Thoughts on leveling up in Diablo 2 vs Diablo 3

Been playing both of these lately, and it's interesting what Diablo 3 gained and lost with its drastic departure.

[@Garryl367]

For reference, respeccing in Diablo 2 was added in patch 1.13, a long time before Resurrected came out. You get one free respec from Akara after you complete the first quest in each difficulty (no gold cost or anything). If you want to respec more than those 3 times, you need to farm up a special item from each of Diablo, Baal, and Mephisto, plus a similar one that drops from either of Andariel or Duriel and combine them in the Horadric Cube.

[@RMJ1984]

Which is a good way of doing it. It should require some work and effort to respec.

[@wardogblaze5106]

@@RMJ1984 the problem there is that if you have a character that can't clear that content then it's just a dead character. It's part of why I just can't get into path of exile. There's so much complexity and room for player expression in the builds you can make, but unless you look up one of the top builds for a season or have a lot of prior knowledge about later game systems and mechanics, you can easily end up making something that can't progress, and also can't efficiently farm the respec orbs. It feels less like a reward for having a good idea for a build and more a punishment for not having having complete information on not only how the game works, but how it *will* work at endgame

[@WizardofWestmarch]

I feel like the problem is having skills drive something as fundamental as damage is always going to lead to an optimal build. The way I look at is have skills drive tactics not damage.

[@mikepence1933]

this is implemented partially but (imo) very well in Battle Brothers, each weapon type has a different moveset of skills, each skill has a different fatigue cost, base damage, armour damage, armour penetration, and (rarely) bypasses certain enemy defenses or inflicts debuffs as a result you will often see experienced players carrying half their bag space or more in spare weapons, and swapping them out to suit the enemy they find themself fighting this combos with the perk system (your immutable build) and your stats (which you control the improvement of, but not the base stats your boys start with), so that while certain lads will lean towards one role or another, everyone can pinch-hit if you suddenly need someone in a role, albeit at lower effectiveness while several perks are straightforward stat bonuses, the majority unlock tactics or offer incentives toward certain tactics, such as swapping positions with friendlies allowing for both offensive and defensive movement through otherwise impassable scrums, movement traits freeing up a world of terrain advantages, ganging up on enemies going from mildly advantageous to cruelly hilarious, or breaking enemy morale en masse through chip damage that would normally just enrage them

[@doodledibob]

My favorite progression (at least for the base game, leading up to max level) is Guild Wars 2. The sense of exploration, where every thing you do always gives you progress, is absolutely incredible. There is some problems with UI bloat, but the hero point system and build/specialization screen selection/customization is nothing short of incredible. It does a beautiful job combining mechanical progression (skills and passive perks are both learned through the same system) as well as presentation (using that gorgeous stylized paintbrush look). I don’t know what kind of game you want Tactical Breach Wizards to be in terms of progression, but I think it’s worth looking at GW2’s UI when it comes to this. I really do think it’s one of the best examples of progression done right.

[@javi7636]

I think the best thing about GW2's progression system is that you do have to commit to a build while leveling up--you have to choose what you unlock with limited skill points--but you can always switch between skills you've already unlocked, so as you level up you're not just increasing your power, you're also expanding your scope of options. By the time you're really familiar with the combat system you also have the maximum flexibility to adjust your build. It dovetails really nicely.

[@TwiiK]

Haha, I love how abruptly your videos end. Amazing that you're able to put these out with zero editing. Your ability to speak without a script is great. For me being able to have a unique identity is the most important thing in an RPG. I remember the Asheron's Call MMO from ages ago, one of my all-time favorite games, where it would take years of playing to reach high levels and I don't think you had the ability to reset any choice you had made along the way, not even your character's name. And since there was no KZbin or barely even any internet nobody had a clue what was "the best" so every single person you met was completely unique. And if you wanted to play on a different server you could sell your own character on Ebay or whatnot and try to find a character you wanted to play on the other server, but that would be like inhabiting someone else's body. You would be buying their exact character with their name and every mistake they had made along the name, not to mention their reputation in the community. Sadly replicating this experience today is impossible due to countless factors.

[@agentjwall]

11:40 Is there a German word for the feeling that there should be a German word for a specific concept?

[@Hanafuss]

wort neid?

[@alastairvanmaren5243]

Fuuuuck, I didn't realize how much I needed a Tom Francis ARPG until now . . .

[@TheLordCrimson]

The problem I see is that optimizing a build is actually just playing a different game than actually engaging in combat. When optimizing a build your player is trying to find a solution to a puzzle and while playing you want them to experiment and these two things just clash. Infinite respecs solves the gameplay issue but just removes the puzzle. No respecs means that they'll either look it up (thus also removing the puzzle) or end up being less effective. (or if the game is badly designed even stuck) The feeling I'd go for is to allow the player to respec infinitely but have it tied to something that isn't trivial but is repeatable. Maybe give a single respec as a reward for an optional mission or give them out as random rewards through the game, something like that.

[@oldvlognewtricks]

Metagame is a thing in every game, to a greater or lesser extent

[@peezieforestem5078]

A simple way to resolve this issue would be a shared skill pool, as in Trials of Fire. Part of the skills can come from the character and be unique and permanent, and part of them could come from equipment that is universal and easily interchangeable. Another way to approach the issue is to make the permanent character progression be extremely broad, with specific abilities coming from easily interchangeable equipment. To give you an example in Breach Wizards, the Witch could have a permanent skill "Excessive force: all movement abilities move targets 1 more tile", and the Riot Priest could have a permanent skill "Police brutality: all knockback damage is increased by 1". And then you equip Gale Grenade on both characters, which provides the active ability. You also have the option to restrict equipment to certain categories of characters, e.g. wands can only be equipped by witches, and rings/scrolls/grenades can be equipped by everyone. The next approach is to divide the skill tree into layers, and in order to unlock the abilities in the next layer you have to permanently lock an equivalent amount of abilities in the previous layers. This will let players experiment initially, and will gradually solidify the build as they level up. Finally, you can tie abilities to easily interchangeable equipment, but level the abilities by using said equipment. This will make it more efficient to stick with the early equipment/abilities, but will not prevent players from changing the build if they really want to. Forcing players to level equipment from 0 will mean that builds power will drop when it's changed, which can incentivize players to revisit earlier areas. Overall, when you want to be "different from any other Warlock", it means others cannot reliably recreate your Warlock, which in turn means you cannot reliably create any given build. The reliability here can come either from overcoming randomness with deterministic design, or with brute force (time investment). Since determinism will negate unreliability, we are forced to integrate brute force. To translate this into human speech, in order to truly solve your conundrum you have to make a randomized system, where randomness can be overcome with time investment. A simple illustration of this would be the Darkest Dungeon quirks: you can either let all the quirks fluctuate; lock some and leave others fluctuating; lock everything by "rolling with the randomness"; or very methodically grind for specific rolls which you then lock and obtain "the build you envisioned".

[@Garryl367]

Regarding the contention between build permanence and the ability to try out different different options, may I direct you to Path of Exile? It's got a decent mix of the two, especially once you get experienced with the game and realize what resources are more readily available or alterable than you think. There's a massive passive tree into which you assign ostensibly-permanent points. These points can be refunded, with a limited amount of refunds being granted from quests throughout the main story/acts, with additional refunds available from a moderately expensive currency item that you can expect to find or be able to trade for in reasonable amounts. You won't be able to completely respec your entire character until you have endgame-level riches, but it's certainly feasible to switch between different variations of the same general build, such as swapping from elemental damage to physical damage, or between a life-based build and one that has very little life but uses alternative defenses instead. The deeper you go into the passive tree, the more specialized your focus becomes, but earlier passives tend to boost fairly broad categories of skills. At early levels, your passives might be boosting any sort of spell, with later points narrowing it down to spells that deal lightning damage, then lightning spells that care about crits, then lightning trap spells that care about crits, etc. Skills are itemized as "gems", most of which are easily purchasable from vendors. While it isn't free to swap them out for a completely different skill set, you can readily adjust which ones you're using, especially when transitioning between skills with similar requirements or mechanics. During the earlier levels, this lets you freely experiment between different skills that are sufficiently similar to scale from the same stats without any inherent loss in power.

[@normalasylum]

I think the phrase you're looking for is "optimizing the fun out." Quotes: "Many players cannot help approaching a game as an optimization puzzle. Given the opportunity, players will optimize the fun out of a game." (Soren Johnson) And therefore, "one of the responsibilities of designers is to protect the player from themselves." (Sid Meier). I highly recommend looking up Johnson's blog (Designer Notes) on the subject.

[@Poki3]

I personally prefer leveling systems that limit what things you can use, but allow you to eventually get everything so you can swap around. I play Guild Wars 2, and as you level up you have to pick what trait lines you unlock and what skills you unlock, but you can only have a few of each equipped and by max level you have them all so you gain versatility as your character grows. I really like trees with a few options to pick in each tier that you can swap between. Like when you level in Xcom you pick one of 2 traits, except it not being permanent, or in Hades where in the mirror you have to pick between 2 traits in each tier for the given run. Ultimately, given the chance, players will optimize the fun out of a game. I think D&D has optional rules that allow you to change 1 trait to a different one on every level.

[@hantuchblau]

I'd like a game which uses respeccing as a normal part of gameplay to force experimentation rather than undoing mistakes. Transistor kind of does this, when you die you temporarily lose some items, but that reinforces the local maximum because you stop dying if you find a good build. Roguelites are an interesting version of this because restarting is respeccing but players mostly stick to their favorite builds. It's so hard to both offer player agency and encourage varying play styles! Feels like there must be some undiscovered trick to solve this which will be copied a bunch once it becomes popular in one game. Also, small percentage changes always feel boring to me. I'd rather flat upgrades to everything than deciding between 3% more health and 1.5% more damage.

[@diegofloor]

Trying to avoid dominant strategies is a worthy effort. BTW, there's a little noise at the end, from you handling your phone, that sounds like Doom's shotgun. :P

[@Torcher999]

Tales of Maj'Eyal has an undo last x skill points spent when out of combat. It kind of addresses both build viability and character permanence points you brought up. It has a different problem where it encourages spending points in advance and undoing them before permanent choice points.

[@LaughingThesaurus]

I've been on a pretty extended ARPG game design thought experiment tangent for a while, and I've been thinking about it as someone who's played tons of games in this genre. I can testify to this much: Unconditional damage upgrades suck, they're boring, they're degenerate, they kill all of the cool stuff that you try to design into the rest of the system... and yet, it's hard to come up with upgrades and effects that aren't just damage upgrades. I think there's a reason ARPGs tend to offer up a lot of raw damage stuff, and it's partly potentially because of balance oversights like the Assassin Kick thing, but, also because it's hard to come up with more situational effects. It's a lot of extra design work. But basically, I think keeping unconditional damage upgrades out of the equation entirely can lead to a lot of interesting things. You CAN have a damage upgrade on a skill if it is used after another skill to make combos. Maybe you can trigger chain lightning on frozen enemies with lightning skills. Try to design skills like you're designing a card game, where you must use things in a sequence in order to get any valid usage out of them. For the record, +50% cold damage to frozen enemies... does not count as a conditional damage bonus in my opinion. It self-synergizes. Gotta watch out for stuff like that.

[@LaughingThesaurus]

I also like the idea of rewarding players with cool powerups if they switch up their skills often. Warframe's mastery rank system is kind of awesome for pushing people to try out a lot of different weapons, even if it doesn't always work out.

[@Schmaratul]

Will always remember Diablo 2 for playing max Skellys Necromancer and cruising, then getting to the first real boss and she just one hit all off them. So I got out of there, got new skellys from corpses, do like 2% damage, get out of there, rinse, repeat, die, try again, real fun. Had a blast with Diablo 3 on release 😅

[@tbmatuka]

Just giving the player the first level of every skill would guarantee that they will have it in levels and not feel like they wasted a point. Even better, you can tie in the skills with the story (the character knew they would need to do something, so they explicitly went looking for a specific item that gives them the skill, etc.) and you could even give the player new skills at trigger points during combat. Once the player has all the skills, they should be more likely to try them all out and see which ones they prefer and put more points into them. . Respec strategies could be tested during alpha/beta to see how players behave with free/cheap/expensive/single-use/no respecs and how they felt about them. I expect that with a single-use respec, almost nobody will actually use it because we all like to keep single-use stuff forever, just in case :D

[@tbmatuka]

This has been bouncing around my head a little bit since this video. Years ago I had this idea/wish for elder scrolls like progress where whatever you use levels just through use, but I hated having to choose which base stat to upgrade on each level. What I wanted was base stats tied to the skills, so that leveling a skill also raises your stats. This would allow for a completely maxed out character if you're willing to grind, but that's actually OK for a single-player game, since most players won't do it, but those who would like to, can. . In your case, this wouldn't happen because there's no open world to grind in, but then you have the problem of potentially broken/weak builds where people tried every skill for a bit as they got them and never actually leveled any of them up enough. This might still work if they get base stat upgrades through leveling up each of the skills a bit, but this would probably affect level design and balancing, which just seems like the wrong thing to do :D

[@rich_in_paradise]

Locking you into skill/talent choices is a real old-school CRPG thing. I think it reflects a time when the designers looked at these games more for their role-playing aspect than their gaming aspect. I think we can take it as read that if you're going to tune the difficulty to be anywhere near tight, you can't lock people in otherwise they'll have to keep restarting if they realise they wasted a point. The bigger difference between Diablo 2 and 3 is down to having an RPG-style skill tree in 2 where everything is designed to lead from one choice to another, and just a pool of abilities and modifiers in 3 where getting the best out of your character comes from synergies and is much less obvious. Worth mentioning also that in D3, your choices don't really matter for most of the game - it's not tuned at all tightly and you can change the difficulty at any time. It's only end-game rift dungeons where you care, and to max out your capabilities you have to chose talents that synergies with gear effects too. Most of this gear you won't get until the end-game.

[@NeillSmith]

In diablo 3 builds are really the collection of unique buffs you get from gear. Doesn't really start until you get to the level cap and start advancing beyond that. You start having to make more permanent choices like investing crafting materials into improving specific items. The biggest difference is you've played the character longer by the time you start to make those choices. The major major downside is only certain skills are viable as you continue to progress because not everything has a buff from an item that's big enough to compete.

[@chrisdenney2997]

this video comes at a weird time.. I was literally just playing D1. I've put like 100 hrs into it in the last few weeks. lol There's just something special about that first game to me.

[@TheHopperUK]

Recently got Diablo 3 on the Switch and I'm not quite sure what I think of it on console yet. I used to play on PC but I swore off it a while back.

[@AS-ym2bp]

Hi Tom, I am a 3d designer and I'd like to have a talk with you regarding licensing of some of your assets from the game "Heat Signature". What's the best way to get in touch? I do not have Twitter.

[@bokchoy7872]

I think Last Epoch finds the middle ground I enjoy with a good take on how to allow respec-ing without it being a tremendous hassle (PoE/D2) or at a whim (D3). There are a limited number of skills and each skill has its own web containing large, significant modifier nodes (D3-esque runes) connected by smaller % nodes (PoE/D2). Each skill can be individually leveled up to 20? 25? so you can't hit all the nodes BUT the cost of repec-ing is just skill levels. For each node you respec, the skill level just decreases by 1. So it's not particularly punishing but there's juuuuust enough friction that keeps you from doing it every 20 minutes but you know it's available if the build isn't working out. I also rather like the fact that re-leveling the skill is just playing the game instead of using hoarded resources like gold, etc. There's a ton of other stuff that Last Epoch has taken inspiration from but the skill implementation is definitely the most interesting imo.

[@StanNotSoSaint]

How excited I felt at level ups in D2: "Nice, I get to click on Vitality five times, and click on Blizzard one time whoo wee, and if I do something else my character will suck".

[@oldvlognewtricks]

The existence of those optimal strategies has killed every period I’ve spent playing D2. Either because I’m falling behind the curve and it becomes thankless, or because I’m playing the boring build and not having enough fun.

[@ExplosiveDisregard]

I agree that wiki builds are boring. It's a big reason I have never really gotten into Path of Exile. It does have a partial respec system though.

[@Notemug]

The biggest flaw of D2's system, to me, is not the 'you've made a big mistake and you won't find out', it's the fact that the gameplay is the same, if I'm stuck to one, very slowly evolving build. It's hours upon hours of doing the exact same thing. D3 solves this, of course, but there, it's too easy to keep changing your playstyle all the time, thus devaluing it, as Tom points out. I hope they manage to find a middle ground for D4.

[@SomeAssh0le]

bruh audio balancing, or just set it to mono, ahhhhh

[@dayf722]

"dabbo"

[@MagpieMcGraw]

They way to play Diablo 2 is to look at the skill tree. See which skills receive damage bonuses from each other. And then pick some combination that sounds good to you and you play the game focusing on those skills. And if you don't like it, use your respec to try a different combination, or you make a new character. The people who originally designed Diablo 2 are playthrough gamers. They weren't focused on what the best build is, but on a particular characters journey through the game, mistakes and all. It's a really old school design compared to todays live service games. Well Diablo 2 is a live service game these days, but it has that archaic core to it. Regarding assassin: nobody uses dragon talon in normal. Although in the new patch it should actually synergize very well with the elemental martial arts. Dragon Claw releases 2 charges, Dragon Talon would release 3 if you put enough points to get 3 kicks. The problem with Diablo 3 is that nothing you do before level 70 matters. At level 70 you get items that are drastically more powerful that anything you found before, you so you just have to throw away all the gear you found previously. And your skill choices don't matter either because you can reassign them at any time. They should just remove leveling from the game. You should try Grim Dawn. It has respecs, and they increase in price the more you use them. You can try several different skills with the same character before the price becomes untenable.

[@RMJ1984]

I honestly don't know which is better. But i know which one is inferior and that would be Diablo 3's skill system. It's simply not possible to balance 1 point skills, runes or no runes system. Every single skill in Diablo 3 is either underpowered or overpowered. Nothing is where it should be. Hence why they have to use sets to give your abilities 5000%, 10000%, 25000% damage, i even think there is one that is like 45 or 50000% damage. It's just absurd and stupid. Abilities having many levels, mean that you find item with + skills and that even if you are unlucky not finding items, you can improve damage and or functionality when leveling up. I will personally take Diablo 2 skill system any day of the week over Diablo 3's. The developers are so proud of their many billion possible builds, and yet every class has 1 or 2 builds that is viable at the highest level. It's Borderlands all over, oh our game has 17 million weapons. Yeah and so what? maybe 500 or 1000 of them at the most is fun, interesting or worth using. Same with open world games, our open world is so big you will never believe, yeah wide as an ocean deep as a puddle. Not sure what has happened in game development, but developers are focusing on quantity above quality.

[@jabberw0k812]

I do tend to prefer the risk/reward of older RPGs. Allowing the player to make mistakes can also make them feel good for making a wise choice. Assuming they aren't just using a wiki, but for a game like Diablo, I feel like that's basically the same as just spoiling the game with a walkthrough.