Try giving the player an ability to run your own settlement :D Defend if from attacks and personally recruit and train NPCs, from farmers and blacksmiths to guards and alchemists. NPCs will have permadeath, but will become better after each attack wave. You can personally invest gold in their training and buy them equipment.
Then you can take some of them on adventure and they will act as followers. Attack waves on your settlement can be either random or initialized by the player.

It has almost unlimited possibilities for future features expansions: like adding more resources and worker types or numerous buildings types and their upgrades. Later you can even add AI controlled settlements and add diplomacy on top of it with trading, war and battle alliances and even large scale battles inside settlements with pillaging and murdering the innocent. Then your settlement will have a reputation of evil raiders or heroes and others will react to your reputation, including your own NPCs.

IMHO there should be two advantages to crafting that make it viable:

1. Cost - items obtained through crafting are more affordable than purchasing the same item from a shop. The components used to craft an item are, in total, less than the cost of the final product.
2. Exclusivity - for the most part, you can obtain any item from a shop or random loot drop that you could from crafting, but a few select items can only be obtained via crafting. Although the components might be found in a shop, dropped by a mob, or randomly found in a treasure chest, the item itself never will be.

You defiantly have to pay attention to things like level design and loot if you want the game to be fun, not to mention having enough variety of enemies. You can keep the same basic game play throughout more or less, but it is important that you do not get stuck in the same area for too long before you can move on or people will get bored with your game and quit. That is why I quit playing the mana world after a very short time back when I played it.

It is also important to have different enemies and areas so it doesn't get too repetitive. Sometimes, all you really have to do is recolor sprites rather than make totally new ones and mix them up a bit.

The most important part to keeping it fresh is to add new stories. Otherwise, people might wonder why they should go through the whole game in the first place. Sometimes, that can actually be the hardest part. The best stories have some sort of symbolic representation of things that isn't directly stated. It is hard to describe what I mean, but its like the difference between Diablo and Depression Quest. One is a a symbolic representation of going to hell, and the other one is very literal. One is considered to be a more fun game than that other, and you have to think about why that is before you can answer that question.

That isn't a crafting mechanic, it's an enhancement mechanic. You aren't using components to create a new item, you're taking an existing item and "improving" it. Also, diablo 2's gems violate both of my rules: #1) components (gems) aren't cheaper than the final product because components can't be purchased, making them priceless and too valuable to ever actually be used, which leads into #2) it doesn't work nearly as well if the results are 100% exclusive to "crafting". Most of the items should have multiple ways to obtain them, but D2's enhanced items are only ever available via gem sockets. And socketed items are not reliably obtainable. It's more like gambling than crafting.

Money in Diablo2 was pretty worthless when you got right down to it. Players almost never used it to trade. Instead things like gems, ruins, and the infamous soj were used. Just because something is the official currency doesn't mean it is worth anything or you can buy what you want with it.

I disagree. I like the idea of items exclusive to crafting. It makes crafting feel worthwhile rather than some stupid side add-on that no one ever uses because you cannot make anything better than what you get from the elite boss drops or trading with npcs, or gambling. I know there is always the idea that you should have multiple ways to obtain the item, but sometimes socketed items could be found in shops or dropped various places. (if you have ever tried to buy or find crafting supplies irl, you may find out it isn't as easy as you think either, depending on what you want to craft.)

I mean, yeah, diablo2 didn't have a boring mining skill where you hit a rock 100 times, get some ore, and then craft socketed items on a forge. I will be honest that style of crafting is actually kinda boring. It would be better if there were more randomness, both in the quality of items you make and the properties they have. Also, always getting ores and making mining too safe.

Real mining is dangerous, there are cave ins, underground infernos of flame, noxious gasses, not to mention that there should be random monsters attacking you in an rpg. It is stupid that rpg mining is so much safer than real life mining. (My great-grandfather died in a mining accident, he went to rescue some fellow miners and he didn't make it out.)

hmm, I realize I spoke as if my opinion is fact, when there is a huge spectrum of philosophy on crafting and what makes a fun loot mechanic.

I agree that money quickly becomes worthless in D2, which kinda means my idea can't be applied to any clone of Diablo's item and wealth system. If money has no value but is the currency with which you purchase items from shops, none of what I was proposing makes any sense.

I agree with your disagreement, actually. I wasn't suggesting crafted items not be exclusive to crafting. I said there should be items exclusive to crafting, just not every item produced by crafting should be exclusive to crafting. Then again, I forgot about socketed items with gems in them appearing in shops occasionally, so I guess D2's gemmed items aren't technically exclusive to it.

Crafting components should indeed be rare and valuable. The problem I have with D2's system, is it almost feels like they intentionally made the gems too valuable to use. As Danimal mentioned, I would hoard my gems. Waiting for that one weapon that was both socketed and had the skill bonuses I use most. Yes! I found it finally! Plug in my top-tier gems to make it a few percentage points better. Then, inevitably, in the very next dungeon, I would find the same weapon that did slightly more damage, had one additional slot, and had a slightly higher skill bonus. I can't count the number of times I used my gems only to immediately find an item that I "should have saved them for". Did the designers intentionally make you regret using gems?

I guess I think crafting should be a game of resource management. You can either spend more gold (which has to be hard-to-come-by for any of this to work!) buying potions, or you can carry around more potion ingredients all the time and make your own potions on-the-cheap. One is not better than the other, you're just using different resources. Gold vs inventory space. Some times the item you want takes too many rare components, so it's better to spend a buttload of gold on it. Other times, it's way too expensive, so it's easier to kill a few extra mobs that drop that one component you need to craft it yourself.

Dungeon Crawl Stone Soup (DCSS) is a roguelike as is Diablo, but DCSS's random loot seems so much more interesting to me. There are distinct classes of weapons with mostly predictable stats. "crappy hand axe"? It's going to, by default, deal 4 to 8 damage. There are multiple factors that could increase or decrease it from the default, but those factors are apparent. Quality of craftsmanship, enchantments, Materials. A mithril axe is going to deal more damage than an iron axe. A masterwork axe is going to have a better chance-to-hit. Enchantments are random, but they're from a common list of enchantments, so you know a Masterwork Mithril Axe of the Phoenix is going to have the same enchantment bonuses as any other item enchanted with "of the Phoenix". Every weapon type also has something that makes it special. Axes deal 'splash' damage to all targets adjacent to the one you're swinging at, sabers have a chance to riposte when a monster attacks you, spears allow you to hit your enemy (and occasionally push them back) before they're in range to hit you. Whips can cause additional "pain" effect, interrupting enemy mages from casting spells at you. I know Diablo kind of has that too, but it's pretty weak, IMO. A weapon's worth in Diablo can, for the most part, be reduced to a single stat: Damage Per Second. Yes, each class in Diablo has a favorite weapon type that gives bonuses, but it isn't anything I find really tactically interesting based on play style. Just job class.

There is a relevant discussion about Clint Bellanger's mechanics seeking to reduce degenerate play in FLARE: https://opengameart.org/forumtopic/possibly-a-bug-selling-items-by-ctrl-... Clint talked about discouraging players from the boring task of hauling every bit of loot back to town to sell it. Being more based on Diablo than DCSS is, FLARE has the same issue of gold suffering from inflation in the late game that Diablo 2 has. This is, in my opinion, exacerbated by Clint's solution to the back-to-town-treck problem: Insta-sell items directly from the inventory. No need to head back to town just to sell, but now gold has even less value now that its easier to obtain. DCSS's solution to both problems is that you cannot sell items. You can find gold, you can find items, and you can buy items with gold, but you can't get gold by selling items you find. "Wow! this staff would be amazing if I were a spellcaster. Alas, I'm a Berserker who has no use for magic" *item dropped*. This solution is pretty drastic for FLARE, and especially ineffective in multiplayer environments where players can simply trade items they can't use themselves. My point is, there isn't just one solution, and each solution come with another problem to solve. :/

@Jastiv: That's a good point about mining mechanics. I actually didn't know Diablo 2 had a mining mechanic. Must have really been boring if I missed it entirely! :P I don't know how you would make traditional mining risks be fun, but there certainly should be risks involved. I wouldn't say it's stupid that video games are safer (for your character) than real life is (for yourself). The point of games is being able to do something you wouldn't ordinarily have the skills to accomplish in real life, so it still needs to have gentler consequences for your character than it would for an actual miner. Real mining means collapse tunnels, toxic gases, etc, but any mechanic that incorporates those things will need to be more advanced than arbitrarily killing the player because 'real life be like that sometimes'. Reaction time, maybe? The game gives some hint that the tunnel is about to give. If you miss the clue or don't move your character fast enough, you take damage and/or lose a big chunk of the lode. I think making mining risky has some potential!

Some may disagree, but I think that, ultimately, it boils down to world building. There's a lot of things that make the Diablo series fun, and many of those things have been mentioned: killing monsters, getting loot, a cool story, nice levels, etc. One could also add randomized levels to the list, this alone added to its huge replay value. Yet I think that all of these separate elements have to be glued together so that they form a cohesive whole, something along the lines of the old saying "the whole is more than the sum of its parts". What I mean is that, when you're playing Diablo, you're not only killing monsters, getting loot, and following a story. You're also getting a glimpse of Sanctuary, which, as you know, is the world of Diablo. What is Sanctuary like? It's a grim, sinister world, inspired mostly by Christian mythology. For example, notice that the player classes (Barbarian, Sorceress, Paladin, etc.) are all humans. And notice that the enemies are always almost either undead or demons, there are no orcs, or goblins, or ogres, etc. This was a deliberate decision that the Diablo team took, in order to differentiate Sanctuary from other fantasy worlds. Not that there is anything wrong with including elves, dwarves, orcs, etc., on the contrary, it can be awesome. Just look at Lord of the Rings, or Dungeons & Dragons. But my point is that the Diablo team wanted to create a different world. You can already get a "feel" for the grim and sinister aspect of Sanctuary even before you start adventuring: remember the character creation screen. It's a night scene, and in Diablo 2, the characters are gathered around a fireplace. The font used for the letters of the menus are gothic, and the selection symbol is an inverted pentagram, which evokes the diabolical and demonic traits of the mythology of Sanctuary. Imagine if the scene was in daylight, the font was Comic Sans, and the symbol was a smiley face instead of an inverted pentagram: that would have gone completely against the atmosphere and mood that characterizes the world of Sanctuary.

I believe that you should aim for something similar: create a world, and everything else should flow from that. Is your world going to be inspired mostly by Christian mythology, like the Diablo series? Or is it going to be inspired more by Celtic mythology, like Lord of the Rings? Is it a grim, bloody and sinister world, or is it a high fantasy setting with many different races and cultures? Or maybe it's something altogether different, a swashbuckling world of pirates and corsairs, or a "sword and planet" setting were fantasy elements exist side by side with elements of science fiction, such as inter-planetary travel? Perhaps it's a Lovecraftian world, where ancient and unspeakable horrors lurk in forgotten cities and in the ruins of ancient empires, or maybe it's a world where steam-punk meets fantasy, like in Arcanum.

I think that if you can decide on some general features of your world, it will make other things easier.

Wograld has a nice engine that does a lot, actually, the whole thing does too much and the java client needs updating. On the other hand, if we could make it easier for end users to compile/install, it really does have all those crafting features and whatnot. It is actually a retool of old crossfire, but I tried to take on the task of polishing the user experience, and I found it to be beyond me.

I guess that's one way to do it, either make a simple game from a simple engine or code you understand, or try and retool some unwieldy thing. On the other hand Wograld has lots of nifty features like client/server, crafting, gathering skills, random dungeons (ok, well flare sorta has this) etc.

Diablo was originally conceived as a turn-based, strictly single-player RPG in the image and likeness of X-COM.

Blizzard hasn't become money shark nowadays, because Blizzard no longer exists. It's just name and nothing else. First it was bought by Activision? Now I don't even know who owns it. All these people, who took part in making all these great games under Blizzard, don't work there for a very long time. It's like comparing Bethesda's Fallout to classic Fallout and saying that new Fallout is crap. It's just name.

Also, Blizzard always started doing one thing and ended up with something else. Starcraft1 was sci-fi Warcraft2 clone at first, then they remade it completely from scratch. Warcraft3 wasn't even an RTS at first (and it had different name), but something like real time tactical RPG where you controlled a party of heroes, but then it turned out to be an RTS.

Well, since the necromancer has visited this thread...

Don't forget that Diablo 2 was actually a really early iteration of that genre. As far as I know, the genre barely existed until developers started being "inspired" by Diablo 2. Not sure why that didn't happen with the first Diablo.

Don't think that people play these games for any one reason. Some will be more motivated by the story, some by exploration and discovery, others will be more motivated by min-maxing, some will want pretty art to look at (and look forward to looking at). Some will even be motivated by the gambling aspect; more direct gambling is a huge industry for a reason, people who do it are not just motivated by money but by more fundamental biochemical reward loops, and any game with "drop tables" is a kind of gambling with your time towards a different end. The full list of reasons is bound to run into double figures. I expect that the games which aspire to be big hitters in this genre will have to try and tick at least the top ten or so boxes on the list. Indie devs don't have to do that, they should probably pick 3-4 boxes and focus on those.

Since there are hundreds (at least) of "Diablo clones" now and quite large amounts are invested in developing more, it is probably approaching an exact science. Go back two decades and you couldn't even do a university degree in "game development", but now you can specifically do a degree in "game design" at lots of universities. Some of those graduates have probably done entire courses on how to balance Diablo clones for maximum appeal! All of that expertise may be floating around on the internet somewhere, if you can find it.

I've been playing TES:Blades lately, and the fine balancing of the loot system in that game is really noticeable... partly because this is something which TES games have historically done very badly, so it's a huge jump. At the very least somebody has spent some serious time with spreadsheets refining that. I wouldn't be surprised if the most serious gamedev studios are playing with using ML to refine the balance of their games by now, since that stuff has really come of age in the last decade. Hire the right expert and they can show you the whole spread and distribution of ways that players can proceed through the game, including which builds are over- or under-powered and even some of the 'sploits that TAS types will find, before a genuine end user has ever seen it.

Coming back to the original question:

Isn't it the best to take your own preferences and make a game that fits them? Because YOU (as an ARPG player) are the expert on what makes an ARPG great. (That's just a thought by me about this...)

Greetings

Peter

Maybe, but the original question seeks other people's preferences in order to make the game fun to more than one person. 

If I like RPGs for the tactics, but my bf likes them for the story, it's possible to have a game we enjoy together. If I only selected the features I enjoy most, I would be playing alone a lot more. I don't dislike the story, of course, but it may not occur to me to give narrative elements importance unless I ask others what they enjoy about them.

Yeah, good point.

I prefer ARPGs where you have some "safe point" or "save point", you know, like you have in original Flare. Where you don't have to start from the very beginning if you die. And I like opponents which aren't too tough. You could implement this by providing different difficulty levels from which the player chooses one at the game start.

Or to start game like a clan member, and in case of dead, continue the game with next character until all members are wasted. Situation described by @buttons in second post is not bad idea at all.

For me personally the suggestions by @buttons are too much "strategy" for an ARPG. But good thing about those suggestions are: It would be something new and unique :)

Greetings

Peter

After seeing this thread I started working on tactical ARPG for OGA 2022 Fall Jam ~3 days ago, I already made AI, pathfinding and basic mechanics like attacking with weapon, inventory managing & stats affecting things. It's ~15 more days left until end of jam and I'll release it, it won't have much content obviously, but hopefully I'll like the game myself and continue working on it, because my last game for OGA jam during winter wasn't good enough and I lost interest almost the next day after jam ended.

Currently the main idea for my game is to have a hybrid between Baldur's Gate (or Icewind Dale) and Diablo. You can pause game anytime you want with spacebar and during pause you can select your NPC companions and give them direct orders - like what ability to use and on whom or hold position or attack that guy, etc. Your NPC companions will have inventory and skill trees just like you and will gain XP with you, but if they die - they die forever and drop all their inventory. Other than that the game will be similar to diablo clone, but I want to have less mobs (no thrashmobs), but they will be tough, in other words quality over quantity, something like units in warcraft3 where they're few, but hard to kill. So fights will be mostly duels or small group vs small group. The game will also be more slow paced for an APRG and there will be no crazy flashing bullet hell abilties that turn everything on the screen into unreadable mess.

Another idea is to ensure there's no number bloat/inflation. The gap between "fresh level 1 character" and super-hero will be low. Your stats will improve very slowly and their starting value will be just below normalized "average", similar to TES3 Morrowind, where you start with ~40 stats with ~100 being max. The same true for health and mana.

For now the game looks like this, there're just obstacles and 5 angry NPCs who aggro on me to test pathfinding:

@VRS1: Sounds like a good plan. I think I will like your game. (Even if you don't manage to implement all mentioned ideas.)

Greetings

Peter

Yes, it's Unity.

No, I'll use free only and/or self-made stuff for this game.

The UI is partly from here:

https://opengameart.org/content/health-orb-11

and few small parts taken from here (free unity asset):

https://assetstore.unity.com/packages/2d/gui/icons/gui-parts-159068

I simply changed color. to brownish/reddish. Health/mana orbs are animated and very beautiful, itsmars made really great work, you don't see it in his previews, but there're 2 textures in his zip file, which if slowly moved and half transparent, will make really beautiful animation and make those orbs look very 3D.

Edit:

Also for UI background I used this texture:

https://opengameart.org/content/dark-stone-tile

Also even though Unity speeds up things, I still do every feature from scratch, since Unity's built-in features are mostly awful. For example its built-in NavMesh pathfinding is terrible, I had to literally invent my own and spent whole day making only this one feature. Even its culling mechanics aren't perfect - you have to manually disable animators for characters who are outside camera, otherwise they're still eating resources. There're a lot of such issues.

I also use a lot things from game-icons.net.

And here's more UI :D

Hey, I've made it, if you're interested:

https://vrs1.itch.io/oga-fall-game-jam-2022-arena

There're no skills (only 1) implemented, the reason is... I've made a mistake with combat mechanics - damage/defense design was taken from Fallout1/2 and it's just very very bad for an APRG due to %based damage scaling. Not only this, it's also very confusing and user-unfriedly - too much damage types, then there're also damage threshold and damage reduction... it may be ok for TB game where you have only 1 armor slot, but not for an ARPG with multiple armor pieces. I already know how excatly I'll change it, but it requires a lot of time and I can't implement most of the skills before that, so I uploaded the "game" today.

The game is just arena where you spawn friendly NPCs, open chests with loot and fight spawned enemy groups. That's it, nothing else.

Progression of power in a great world; to explore mysteries of a world and yourself.

WithinAmnesia, but how will you solve monotone ARPG combat? Progression and exploration won't matter, if everything you do is moving from point to point clicking moving 3d meshes (or 2d sprites) of enemies with your cursor. Just look at all these ARPGs, they're all the same, in Diablo1 you moved from temple level to level, clicking on demons, in more modern ARPGs you choose some skill or whatever and just spam it like regular attack. Nothing has changed since Diablo1. People laugh at cookie clicker games thinking they're not serious, but ARPGs are cookie clicker games, just more beautiful, usually, and with fake complexity.

I managed to play Path of Exile for a long time only because I played summoner witch (later necromancer). I was just walking from room to room, while my army of zombies, skeletons, spectres and a golem destroyed everything. This was long time ago, before all those nerfs and when summons dealt damage even if you're dead, so I could kill any boss in the game, even if the boss killed me, by just lying there on the floor, waiting until my army kills it. Then I tried to play non-summon build and got bored in less than 1 hour, because I was too lazy to click on every walking 3d mesh on the screen to force it play death animation. Literally cookie clicker game. Summoners at least play walking simulator, it's less annoying, just pointless.

@VRS1: You are comparing Diablo-esque games to cookie clicker, but you were too annoyed with diablo-esque games because you had to click too much? Haha! That's pretty funny. 

But in seriousness, how would you solve monotinous ARPG combat? I've thought about that myself. I have a few ideas on how to make combat feel more like the reason to play and less like an obstacle that gets in the way of the quest. How would you design combat so it isn't just mindless clicking? 

In what way does "monotonous combat" as a negative invalidate "progression of power in a great world" as a positive?

I tried to add tactics to it with pause-anytime-at-will and manual full party control. You even played that game on OGA jam judging from your comment there, so you can even say if it worked or not. But I think it didn't work, as it was mostly melee oriented game (typical fantasy setting) and it was just mindless slaughter. Melee + tactics may work only if you add more mechanics like flanking and zones of control and that's turn based mode or something similar to neverwinter nights or pillars of eternity, but real time + zones of control is very clunky and feels wrong, like a turn based converstion to real time. So in practice it should be turn based only mechanic, which isn't ARPG.

So in the end my plan failed, that's why I asked how to solve monotonous ARPG combat. Btw, even 1st/3rd person melee ARPG combat suffers from stupid combat mechanics - from primitive games like The Elder Scrolls to more "hardcore" like Dark Souls - it's all about AI abuse. At first it's look fun, then you learn enemy/mechanics behavior, then you wait until AI does something stupid and then you act - the essence of every 1st/3rd person ARPG combat. Playing 1st/3rd person ARPG is like feeling like you're the only sentient being in the world and you just abuse everyone you meet, because they all have 0 IQ. Again, only ranged combat will solve this.

In what way does "monotonous combat" as a negative invalidate "progression of power in a great world" as a positive?

If ARPGs are mostly about combat (otherwise it's a walking/dialog simulator and thus shouldn't be called ARPG), then if you don't solve this rather important problem - other positive factors will be wasted, because the game will be a failure. It's like creating a shooter where you forgot to implement guns, but made really beautiful scenery and realistic sounds and other effects. Yeah, sure, it's cool, but people won't be able to enjoy the game, because its main mechanic (shooting) is terrible.