Desktop Dungeons

[The Dash]

[QUOTE=Fengol;255438]You bastards, I've yet to find the poison glyph! Just remember The Dash, that they're explored now so you don't get healed by walking around. That's an important strategy I use to keep myself alive if I'm too low level to attack the available monsters; and I normally keep a corner unexplored for the boss.

But now you know where everthing is, and with that character you also get extra loot. You can tactically plan out your attack. Only tried it once after noticing the achievement, i much prefer Vampire Dwarves :P

[Nandrew]

Heh, yeah, fair enough: 16-bit SNES looks a helluva lot better than Slot RPG. It also requires considerably more skill to convey ideas with just two colours anyway. :P

I wouldn't say I need to worry about getting a leaderboard up just yet. This game is still in a buildup phase and leaderboards won't really be practical until I fix up the more glaring bugs and start marketing it beyond Game.Dev. And on the matter of interest, you'd be surprised how long it lingers: OOTEZ itself still has a fairly long tail and if anything it's been my own negligence that's caused it to quieten down so much. Marketing isn't for the lazy, and I'll have to pick that ball up again when I get back to active development. <_<

Aaaanyway, Desktop Dungeons still has to undergo a LOT of changes before it enters the main phase of its life. In gameplay terms, we'll probably be looking at some more character classes (I penned down a ridiculous number of ideas which I'll be whittling down to the strongest ones), several extra monster types, *maybe* a few more spells (though I'm genuinely happy with the current selection, funnily enough) and very intense playtesting / balance tweaking. The online leaderboard -- and my main marketing push -- will come when I've got the game built to my satisfaction, a few versions down the line.

If I had to commit myself to a schedule, I'd probably look at making one more release, getting some feedback on that, then following it up with a "ready for marketing stuff" version with online score and most of the core systems sorted. Then stuff can asplode and I'll put a very strong focus on score and other competitive play.

I'm making a delicious gaming stew, and this early feedback and first few guides are the browning of the onions before I chuck in all the main goodies. :)

[Nandrew]

But now you know where everthing is, and with that character you also get extra loot. You can tactically plan out your attack. Only tried it once after noticing the achievement, i much prefer Vampire Dwarves :P

You've got a nice idea, but remember that dungeon exploration is a resource in itself. LEMMISI is very potent in the hands of a Diviner, but using it all at once would be like quaffing all your healing potions before fighting any monsters.

[Miktar]

I would personally simplify the character classes and races, but pop them up a little with more exaggerated traits. I swear to God, Nandrew, if you PopCapified this, I'd buy it for $15.

[Fengol]

I would personally simplify the character classes and races, but pop them up a little with more exaggerated traits. I swear to God, Nandrew, if you PopCapified this, I'd buy it for $15.

Oh hell yes, same here!

[Nandrew]

I would personally simplify the character classes and races, but pop them up a little with more exaggerated traits. I swear to God, Nandrew, if you PopCapified this, I'd buy it for $15.

Simplification, as you probably know, can be surprisingly difficult. But it's definitely a worthy goal. Perhaps I'll see what I can do about giving more classes two -- or even just one -- special trait while keeping their "personalities" intact. :)

[The Dash]

You've got a nice idea, but remember that dungeon exploration is a resource in itself. LEMMISI is very potent in the hands of a Diviner, but using it all at once would be like quaffing all your healing potions before fighting any monsters.

But the problem is because it appears to reveal random blocks, you have to reveal a significant section to make any story of strategy. Prahaps it should only reveal empty tiles, which would be interesting as you could get a rough idea of the walls layout, but with the added treasure there will be lots of monstor confusion

[dislekcia]

Metagame the hell out of the classes! Once you finish with one of the left-column classes, you unlock a corresponding right-column class. Even allow unlocking of races as well, that way initial players aren't overwhelmed with choices early on AND you've even got a pay-hook for later.

You know what the game needs? Set pieces. Things that players can learn will always happen when they recognise certain configurations of the board or of enemies. One of them needs to be a shop-keeper ;)

[Aequitas]

Firstly ... fu<king awesome game Nandrew! You have outdone yourself!

^5

I played as a dwarf vampire ^_^
Getting health from blood was super effective!
Using low health to do more damage made for some interesting choices.

Also, I demand tribute for exploring a devils worth of the dungeon!

[Miktar]

[Miktar]

DAMN! Died on the final monster. :P

[Fengol]

I disagree dislekcia; because he's got good explanations for the different races and classes I don't think initial players will be overwhelmed. Maybe the screen can be improved but the first time I played, I read the descriptions and chose a Dwarf Priest who seemed to have the health options.

[Fengol]

Maybe between my stats and the glyphs you can show what effects are in place for the player? I had the first strike effect on, did 21 damage to a creature with 18 HP and it managed to kill me? I think also, you need to manage the "You die..." message after all the damage indicators have been dealt so I know, at least, how much damage was done to me.

I'm on a mission to see how high a level I can get, so far struggling to break level 7.

[BlackShipsFillt]

I gotta say, when you say "perfect little game" you're not exaggerating. It feels like the core of a rogue-like game stripped of all the fat. It plays nicely, I died three times, then figured out what was going on, restarted once, then finished it (screenshot)... same learning curve as Strange Adventures...

I also really like that no random negative events occur, it feels like all the randomness is laid out at the start and it's up to me to solve the puzzle (which is a good feeling).

I reckon you could take this in many different directions... I like the idea of having some set pieces (like dislekcia said), having levels that require different strategies would really extend the re-playability. Miniboss setpieces would be fun. Maybe some traps (though traps that are easily recognizable but then require some cunning to tackle, like summoning a monster on the trap, while some traps could be fun, like an ambush trap that summons low level monsters), and maybe some story set pieces, like save a princess and return to the king or bring a certain glyph to a wizard.

I think the characters are pretty self explanatory

[Aequitas]

.... The Warlock boss is called Aequitas .... and he has a cool beard.

This game wins the universe.
The End.

[BlackShipsFillt]

I don't know if I'm being mistaken here but humans seem pretty good, I was a bit disappointed with the elf (2 or 3 mana doesn't seem be as good as 20% to 30% extra damage)... potions might be useful, but extra damage with the possibility of first strike is overwhelmingly good. (although I like this strategy, especially when combined with the last stand ability, but from my brief experience the other races seem a bit underpowered)

[BlackShipsFillt]

Ooh.. The Diviner is loads of fun if you get that call-a-monster glyph. Killed Frank the Zombie at level 6 (And I thought he was friendly). The alchemist absolutely destroys if you get an easy start...

There are one or two interface things that might be nice... I can't always remember when I'm death safe, it might be nice to have death safe and first strike indicators. Also the medusas still confuse me a bit, do they kill you when you're below half health and attack them? The gladiator perk of extra experience is a little hard to see the effects of (Gladiator's seem to be a less potent version of rogues or warriors, that get to level 10 at level 9 but they still seem a bit weaker than their counterparts), maybe some way of indicating how much extra experience the player is getting.

By the way... Both Desktop Dungeons and this forum can't take my full name.

[Nandrew]

Okay, so at this stage replying to everybody's comments is going to be impossible. :P I've received a veritable sea of feedback that I'm now wading through (here and elsewhere), and there's been some pretty remarkable ideas among them.

I've not got a new version out yet, but I'm working on it, and it'll be including some interesting ideas and new content.

For a start: I like the idea of dungeon set pieces. I mean, Crawl and others have spammed the **** out of set pieces, so obviously I have to do that too (just because). In fact, I wanted to implement them from the start, but it turns out that it's considerably more of a coding task than a game logic task, and I'll probably be leaving that feature for a *little* bit later. So it's not coming yet, but it sure as hell will be at some point! My only real "set piece" for now is that the player always starts in an open room, and that's been there from the first version. :P

Now let's get to the major bit:

I'm revamping the class system in a considerable way (though keeping the core structure intact 'cos I like it, so nyah!). There are now twelve classes penned and I'm busy coding them in. All of them will (or SHOULD) be well-balanced, but I'm also taking into acccount two new ideas: ACCESSIBILITY and PROGRESSION. I'm about to write a freakin' essay on these ideas, so watch out.

As an overview, lemme show you the proposed class structure first:

TIER 1:
Fighter
Thief
Priest
Wizard

TIER 2:
Berserker
Rogue
Monk
Sorcerer

TIER 3:
Warlord
Assassin
Paladin
Bloodmage


This is how I plan to address ACCESSIBILITY:

(1) When a new player starts up Desktop Dungeons, they'll ONLY have access to Tier 1 classes. That's FOUR really basic characters: enough to provide a meaningful selection, but not enough to overwhelm them (and if it does, I'll take that loss. If four classes are too confusing for a person to choose from, they shouldn't be playing this game anyway. Or in other words, GTFO NEWBIE)

(2) Each class in Tier 1 embodies a particular play style or theme which y'all have probably already figured out. The fighter focuses on a mixture of taking and dealing combat damage, priests are anti-undead tanks, wizards cast spells and **** and the thief is, y'know, everything else. They're character archetypes. Each archetype has a corresponding character in Tier 2 and Tier 3 (for example, THIEF -> ROGUE -> ASSASSIN).

(3) While the four Tier 2 characters are similar to the Tier 1 characters presented, they still have completely different abilities and, more importantly, require a more advanced playstyle than Tier 1. A first tier example: the Fighter is similar to the old Gladiator: fast XP gain and killing blow survival paired with the old Warrior's monster reveal. These are what I consider to be three "newbie oriented abilities": the first one assists with faster leveling, the second essentially forgives a mistake and the third is a guide to show players where to go.

(4) Contrast this with the current idea for the Tier 3 warrior, the Warlord: a similar playstyle (rush into combat and fight stuff toe-to-toe) but with far more advanced and "maintenance-heavy" skills for veteran players: he starts with a skill glyph that needs to be manually activated, has inherited the old Battlemage's power of consuming mana potions for attack power, and also now uses the old Vampire's technique of doing bonus damage when on a tenuous amount of health. This is an appreciably more involved/confusing play style, which is why it's reserved for Tier 3. But of course, vets may well find that he's far more useful and interesting than the lowly Fighter.

(5) In summary: any player who even wants to try a Warlord will first have to "prove" themselves with two similar characters: a slightly less complex Berserker and a really basic Fighter. And if somebody wins with a Fighter, tries a Berserker and finds the latter too complex ... well, they know that they can always go back to the Fighter or try one of the other Tier 1 classes. Win? Win.


This is how I plan to address PROGRESSION:

(1) In addition to only having four character classes to start with, players will also only have five different monster types and nine different glyph skills. These will, where possible, be the more basic / less powerful varieties.

(2) Every time a player wins the game with a new TIER 1 or TIER 2 character, they'll unlock one extra monster and one extra character. From that point on, starting a game with ANY character will include the new monster in the game's random generation. Since there's 8 unique characters in the first two tiers, this will eventually bump the monster count up from 5 to 13: the sort of games you have after playing for quite a long time will be far more complicated and interesting than the ones you started out on, and presumably anyone who has managed to win with these sort of characters isn't a newbie anymore. Thus, veteran gamers get to unlock a steadily more exciting and rewarding challenge for themselves, while less experienced players are subtly (BUT QUITE FIRMLY, MIND YOU -- SO SAYETH THE DESIGNER!) given a nudge towards the full game experience.

(3) So what about the TIER 3 characters? Well, in case a player isn't thrilled at the idea of unlocking more monsters and therefore a greater variety of challenges for themselves (in which case they can refer to my previous statement regarding GTFO NEWBIE), and the reward of new characters isn't enough (hah! As if!), some of the most powerful glyphs in the game can only be unlocked when you win with one of the Tier 3 dudes. And I know that they're the most powerful, because I've picked the ones you all love.

(4) For example, playing as an ASSASSIN will start your character out with the APHEELSIK glyph (most of you will already recognise this as the all-powerful poison spell). This glyph won't appear for ANY OTHER CHARACTER unless you win with the Assassin first, at which point it'll be unlocked in a similar way to the monsters and be included in the "random pool" of spells when a dungeon is generated. Similarly tasty spell unlocks are Resist Death from the Warlord, Heal from the Paladin (it's been buffed now, cos it kinda sucked before), and a new one called BLUDTOPOWA that you can unlock from the Bloodmage.

(5) As a final incentive for progression and experimentation: Consider that playing as a Fighter with a fully-unlocked, fully-stocked dungeon could end up giving you a MASSIVELY different play experience when compared to playing for the first time with only five enemies, fewer spells and a piece of bubblegum where your shield should be. Add to this the possibility of unlocking set pieces as well (though only for later versions of the game) and you've got yourself some delightful replayability. Not to mention, of course, that I'm making an achievement worth several thousand points to give out whenever a character wins a "full unlock" dungeon.


Ultimately, the accessibility and progress are themselves interconnected (unlocking higher tier characters helps progression, while starting with fewer monsters/glyphs helps accessibility). Together, they make SCIENCE! I'm pretty keen on this idea, so who's with me on this? Anybody wanna dissect it? Or maybe just shout "WOOOO!" if it's been tl;dr? I reckon it's a good move.

By the way... Both Desktop Dungeons and this forum can't take my full name.

Yes, but I win because my game includes the "h".

[Fengol]

I love these ideas Nandrew and turn the game towards a Completist style similar to the most of the Popcap games.

What about also having abilities or spells that can be unlocked but used across characters, or even custom classes where the player can pick a set skills they've previously unlocked. Say, after completing all 3-Tier classes.

[Nandrew]

What about also having abilities or spells that can be unlocked but used across characters

I'm afraid you're going to have to clarify that one, because the way I'm reading it now suggests that you're asking for a feature I've outlined above. Poison, healing, death resist and a new glyph are ALL unlockables that everyone can use when you win with the correct characters. ;)

or even custom classes where the player can pick a set skills they've previously unlocked. Say, after completing all 3-Tier classes.

That's a tempting idea (very tempting), but after some thought I'm going to say no. First of all, I think that the definition of particular classes and play styles is important here and should be controlled by the developer. A certain level of customisation is cool, but that's what the race combos and random dungeons are for.

A parallel to this can be seen in Diablo. Actually, **** Diablo, let's talk Borderlands! When you pick a class, you have a certain level of flexibility within that class: weapon specialisations, skill trees, whatever. This helps you create a character with your own "mark" on it, I suppose.

If you choose a soldier, chances are that you're going to act as support, medic, or whatever other roles are appropriate for its skill tree. No matter what you do, however, you'll still remain within some hard-coded confines of the soldier class. You can't Phasewalk, you'll rely on your turret in heavy firefights and though your combat specialisations may deviate from the Shotgun and Combat Rifles, the developers have pretty much shown you that they're the most sensible choice to go with.

And that's great! Part of the coolness and value of being a soldier is that you get to do stuff that none of the other players can! Conversely, when I see another character with Brick running in with freaking EXPLODING PUNCHES, I get a teensy bit jealous for a tiny moment, which is also pretty neat. It affords every character value and flavour.

I want to keep that flavour with each of my classes. If I control the variables, I can tweak balance and make extremes more easily. I can build a Rogue class for my players which has +60% damage as one trait and -5 HP per level as another, forcing a very particular -- but fun -- playstyle.

Given the opportunity to select anything, however, I see things going amiss. For a start, people are going to cluster around a few "prime" builds and see others fall by the wayside. Instead of global leaderboards being filled with fighters, paladins, wizards and assassins and other classes that I can more easily control and balance, we're going to have two or three uber custom builds while half of the skill tree falls away because it doesn't perfectly synergise with the most popular spell in the game. Considering that I want to make score at least as important in this as I did in OOTEZ, this is actually quite a grim scenario.

The alternative? More carefully balance the traits of the base classes. Make them less exaggerated so that there's no easy way to get an "ultimate combo" when customisation comes. But I WANT that exaggeration, and if I have to choose between that and custom classes, custom has to go. :P


In general,there is one rule I try to stick to in games, because I've seen how valuable it is from both ends: NEVER give the players ultimate power. It's really, really tempting to give in: players scream for it, developers think it would be cool, but every time I see them yield to that sort of demand, the game loses its magic. You're no longer humbled by the experience, because you own it now. And while that rush of awesome is great for a moment, it quickly subsides and I shelf the bugger.

Other gamers may think differently, but this is how I genuinely feel when I personally approach a game. And since I don't really want to put a feature into a game which I myself would be disappointed by, I'm going to have to call it a no. :P