). Since there is no ā€˜goal valueā€™, and since other skills all have different base values, it is difficult to determine how big an upgrade really is in comparison. Furthermore, the out of combat talent tree tooltips almost never align with what we see on screen. In the end, all this leads players to rely on a simple ā€˜more is betterā€™ formular, and to ignore the obscure numbers.

This is no problem per se, but if a player does not understand the impact behind a system, then there is no imagination how progressing it will change a playerā€™s experience, and ultimately no excitement in explorating the system with investments. In short: it is not fun.

In recent interviews, the developers have indicated on an overhaul of the skill tree. From what I have heard, it seems to implement some transformative things in the tree for each skill. That sounds good at first. And while they are at it, I want to provide feedback for another related issue I like to see improved: skill leveling.

In this post I suggest some additional properties/stats to increase for each skill point a player has invested through various sources. My intention was to find properties that are more interesting than just percents of base damage, and therefore much clearer to the player within the first few points. Clear in communicating that a level 20 skill will feel different compared to level 1 or 5, making a player curious to try and play it.

However, the suggested properties shouldnt increase stats easily to the max limit; just a little is enough to make the power growth noticable, and also to free some equipment affix slots, which is very helpful for customization.


Optional skill properties to increase with each point
+ Cooldown Reduction click on it

Not new, but this is what I am talking about, so I put it first. Cooldown Reduction not only gets powerful skills closer to a spamy state. The difference in seconds is clear, it is fun to imagine the upgraded effect in advance, and it feels good to invest points in it. Thats all I ask for.

For some reason, improving the cooldown via points is not enabled for all cooldown skills in Diablo 4, but mostly reserved for the Defensive and Ultimate branches. This is another reason why I want to mention it here; to suggest a removal of this outdated limitation.

+ Appearance / Transformation click on it

Improved visuals are ideal to display how far a characterā€™s progression in power is. Skills progress as well, and they tend to look very basic without any buffs, so why not let them show their power level this way.

  • Imagine Necromancer skeletons transforming into a stronger looking version with more and more armor parts being added.
  • Or Lightning Spear (Sorcerer) grows into what could look a bit like an asian dragon.
  • Or Druid wolves mutating into werwolves and even bigger creatures.

All of this not by equipping some unique item oder aspect, but by investing skill points.

Actually, this feature exists to a small extent for (some) skills, but it is limited to a single visual change that is also very restrained in its intensity. I would love to see this more developed and visible, with many incremental steps from level 1-30+.

+ Number of effects / units click on it

Related to appearance, but it hits different for me. It is a feature that I dearly miss from Diablo 2: Increasing the number of skill effects with skill points. This can be very appealing to explore, even exclusively for the visual addition, as the sheer amount (minions, arrows, charged bolts, hydra heads) feels like a symbol of strength. And, of course, it looks very cool at high skill levels.

+ Size click on it

Also related to appearance. Very impressive are size upgrades as they expand a skill effect to be more screen filling. Personally, I always have fun to increase these properties with Tempering, even if it is just to see how it looks.

+ Duration click on it

Provides more comfortable handling of buffs and debuffs. For example, this property can increase the duration of war crys, but also the duration of Crowd Control, Vulnerable, Damage skills, Bleeding, Curses, etc.

+ Lucky Hit click on it

Granted, Lucky Hitā€™s effectiveness is a bit harder to grasp as each skill has a different base value. Furthermore, the stat is rare on items and these appearances tend to be small, so most players ignore Lucky Hit entirely.

My reason to suggest it here is the very high cap of 200% Lucky Hit. Thats why I think it couldnt harm to let skill points contribute a bit.

+ Attack / Casting Speed click on it

Attack speed adds a lot to the gameplay feeling. This would also only benefit the specific skill, e.g. Pulverize will get Pulverize Attack Speed.

+ Chance to cast twice / deal double damage click on it

The chance to cast twice / deal double damage is quite popular for offensive skills since it was introduced. It is strong and, whatever number the damage result may be, always clear about how big its effect to the skill is.Well, it would be even clearer, if the character stats list was up to date with showing double cast/damage values. Currently the list is very incomplete and seemingly random in what affixes are shown.

+ Any skill modificator stat click on it

Modificators are the Y-node options behind a skill a player can choose from. May it be about Fortify, Healing, Resource, Crowd Control chance, Damage reduction, different thresholds or prerequisites to trigger another effect; These things could be included in the skill levels.



Skill Examples

Here are a few examples with Druid skills where some suggestions from above are illustrated. You can see how the properties/stats grow along with the skill level from left to right. Note that values for cooldown are not exact, level 20 is not meant to be the maximum, and the camera icon :camera_flash: is for visual transformations without other effects.

Landslide lvl ā†’ 1 2 3 5 10 20
number of pillars 4 4 4 5 6 8
chance to cast twice 0.0 1.5 3.0 6.0 13.5 28.5
attack speed 0 2 4 8 18 38
pillar size % 100 103 106 112 127 157
lucky hit 20.0 20.6 21.2 22.4 25.4 31.4
:camera_flash: diamond skin transformation % 0 4 8 16 36 76
Tornado lvl ā†’ 1 2 3 5 10 20
faster tornados % 0 5 10 20 45 95
enemy tracing % 0 3 6 12 27 57
size % 0 2 4 8 18 38
chance to cast twice 0.0 1.5 3.0 6.0 13.5 28.5
lucky hit 8.0 8.2 8.5 9.0 10.2 12.6
Lightning Storm lvl ā†’ 1 2 3 5 10 20
maximum strikes 5 5 5 6 7 10
strike accuracy % 0 3 6 12 27 57
dancing bolt spawn % 0 2 4 8 18 38
chance for double damage 0.0 1.5 3.0 6.0 13.5 28.5
lucky hit 15.0 15.4 15.9 16.8 19.1 23.5
:camera_flash: lightning thickness % 0 4 8 16 36 76
Boulder lvl ā†’ 1 2 3 5 10 20
cooldown in seconds 10.0 9.7 9.4 8.8 7.3 4.3
rock size % 100 103 106 112 127 157
rock speed % 100 103 106 112 127 157
lucky hit 5.0 5.2 5.3 5.6 6.3 7.8
:camera_flash: magma rock transformation % 0 4 8 16 36 76
Wolves lvl ā†’ 1 2 3 5 10 20
number of wolves 1 1 1 2 4 8
cooldown in seconds 11.0 10.7 10.3 9.7 8.0 4.7
movement speed % 100 104 108 116 136 176
cleave damage chance % 0 3 6 12 27 57
wolve attack speed 0 2 4 8 18 38
chance for double damage % 0.0 1.5 3.0 6.0 13.5 28.5
lucky hit 50.0 51.5 53.0 56.0 63.5 78.5
:camera_flash: werwolve transformation % 0 4 8 16 36 76
Ravens lvl ā†’ 1 2 3 5 10 20
number of ravens 1 1 2 3 5 10
attack rate in seconds 5.0 5.0 4.9 4.8 4.5 4.0
cooldown in seconds 15.0 14.6 14.1 13.2 10.9 6.5
crit bonus 8.0 8.8 9.6 11.2 15.2 23.2
active attack size % 100 103 106 112 127 157
chance for double damage % 0.0 1.5 3.0 6.0 13.5 28.5
lucky hit 45.0 46.4 47.7 50.4 57.1 70.7
Cataclysm lvl ā†’ 1 2 3 5 10 20
cooldown in seconds 60.0 58.2 56.4 52.8 43.8 25.8
duration 8.0 8.4 8.8 9.6 11.6 15.6
tornado frequency % 100 103 106 112 127 157
chance for double damage % 0.0 1.5 3.0 6.0 13.5 28.5
lucky hit 62.0 63.9 65.7 69.4 78.7 97.3
:camera_flash: rain density % 0 4 8 16 36 76
Grizzly Rage lvl ā†’ 1 2 3 5 10 20
cooldown in seconds 50.0 48.5 47.0 44.0 36.5 21.5
maximum duration 10.0 10.4 10.8 11.6 13.6 17.6
damage reduction 20.0 20.5 21.0 22.0 24.5 29.5
fortify per second 8.0 8.8 9.6 11.2 15.2 23.2
:camera_flash: minotaur transformation % 0 4 8 16 36 76
13 Likes
2

Hey, plus one heart for getting that formatted on the forums. That alone is impressive!

Before I checked out guides and started reading the forums, I definitely did not get why adding plus damage to skills in the paragon board or maxing out skill ranks didnā€™t seem to matter at all. Iā€™d boost a skill all the way and it still wasnā€™t great. My problem was I wasnā€™t shooting for 17 different multipliers. I can see how your examples actually make points meaningful.

That said, they canā€™t even get what we already have to work right sometimes. Just one ring was too much, let alone all these options.

I would love for them to get everything thatā€™s already in the game working first, but you have some fun ideas there!!

3 Likes
3

A long time ago, before the game launched, skills were supposed to work like that. Put more point into a skill to gain more flat damage and ability. I would like that.

Now we start with like 12 points but it doesnā€™t really matter at low level because skill damage is based on weapon damage. Canā€™t wait to see what they come up with in the future.

1 Like
4

Most of us want to see that kind of improvement but it is big no for blizzard. 2 years of experience says there will no change for their kindergarden level skill tree and passive tree (paragon).

Sorry for my low expactations from blizzard. It is sad. Oh god i miss d2.

3 Likes
5

Very impressive :slight_smile:

Believed it or not I still would like to see changes to the Pit though. In other words fix what is bad rather than delve into big improvement/s (although yes, this would work too)

P.S. donā€™t forget the flexibility one of the S9 features can bring to the game (either in the form of Arcanas or Infusion)

1 Like
6

In season 6, ultimates gained a bonus when reaching level 5. Spiritborn has some abilities like this. I thought this would be added each season, changing the abilities.

It would be interesting to have these points at 10ā€¦15 levels. It would give a great power when reaching a level, which would value itemization and the end game.

It would be a way to make abilities that are strong at the beginning more interesting in the end game without unbalancing the game. Basic abilities could become powerful at level 30. With more shots, interactions with other abilities, etc. It would be really cool if firebolt at level 30 summons 2 hydras that shoot firebolts. It would be really cool.

  • I would like these interactions instead of just more multiplicative damage at each level
2 Likes
7

This looks like a very fitting system for D4, but would you just display all the information in a tooltip?

1 Like
8

I had something similar to this before, where the skill points were more based around whether you just wanted the skill, you wanted it to proc something, or you wanted to go all in.

Example: Ring of Fire
1 pt: deals X damage in a circular area for Y seconds
(if I just want the skill, I stop here)
2 pts: enemies damaged by this skill burn for Z seconds
(now if I run a pretty general dot build or have something proc off of burning, this is a good place to stop)
3 pts: Base damage increased by W%
4 pts: Area of effect increased by Q%
5 pts: Burning damage caused by this skill increased by T%
(3-5 points are based around how important this skill to my build; 3-4 is a good damage dealer. 5 if I am super into dotting as many foes as possible as hard as possible.)

Then I can still have modifiers, as it is now or like D3ā€™s runes.

Ring of Ice: Damage becomes cold; burning replaced with frostbite.
Chilling Ring of Ice: Frostbite applied by this skill also chills.

Channel: This is now a channeled skill, with cost/second equal to C.
Growing Channel: Area of effect increases while channeled to a maximum of D% after E seconds.

Explosive Fire: Ring of Fire periodically spawns small explosions that can strike nearby targets.
Exploding Explosive Fire: Ring of Fireā€™s explosions spawn shrapnel that travels a short distance and then explodes again.

etc.

In any case, making it less about numbers and more about what it does is fun for me, at least.

9

So where do Aspects fit into this model? Looking at Wolves the Aspect that grants an extra companion now seems like it should just add extra skill levels since it would be ineffective at higher levels of Wolves.

1 Like
10

Aspects should give multipliers.

11

I canā€™t back that one. Thatā€™s the most boring system in the universe.

12

Iā€™m pretty sure some aspects can get sacrificed for a greater good :slight_smile:

However (again), Iā€™m more hopeful for the potential of the skill-crafting/altering mechanism/s (especially Arcana)

Imagine if there are gems/sockets/whatever post S9 where you just ā€œslapā€ it on a skill and it does stuff:

  • 5% increased Lucky hit chance and 25% radius
  • 25% increased damage and AoE but adds (or increases) CD of 4 seconds
  • Adds a second charge and reduces CD by 3 sec

e.t.c.

Iā€™d most definitely love to see that at work (for example do I put the first one on a Meteor or LB, or say if Barb do I add that to DS or EQ ?, do I put the 2nd one on a Mastery skill or an Ult, do I put the 3rd one on a defensive skill or mobility one ?, e.t.c.),

Point being ? - I just like that potential the system of ā€œArcanasā€ long-term can do ā€¦ :thinking:

13

Skill modularity is a weird space because thereā€™s a utility function to it where after a certain point every element of modularity actually decreases the diversity. For example in your options the second charge is the best and the increase of CD is the worst. I guarantee you that will be gamed so hard that it wonā€™t matter what the other options are. Thatā€™s what makes it tough.

A flat reduction to CD will always be a best option too; it should almost have a drawback like a reduced amount of power.

14

Iā€™m not saying no but the thing is you define/design a system (expand the imagination with a mechanism of ā€œfairnessā€ on top of that)

For example - can contain limitations, based on ā€œcurrencyā€ (i.e. some sort of energy), can have limit of total amount, e.t.c.

Limitations ? - each of those can have a pre-assigned skill type (or set of limited skill types) they can alter
Currency ? - think of Spirit from PoE2 (letā€™s call it Atunement for ex.), the one that gives 2 charges and CDR can simply cost more of that
Limitation of totals ? - can equip/alter-skills only 2 at a time

See the point ?, yes, I get your ā€œbut A is better than Bā€ point, but I wasnā€™t talking about the ā€œmarketā€, I was just talking about the featureā€™s potential (not just at S9 but long-term)

15

I understood what you meant and agree with you but I guess I am trying to caution against the inherent blindness to other people we have when we implement ideas into things. We never properly imagine someone smarter than us or dumber than us (because we are us, a hard limit) when designing things so while the ideas may flow like water youā€™ll be tailoring and scaling back a lot if you miscalculate how effectively someone youā€™ve never heard of can outthink you.

When I did create things for other people in the past as often as I was shocked how dumb people could be I was always blown away by how much more clever people could be than I am and how many times someone made something that wasnā€™t made for a task fit really, really well. So your dimensions are fine but kind of like PoE2 the oversights there in those builds were severe during .1.0 I mean I think there was, unbugged, like a way to create thousands of reflective ice shards that caused burning and burst from anything they killed but also split into 6 or something. Backroom testing will never find that due to the bias of knowing how itā€™s supposed to work.

2 Likes
16

Basically go the Last Epoch route?

17

The aspects that are practically multipliers. Nothing will change

19

I think this is a fantastic idea but the ptr should be like 2-3 weeks or even a full season to flush out all the bugs for this type of overall.

Unless you do it one season at a time and one part of the tree at a time. Start with basic, then core, then on down the treeā€¦. Alternate this with powers/gens and then when all are done put it into eternal. This will create different metas for different seasons too.

1 Like