After loving both Diablo 4 beta weekends, I booted up Diablo 3 again after 10 years
In short, it was great!
I had so much fun, and after 20 hours of play I bought Reaper of Souls too. Diablo 3 has come a long way since I played it on launch 10 years ago. Back when Diablo 3 first came out, I played the hell out of it, and was even able to beat Inferno Diablo on Softcore about 12 days post-launch and before they nerfed Inferno. However, it didn’t really click the way that Diablo 2 did for me, and eventually I stopped playing after a few months. For 10 years.
After spending a bit of time with the game and thinking about it, I like how the game has changed and think it’s in a great spot. Kudos to the developers who kept it going over the years and iterating on the original design. So far, I completed the Altar of Rites, Season 28 Journey, and got a Wizard, Necro, and Crusader decked out in their complete sets and was able to clear up to GR140 solo.
I’d give Diablo 3 a solid 9/10 as it stands today. However, in my opinion, not everything is as good as it could be. There are some design choices in the game which are questionable and still some annoyances with the gameplay, UI, and controls that only really stood out to me until I put in about 40 more hours.
Obviously, we all really like this game, otherwise we wouldn’t be discussing it on this forum, so instead of going over what I liked, which are numerous and uncontroversial, I’m going to go over the things that I thought are still a problem in the game as well as suggestions to fix them. These problems may or may not be that pressing to the community at large, and my proposed solutions might not be everyone’s cup of tea, but I think they’re fair, and backed up with real reasons.
Problem 1: HUGE power imbalances that hinder cooperative play and stifle build diversity and choice.
Power imbalance is too big between the low paragon players and the high paragon players. It’s also too big between players who have a “good” build vs players who have a “bad” build. I put good and bad in quotes because it all comes down to the damage multipliers. Players with “bad” builds don’t have any gameplay-centered, logical reason why their build is bad. The only reason their build is “bad” is because the skills they chose to use just don’t have enough items that give +damage. These builds are essentially chosen for you before you even play because the itemization has too many +damage modifiers on only a select number of skills, leaving the rest in the dust and irrelevant past Torment 5. There shouldn’t be so many “obviously good because of the insane damage buffs” choices. If you were using a guide beforehand, maybe you could see this coming, but what of the new player who wants a fresh experience and levels up using skills he or she enjoys, but then later discovers that there aren’t any items that buff that skill into usable territory in late game?
Oftentimes, public GR games will have players who want to play at GR120+ in the same game as players who can only play at GR90, and the GR90 players are essentially doing zero damage. This leads to a frustrating gameplay experience when the higher level players have to carry the lower levels, and the lower levels feel useless.
Set item bonuses are too strong, and (most) legendaries are too weak. Some of the sets aren’t even in balance with each other, with certain sets being obviously better than others but not even necessarily harder to obtain. You can see this pretty easily on the season leaderboards.
At the endgame, the damage numbers that are popping out are utterly ridiculous. 100 BILLION? 15 TRILLION? I mean, this is crazy. What actual purpose does that have other than filling up my screen with ridiculous numbers? When set items need to have affixes like “adds 10,000% damage” as a modifier, that’s pretty telling that something went off the rails in balance and power creep.
With the way things are, at T16 and especially at GR90+ I can have a build with one skill that does decent damage and another that does basically zero damage. All because skill A can be potentially boosted 1000% with the right items, and skill B can’t be boosted that high because those items don’t exist for it. The beginning of the game on a fresh character on Normal difficulty feels great. You can use resource generators and resource spenders, and they all do relative amounts of non-trivial damage that make sense given the context of their utility. However, after a certain point, the “monster health” and “player damage” curves become unhinged, a player’s skills are no longer balanced relative to each other, and the entire feel of the game changes.
The current meta is basically a race to find the most multiplicative sources of damage, stack crit, and ignore everything else. You can get most of the way there with simple math theorycrafting rather than actually playing the game. Like, it’s pretty obvious that meteor is going to be an OP skill when there’s at least 4 different items that boost its power by multiple hundreds of percent. Is that by design? Are we supposed to pick meteor and ignore most of the other skills that don’t get that same kind of power boost? Why even bother having other skills on a class if they’re only relevant up to T5? That doesn’t seem like good design to me. Shouldn’t all skills be more or less viable and have usefulness even at endgame?
Speaking of which, there are way too many difficulty levels, and it’s very cluttered. It’s indicative of bad balance design when you have to make the player individually fine tune the balance for themselves among 16(!) difficulty levels because relative power levels are so out of whack. Player-driven balance tuning should be left to custom events like Greater Rifts.
Potential fixes for problem 1:
Solution 1a:
There are two general ways in which I can imagine you can fix this. One way would be to eliminate most, if not all, of the +damage modifiers on all of the legendary affixes and set bonuses. One of the main contributors to power imbalance is that the itemization is not equal with respect to skills. Some skills get multiple items with heaps of +damage. Some skills don’t, and are then useless. Then you can rebalance the difficulty levels to be something more reasonable (4 difficulty levels), such as Normal, Nightmare (higher chance of legendaries), Hell (possible to drop ancient items), and Torment (possible to drop primals).
I have a feeling that this could be too much work to program and too disruptive to the status quo, so in lieu of that, you can do this instead -
Solution 1b:
Add more +damage affixes onto legendaries for skills that don’t have enough. Why does the Meteor skill have 4 items that give it 300%+ damage each, while the Blizzard skill has….zero?
Add more legendary affixes that buff the currently neglected skills. But does that mean we need to make even more legendaries and dilute the item pool? No. There are more than enough useless legendary items already in the game that are basically auto-salvage because there is no point to their existence due to very weak or irrelevant affixes. Change those affixes to buff the currently useless skills so that we have more viable skills to choose from.
Either method could achieve this goal. I would prefer the first way, but I understand if that’s too hard. Moving along, here are other potential fixes that could help -
Solution 1c:
Keep Paragon levels in check. Cap Paragon levels at 2000, and introduce a very harsh leveling curve so that it takes longer and longer to level up past level 1500. Similar to the EXP curve in OG Diablo 2. Let people’s skill and accomplishments be determined by their build choice, itemization, skill choice, and gameplay - not by how much main stat they grinded.
Solution 1d:
Buff primary/signature/generator skills by letting players pick more than one rune after level 70 (maybe 2 or 3). This would make some novel skill combinations possible and give people more reason to use them even after you don’t care about resource generation anymore. Some of the most fun I’ve had with the items in Season 28 are the ones that give access to multiple runes. This theme should be expanded upon.
Solution 1e:
Items that drastically boost only multiple target DPS and not single target DPS, and vice versa, should be adjusted. Nilfur’s Boast is a good item because the “+900% if Meteor hits 3 enemies or less” legendary affix keeps the skill relevant for both multi and single target DPS. However, items like Bloodtide Blade only boost multi-target DPS for Death Nova, so you get into situations where a season 28 Death Nova Necro is great at mowing down mobs but utterly terrible at killing the boss in a Greater Rift. This feels bad to play. Players should be able to excel at either single or multiple target DPS, or be decent at both, but never be glaringly terrible at one of them.
Solution 1f:
Do a general rebalance of some of the least used skill/runes. I’m sure you have data on which skill/runes players are picking and know what the least popular ones are.
Problem 2: There are too many clicks required to do things. It’s annoying to have to repeat the same recipes over and over again, or upgrade jewels, or gamble.
Solution 2:
-
Let players upgrade all gems to their highest tier with available materials in one click at Shen.
-
Let players gamble 30 items at once and auto-salvage the magics/rares.
-
Let players pay more to enchant with 10 options at the Mystic.
-
Make a “salvage all non-ancient legendaries” button at Haedrig.
-
Make a Kanai’s Cube recipe to convert gems 50 at a time, or convert crafting mats 1000 at a time.
-
Allow players to gamble or craft as if they were other classes. In other words, if I’m playing a Wizard and want to craft Templar gear, let me tell Haedrig that I want items with STR main stat instead of INT.
-
Give pets an option to auto-salvage magics, rares, and/or non-ancient legendaries. This option in the altar of rites is amazing, and I feel sorry for everyone else who played these past 27 seasons without it.
Problem 3: Inventory management is a hassle, especially when you’re playing multiple characters.
Solution 3:
- Well, the obvious one - add more stash tabs. I’d even pay for more as DLC. How are we supposed to manage multiple characters with multiple sets given the limited stash space?
- Add a gem tab in the character inventory that exclusively holds gems. The upgrading process takes up way too much space.
- Add a lock toggle on items you wish to keep and not accidentally salvage.
- Let us search for legendary gems in the inventory.
- Let us search for legendary gems even if they are socketed in something.
- Let us search for specific stats or affixes on items in our stash.
- If stash search has a hit, highlight not only the item itself, but the tab that it is in.
- Armory/Wardrobe should also save Paragon setups as well.
- Give players more ways to increase their bloodshard cap, such as doing certain quests or getting certain achievements.
Problem 4: Too much lag in multiplayer games, especially when there are multiple Tal Rasha Wizards.
Solution 4:
- Get rid of the “area damage” stat. It’s too esoteric in concept and contributes a lot to lag. At first glance, it’s hard for most players to intuitively grasp what the heck area damage even is, how it works, and why it’s important.
Problem 5: Bounties are tedious, Ubers are a joke, and there’s really nothing significant to do besides grinding Greater Rifts and the occasional Echoing Nightmares.
Solution 5:
-
Keep the 2X bounties from season 28. Let solo players only have to do 4 out of the 5 instead of all 5. Get rid of the tedious bounty quests like “kill all enemies on level 2 of some random dungeon”.
-
Have extra strong “boss bounty” quests with nice rewards that pop up after you turn in everything to Tyrael. These are extra hard boss fights that actually require players to team up during multiplayer bounty runs.
-
Buff the Ubers so that they’re actually a challenge. This problem goes hand in hand with the overall power imbalance/power creep problem I talked about in the beginning. I imagine that there used to be a time when the Ubers were actually difficult? But then power creep affected the game and now they’re a joke. Make them hard again, and buff their rewards respectively.
-
Create new rift types that are challenging and different, like a boss rush rift.
-
Have the option to receive GR rewards in the form of a gift box, like how you get bounty rewards. It would help streamline GR farming runs and save inventory space.
Problem 6: Legendary gems are cool, but have issues. Only a small amount of them are actually useful.
Solution 6:
-
Either remove all damage buffs and make them all utility gems, or keep the damage gems and buff the useless gems.
-
Let Ramaladni’s Gift be usable on jewelry too. They all NEED a socket because of the existence of legendary games, so let’s make it easier to socket them.
Problem 7: Followers are nice to have, but are basically just glorified stat sticks and are hard to manage with multiple characters.
Solution 7:
-
Have followers be account-wide, so we only really have to gear them up once. It’s frustrating having to either transfer gear over or craft the same setup each time on each different character.
-
Give followers an armory/wardrobe function as well.
-
Give followers more survivability so that the “followers cannot die” relic is not the only rational choice to make.
-
Give followers more DPS so that their damage output is not just an afterthought.
-
Let players select from certain AI patterns for their followers to execute.
-
Let solo players use all 3 followers at the same time. You might as well. It’s possible since the Asheara’s Set exists. Let them do combo attacks with each other like in Chrono Trigger that would be awesome.
Problem 8: Convention of Elements is a stupidly designed ring, and its gimmick is annoying to work around. It’s also doubly annoying that it’s so powerful that it’s best in slot for many builds. Windows of opportunity are more engaging when it’s based around the context of the environment or enemies, instead of a self-timer.
Solution 8: If the object of the design is to encourage players to use multiple elements, then do this instead:
- Using an elemental skill on an enemy increases its resistance to that element by 2% per hit, but also decreases its resistance to the other elements by 5%. Capped at 100%.
- Or maybe - Casting an elemental skill reduces a player’s resistances by 2% per cast for that element and increases their damage for all other elements by 2% for 20 seconds. Stacks up to 50.
- Or, just keep it the way it is, but make other rings strong so that there are other viable options for best in slot rings.
Problem 9: All sets aren’t created equal. Some sets are obviously better than others. Is that on purpose? The stronger sets don’t seem harder to obtain, so there is no strength/rarity curve either.
Solution 9:
-
Rebalance all current class sets so that they can compete with each other using standardized parameters. For example -
-
Haedrig’s Gift starter sets - Can reliably clear GR 70 under 10 minutes wearing only the set and rare items and zero paragon. There should be a variability of no more than +/- 5 GR tiers of each other.
-
All other class sets - Can reliably clear GR90 under 15 minutes wearing only the set and rare items and zero paragon. There should be a variability of no more than +/- 5 GR tiers of each other.
-
There should be more than enough data via the leaderboards about which sets are under/overperforming and why.
Problem 10: Current set design is too one-dimensional, without much room for flexibility, and in most cases are overpowered compared to legendaries. This is an overly rigid design that stifles creativity and build diversity. It is a top-down design that doesn’t give players much freedom.
Solution 10a:
-
Add another 2-item set that incorporates legendary items, similar in spirit to the Legacy of Nightmares set. For example, a set consisting of a Wizard Hat and Source. You get the 2 set bonus for having the 2 base items. You get a 4 set bonus for equipping two additional legendary items. You get a 6 set bonus for equipping another two legendary items. Maybe add an additional bonus if all added legendaries are Primal.
-
2-set bonus: All signature spells have 25% faster cast rate. Primal:50%
-
2-set 2 legendary bonus: For each second a spell is channeled, increase max energy by 1% for a duration of 5 minutes, max 300%. Primal: 2% for 10 minutes max 600%.
-
2-set 4 legendary bonus: Familiar now gains all runes and is always active even if not on the skill bar. Primal: Familiar morphs into a mini-Archon
Necromancer version -
-
Necromancer 2-set bonus: +10 Essence on crit, and +10% critical hit chance on Primary Skills. Primal: Secondary skills cost 2X essence but also do 2.5X damage.
-
2-set 2 legendary bonus: +2% maximum essence for every active summoned minion. If no minions are active, gain +10/sec passive essence regen. Primal: double the bonus
-
2-set 4 legendary bonus: Blood Rush gains all runes. Primal: Blood Rush cooldown is 1 second and detonates a Blood Nova on arrival.
This is of course just a concept and would have to be worked into a greater rebalancing. These set bonuses are designed to be very general in nature so that they can be used creatively in a variety of builds. However, this still doesn’t even come close to the power of Legacy of Nightmares as it is now because LoN grants you +9750% damage and my concept doesn’t (damage is king). This is just another testament to why +damage is so important that not having it makes you useless.
Solution 10b:
Also, buff the current legendaries that are currently auto-salvage tier and give their primal versions extra juice to make them interesting. For example, make sure legendary effects’ damage scales well into the endgame. Many legendaries suck because while their effects are cool and useful in the beginning, there is no way for them to scale with player power, and their damage output falls off. Let legendary effects have a small proc coefficient. If a legendary effect spawns a minion, like the Maximus or Skycutter sword, let that minion stay active forever until it dies. If a legendary power affects a skill but doesn’t add damage, give it a damage buff so that it falls in line with the other good legendaries like Mempo of Twilight. If a legendary has an utterly useless affix like Gladiator Gauntlets, scrap the affix completely and at the very least change it to buff one of the low tier skills.
However, the problem with power creep isn’t solved - if you buff legendaries, you’re inevitably going to buff sets too because you can use legendaries with sets, and those can still benefit from both set and legendary bonuses. I suspect that this is the dynamic that’s one of the root causes of this 10-year journey of insane exponential power creep, which is why I feel like all the damage percentages should be cut across the board to begin with. In lieu of that, maybe implement a way in which legendary power is lowered if used with a set, but that also seems like an inelegant solution. Maybe you just have to make class sets be 8-piece, but that also seems inelegant and restrictive (even though the concept of set items are restrictive by design)
Solution 10c:
To encourage gear diversity and flexibility, give crafted sets up to 6 items in the set, but keep the max set bonus at 4-set to encourage gear diversity and flexibility.That way I’m stuck not having to use bracers + waist for Guardian’s or pants and belt for Crimson’s 90% of the time and have the option to maybe make room for different items.
Solution 10d:
Alternatively, set bonuses could be more general and less specific. Instead of adding 10,000% damage to specific skills so that those specific skills are the only ones under consideration, make that 10,000% damage buff apply to a category of skill instead. For example, Season 28 Aegis of Valor boosts Heaven’s Fury and Fist of the Heavens ONLY. Instead, Have it boost both Secondary Skills and Conviction Skills, and have all Secondary Skills generate wrath on hit. That way, it operates more or less the same as it is now with HF and FoH, but you could also experiment with different skill combinations as well.
Problem 11: Kanai’s Cube legendary affix storage is a great way to add flexibility and power to a build, but sometimes it’s not enough flexibility.
Solution 11:
New gear idea and Cube recipe - 1 rare ring or amulet + 3 gems of any type + any legendary item = Legendary ring with that item’s legendary affix + the original ring’s rare stats. Using higher quality gems increases chances of better rolls on the legendary affix and the original stats. This provides more options to promote build diversity. This also provides more “best in slot” alternatives to Convention of Elements and Squirt’s Necklace.
Problem 12: The UI is decent, but could use more Quality of Life upgrades to give more relevant information to the player.
Solution 12:
-
Give players the option to see the actual numbers of their life and resources on the UI bar.
-
Give players the option to clearly see how much damage each of their skills do based on their current build - put the numbers in the skill description themselves.
-
Give players the option to clearly see what their current cooldown timers are based on their current build. Skill descriptions are based on naked character numbers, and don’t take into consideration equipped gear.
-
Give players the option to see all their crafting materials from the inventory screen without having to make an extra click.
-
Give players the option to see what’s in their Kanai’s Cube without having to go back to town.
-
Indicate Elite mobs on the minimap. You already do it for quest monsters and bosses. This would be very useful. Just make it a blue skull for champions and a yellow skull for uniques.
Problem 13: Many skill/runes are too weak, and are not even considered in “serious” builds. Aside from just buffing their damage, there should be a way to make them more interesting and create new experiences even for 10-year veterans of Diablo 3. I really enjoyed the gear that enables multi-rune skills, and I think that should be expanded upon.
Solution 13a:
Droppable skill runes as items. A select number of runes can drop as items. These runes go into a separate inventory tab. Runes can be augmented onto non-socketable ancient gear (shoulders, bracers, gloves, belts, boots), and will allow you to use both the rune selected on your skill bar + the rune you augmented into your gear at the same time. You can choose either the stat augment or the rune augment. Primal gear can augment both at the same time. There are an average of about 125 runes per character, so they will have to be chosen carefully so that we have a manageable number. Choosing the least popular runes to add as augments to gear would be a creative way to buff them and make them relevant. As I mentioned before, I think the items that enable multiple runes on a skill are very interesting, so I would like to expand on that theme.
Solution 13b:
Cross-class skills - If you use an item that was augmented with a rune from a different class, you can use that skill/rune yourself. This obviously has a lot of balance and logistical implications, but the scope can be limited in order to reduce the amount of chaos it will cause, to reduce the amount of work it will take to implement, and to preserve class identity. For example, you can only use one cross class skill at a time. Only certain skills can be used cross class. There are a lot of skill/runes that are impossible to use on other characters, like DH bow skills on a Wizard. Certain skills like Command Skeleton will remain Necromancer only because a skill like that is too integral to a Necromancer’s identity. Keep the scope small at first, with mostly utility skills and a few offensive/defensive skills, and adjust as needed. This will provide a huge wave of theorycrafting for Diablo 3 players that will provoke lots of interest from people who may be bored of the game.
Problem 14: Endless grinding for Paragon and +main stat passives are boring. Grinding is fun, but making players do it forever in order to brute force their build to work is not. There needs to be more than just grinding for gradual increases in damage, which is what main stat gives you.
Solution 14: Paragon specific endgame passives that cost 100/200/300 paragon points each to activate.
Cap Paragon levels at 2000 with an increasingly harsh EXP curve that ramps up after 1500, so that you’re technically still growing your paragon, but you do it much more slowly.
Here are some examples of endgame paragon passives that are more interesting, in my opinion, than simply +100 main stat:
Wizard: Activate all 3 armors when using one. If you don’t have the other armors on your skill bar, they will just use the default armor, similar to how Belt of the Trove uses the default Bombardment skill even if you don’t have it on your skill bar. But what if you want to use the runed skills? Then put them on your active bar. But what if you don’t have space? Then you can use the augmentable runes system to add those runes to your gear so that they will be active.
Wizard: For every 1% of max life in raw HP you lose while using Energy Armor, your next cast of Black Hole will be 1% more damage/duration/area. An arcane chain will also extend between your Energy Armor and the Black Hole for its duration for any Black Hole buffed beyond 200%. Any other players or minions that use an arcane skill while the buffed black hole is up will also generate an arcane chain between them and the black hole. 500% cap.
Wizard: Actual Magic Weapon - Magic Weapon turns into a togglable skill that turns your equipped weapon into a giant, glowing version of itself. Your Signature skills change to swing this melee weapon on click, but also proc your Signature skill on hit. Perfect for players who want to try a battle mage.
Wizard: Mirror Images last forever, but each active mirror image costs 1% life and 2 energy per second. Max 4. Mirror Images have a 10% proc coefficient
Wizard: Signature spells now cost 10 energy to cast, but the damage is doubled and it shoots double the amount of projectiles/hits/bounces
Necromancer: Casting a curse also casts Devour and Corpse Explosion and Revive at the same time.
Necromancer: Casting Army of the Dead on a target also causes all active minions to run at the target and blow up.
Necromancer: Add +5 charges to Bone Spirit and double its speed. If no charges are available, Bone Spirit can be cast for 200 essence.
Necomancer: Every bone skill you have on your skill bar adds +4 max skeletons.
Necromancer: Every 10,000 essence you spend spawns an additional golem of a different type for 10 minutes. Max 5
Crusader: Proc rate on Blessed Shield is increased and also applies to bounces.
Crusader: Akkarat’s Champion duration is increased by 0.5 second for every enemy killed and every 20 Wrath spent.
Crusader: Spending 10 Wrath spawns a Blessed Hammer
Crusader: Running into enemies with Steed Charge knocks them back and does a small explosion of damage that scales with player damage. Every enemy knocked back reduces the cooldown of Steed Charge by 2 seconds.
Crusader: Killing 50 enemies activates Laws of Valor for 5 minutes. Cumulatively losing 50% life activates Laws of Justice for 5 minutes. Spending 500 Wrath activates Laws of Hope for 5 minutes.
Crusader: Hitting a blinded/frozen/stunned enemy with any primary skill calls down a Fist of the Heavens.
Crusader: Blinds from Shield Glare last 12 seconds, and cooldown of Shield Glare is reduced by 1 second for every enemy hit.
Crusader: Condemn’s vacuum effect is always active at the cost of 5 wrath per second and scales with gold pickup radius. +2% damage and damage reduction for every enemy within 15 yards, up to 10 stacks.
Random other suggestions -
Speed shrines are kinda useless. Make movement speed more relevant by adjusting some skills or sources of damage that scale with move speed, like the Shadow Pinions skill from Vampire Survivors. For example, skills or items that leave a fire trail have a larger effect with more move speed.
Do something to ancient items to make the easier to differentiate visually, from other legendaries.
The current situation for regular gems going in helms/weapon/other feels a little too simple. It could be spiced up a little more.
This is about all I can think of at the moment, so if you read it all, thanks! I didn’t expect it would be this long, but once I started to think about things, I just kept typing. If you skipped it completely because it’s way too long, then, well, I understand lol.
this is for you, naksiloth. I made it about halfway through when this thread first appeared, saw most of it was nonsense (Paragon cap? Really?) or struck me as someone who doesn’t have a lot of time to play but wants to compete with people with more time so he wants to throw a throttle on them to “even the playing field”.
I’m thankful for the acknowledgement, though, cheers.
