Unlocking Species

[vintermann]

[MOD - danr - I split this off from the Spellcaster background thread]

Species are best "simplified" by letting them be a prize target of unlocking. All the races in category 3, and many in category 2 as well, are choice candidates as achievement awards.

[mad]

I'm kind of opposed to the idea of unlocking races because one of the things that drew me to crawl was the fact I could play a Centaur, which nethack didn't allow. It was cool and exciting to have my centaur run around and be awesome even though I hadn't quite gotten the hang of roguelikes. I kind of credit that "OMG I CAN PLAY A CENTAUR" happiness with getting me into roguelikes in the first place. (I don't know why, I just love centaurs)

Unlocking as a mechanic is kind of annoying in general - when Mario Kart Wii was released we got a whole group together to go play it, and it was really annoying that most courses had to be unlocked in single player mode.

Then again, I'm sure someone out there loves unlocking! I think Acheivements are a fantastic idea, but using them to unlock things is unneccesary imo.

[zasvid]

Species are best "simplified" by letting them be a prize target of unlocking. All the races in category 3, and many in category 2 as well, are choice candidates as achievement awards.

I think that this unlocking thing is a rather misguided side of achievement systems. While, I guess, there are some people that need to have the game telling them they achieved stuff to keep them interested, I feel that decorative achievements will suffice to please them. On the other hand, there are people who don't like unlocking and might be turned off the game altogether because of it (for example, I tried playing ToME 4 and quickly quit, because I was bored with restarting the same character over and over and didn't manage to unlock any interesting options - and that's coming from someone who was a big fan of the previous installment of that roguelike).

[mumra]

I think that this unlocking thing is a rather misguided side of achievement systems. While, I guess, there are some people that need to have the game telling them they achieved stuff to keep them interested, I feel that decorative achievements will suffice to please them. On the other hand, there are people who don't like unlocking and might be turned off the game altogether because of it (for example, I tried playing ToME 4 and quickly quit, because I was bored with restarting the same character over and over and didn't manage to unlock any interesting options - and that's coming from someone who was a big fan of the previous installment of that roguelike).

Crawl has achievements. Things like, clearing Orc or Lair for the first time. Getting to XL:10 for the first time. Going to any new branch for the first time. And so on... Can't say what it's like reaching the Orb for the first time, I'm not even close. The game even summarises your character's key achievements in the morgue file. Any more than this is just fluff; I know when I've reached a milestone - I don't need a popup message to congratulate me, or some unlockable as a reward - I've probably already found some great new items or a new monster, that's enough reward.

But then, I've always hated the idea of "achievement systems" full stop Crawl is exactly the kind of game that doesn't need one.

[vintermann]

for example, I tried playing ToME 4 and quickly quit, because I was bored with restarting the same character over and over and didn't manage to unlock any interesting options - and that's coming from someone who was a big fan of the previous installment of that roguelike.

I loved the unlocking mechanic of ToME 4 - although some of it was poorly thought out (dying to a ghoul to get to play a ghoul? dying shouldn't be an achievement, and running to a certain dungeon to get killed is kind of bizarre) it was a wonderful way to get into the game. I have picked up very few roguelikes beyond Crawl, which I was also initially very skeptical to. I suspect I'd have a poorer first impression of ToME of it all had been available from the start ("Oh great, 'wyrmic'. What kind of half-baked pseudo fantasy thing is that?" I would probably have said - but after unlocking it in the game, I thought it kind of cool).

By the way, the tutorial races and classes should obviously be unlocked from the start, so don't worry about Centaurs! What can benefit from being hidden are not necessarily difficult choices, but exotic choices. Undead, Deep Dwarves, Felids. Classes starting in the abyss with an exotic religion. Things that reveal surprising game features, or stretch the game's mechanics. Things a new player has no reason to suspect is fun unless he's played the game for a bit to verify that the devs haven't gone totally overboard.

[mumra]

A big problem with an unlocking mechanic is if, say, you've been playing for ages and unlocked everything - then your hard drive dies and you have to reinstall. Or for online play, you'd need a system where different servers all shared players' unlock information. Just wouldn't work for crawl IMO and to be honest it should probably be discussed in a separate thread, this one is WAY off-topic already

[OneTrueFelid]

Crawl already uses a central scoring/log-in system, I thought? Well, for two different servers.

The tournament from August had achievements of a sort!

[Zicher]

The tournament from August had achievements of a sort!

Which most of newbies are unlikely to reach within weeks for some and never ever for others .

[TGW]

Mumra's got the right idea.

Also, people should note that the variety of possible characters is just as big a draw for many new players. When I showed Crawl to a friend, the first thing he picked was DSCK of Xom.

Correct approach is to help the player choose a character if they want. There are proposals for this on the combo choice wiki page.

[Tiber]

The point of unlocking things is to give the player something to work for in games where the content is the game (i.e. fighting games, where the only thing you really do is pick a character and fight other characters), or to create a sense of progression in games that are essentially a playground (i.e. a multiplayer-focused shooter). It doesn't really work with this game, since the point of the game is to try and take a species and a starting skill set, and make it to end, which the vast majority never will. The purpose of unlockable content is to extend the amount of playtime a player gets from a game, which is usually only a problem for commercial games or games where each round or playthrough is generally short. Crawl fits neither of those descriptions.

[danr]

Also, how would it work - would crawl have to put something in the Windows registry so that it keeps track of what is unlocked everytime you unzip a new Trunk build? And what would you have to do to unlock things - hopefully not ascend with the orb! And how would you decide which order things get unlocked in?

I would really NOT relish having to unlock all the characters again every time I updated my installation.

This is a singularly bad idea, I don't see any support for it, I'm not sure why I'm even discussing it except that I want to kill any tiny remote chance of this ever happening.

But this will never ever ever happen.

[Curio]

just NO
simple as that

[Grimm]

On the other hand the ability to *lock* species might be an interesting thing. This would mean that when you rolled a random character you could turn off races or classes you didn't want to show up.

[danr]

I haven't done it, but I think there's already an option in init.txt for which races / backgrounds you want it to choose from randomly.

Though personally I don't see the point of anything other than Halfling Hunters

[Grimm]

HERE WE GO

[vintermann]

Also, how would it work - would crawl have to put something in the Windows registry so that it keeps track of what is unlocked everytime you unzip a new Trunk build? And what would you have to do to unlock things - hopefully not ascend with the orb! And how would you decide which order things get unlocked in?

I would really NOT relish having to unlock all the characters again every time I updated my installation.

This is a singularly bad idea, I don't see any support for it, I'm not sure why I'm even discussing it except that I want to kill any tiny remote chance of this ever happening.

But this will never ever ever happen.

dpeg has suggested it in the wiki, so it's not all that impossible.

Windows registry?! Why on earth. Since I use Linux, I wouldn't be happy with that at all. The sensible thing would be to track it in a file same place you've got your config file.

No one would be forced to play with unlocks/achievements unless they wanted to. It would be easy to disable, and builds supporting wizmode should probably have everything unlocked by default.

What would you have to do to unlock things? A few ideas:

* Entering the Snake Pit could unlock Nagas.
* Entering Shoals could unlock Merfolk.
* Killing your first Ogre could unlock Ogres (all of these would maximize the "wow, you can play as these things!" factor - though Ogres would need to be improved to prevent subsequent disappointment .)
* Being cast into the Abyss could unlock Abyssal Knights.
* Reaching Crypt could unlock Ghouls.
* Reaching Tomb could unlock Mummies.
* Finding an altar of Beogh/reaching Orc:4 could unlock Beoghites.
* Reaching the Temple could unlock one or more of the religious backgrounds - or perhaps demigods.
* Picking up spellcasting from reading scroll could unlock one of the more exotic hybrids (Arcane Marksman?)
* Casting a Necromancy spell could unlock necromancers.
* Converting to a certain god could unlock the starting religious background of that class (though I think Xom, Trog and possibly Zin should not need any unlock at all)
* Gaining a certain level of skill in Evocations (5?) or using a recharging scroll on a wand/rod could unlock Artificers.
* Gaining a certain level of skill in Unarmed Combat (5?) could unlock Monks.
* Meeting your first dragon could unlock draconians.

The reason unlocks are meaningful in a content-rich game like Crawl, is very different for the reason they are meaningful in "LEGO: Star Wars" or similar. There, they are just to sustain interest in a finite game a little while longer at the point people start to get bored with it. In roguelikes, unlocks serve the purpose of softening the disappointment at losing your best character so far (which is likely to happen in connection to many of these milestones!) and introducing the game's tremendous complexity and diversity in a gradual manner. So, for experienced Crawl players, unlocks would matter little - but experienced Crawl players would also know how to turn off unlocking, which they probably should.

[Jenx]

No.

[vintermann]

I don't get the empathic opposition to a feature that would be 100% optional...

[galehar]

Someone would have to waste their time coding this uninteresting feature. I don't see any of the current active devs doing it. I really don't like the idea.

[vintermann]

It's not unthinkable that I could code it myself, I've been wanting to get into Crawl development for a while (always something which comes up, though... including actually playing it!). Anyway, there's preciously few arguments against it.

[galehar]

It's not unthinkable that I could code it myself, I've been wanting to get into Crawl development for a while (always something which comes up, though... including actually playing it!). Anyway, there's preciously few arguments against it.

Well, if someone submit a patch with clean code easy to maintain, there's no reason not to include it as an option. I would default it to off though.

[mumra]

I don't get the empathic opposition to a feature that would be 100% optional...

The thing is, if it's not needed, then it's basically "code bloat". Potentially introducing new bugs for a feature that isn't even enabled by default - and most people will never enable.

Besides which if you want to get into development on any new project, you're probably better off starting work within a known and understood system, rather than trying to create a whole new one. That way you have an easier learning curve and start getting a feel for development style etc.

But anyway, there's no harm in having a look at the code to see if or how it could be done. Be wary that the start menu code will be changing somewhat by the looks of things.

[Kate]

It's not unthinkable that I could code it myself, I've been wanting to get into Crawl development for a while (always something which comes up, though... including actually playing it!). Anyway, there's preciously few arguments against it.

Well, if someone submit a patch with clean code easy to maintain, there's no reason not to include it as an option. I would default it to off though.

Personally I'd say that alone is never a strong enough reason to add any feature or option - if it's going to be part of the game it should have a good reason to exist other than "there's no argument against it so why not make it optional", or even worse "you can't complain if it's optional" (as a general principle, not just specific to this idea although I'm opposed to it too for the same reasons others have mentioned).

[vintermann]

Sigh. If unlockables are implemented, it will be for the benefit of players starting out. I realize they are completely uninteresting for most players posting in this forum. The current dev team are also experienced players, so it's surprising there are anyone talking about unlockables there at all.

Desktop Dungeons uses unlockables to great effect, so does DarkGod's ToME4, but both of those are one-architect projects. One of the dangers of a game designed by a committee of players who have already gotten very good at the game as it is, is that they (or should I say we - I've won three times too) may easily lose sight of what the game looks like for a new player. I see a lot of really well-though out games made by rising indie stars that embrace the idea of unlockables. I ask again, why shouldn't Crawl? (Technically, Crawl also started out as a typically "indie"-game rather than a traditional free software game, and I think it's a reason why it has so many features in line with a sound philosophy).

Crawl, unlike most team-managed free software games, has an ambitious vision and a philosophy - it doesn't just exist for the benefit of its most hardcore players. But when evaluating what changes might be sensible, it's good to look at what's happening in the indie game scene, and consider picking up the ideas that gain ground there.

So, that you would be totally uninterested in the feature is not really an argument, since it wouldn't be a feature for you any more than the tutorial is. What I want to hear is why you think unlockables would be bad for attracting and keeping the attention of new players.

[Zicher]

-1 for unlockables.

Unlockables simply mean a limitation of available playstyles to me, it's that simple. I'm certainly not the only player who found his niche within Crawl in successful yet not straight-ahead-notable builds, and I would not find it if not for a friend's recommendation.
If Spriggans and/or Venom Mages were locked for any reason and by any requirements needed, I think I'd be a worse, or at least, less experienced player. And I certainly would not be able to fully promote the Poison Arrow Gatling Gun glory to newer players, as I do now.

All of this is IMHO, of course .

[galehar]

One of the dangers of a game designed by a committee of players who have already gotten very good at the game as it is, is that they (or should I say we - I've won three times too) may easily lose sight of what the game looks like for a new player.

Wow, that's not really fair. We have just overhauled the tutorial and we are constantly refusing features that would make the game less accessible. We are really considering new players. That's why I would actually consider adding the feature as optional if a patch were submitted (only if it's good quality). Seeing the length of the wiki page about achievements, there does seem to be some people interested in that kind of things.

[danr]

(ninjaed somewhat:) New players are already helped by the tutorial, and the many places (including this forum) where you can find out recommended builds for beginners.

Unlocking is a gimmick used by games to manufacture replayability. Crawl is infinitely replayable, it does not need a gimmick to achieve this.

Also, an optional unlocking scheme is kind of pointless - who would turn it on? You could always impose on on yourself if that would give you joy, I suppose.

Another thing: you'd have to make the "easiest" species available first, and typically, unlocking gives you access to more powerful stuff later. It would not be very exciting to unlock Ogres and Centaurs after playing a lot of minotaurs and spriggans.

I think you have to let go of the dream, vintermann. I've never seen clearer more unified opposition to an idea. Sorry!

[KoboldLord]

Sigh. If unlockables are implemented, it will be for the benefit of players starting out. I realize they are completely uninteresting for most players posting in this forum. The current dev team are also experienced players, so it's surprising there are anyone talking about unlockables there at all.

With all due respect, how could it possibly be for the benefit of new players to lock all the easiest combinations behind achievements? Deep dwarf necromancers, spriggan enchanters, and felid berserkers are on the 'newbie combinations for absolute beginners' list, and are common first wins.

Unlockables work in Desktop Dungeons because they are fully integrated as part of the game. A level can be cleared in a fairly short span of real-time, it is easy to predict what actions will unlock another set of options, and additional unlocks ramp up the difficulty curve for experienced players. Crawl has not been built from the ground up with unlockables in mind, so it is not reasonable to expect them to work seamlessly.

Unlocking everything every time we get a new trunk would be an annoying slog to experienced players, but it would be downright abusive to newbies.

[mumra]

So, that you would be totally uninterested in the feature is not really an argument, since it wouldn't be a feature for you any more than the tutorial is. What I want to hear is why you think unlockables would be bad for attracting and keeping the attention of new players.

I'm a pretty new player, so I think I'm in a position to comment with some authority (Of course diminished by the fact I've played other Roguelikes and RPGs, but I still wouldn't describe myself as "experienced")

Yes the choice of species and backgrounds was fairly mindblowing at first. I think this will be largely exacerbated by categorisation as discussed in the other thread.

You draw comparison to indie games and Desktop Dungeons in particular (actually that's precisely the game that let to me playing Crawl, when I read that DD was inspired by it). My personal feeling with that game was that the unlocking process was the game - once I'd unlocked everything, I completely stopped playing - there was nothing left to discover. The unlocking process there served mainly to extend the playtime. Without the objective of unlocking features, I frankly wouldn't have bothered completing it with every single character. It's a game that's very narrow in scope with actually very little content. It's basically a series of puzzles, i.e. "how to manage the resources to complete the dungeon with x character". Once you've solved the puzzle for a particular character and dungeon, there's no real fun in replaying it. But I still wanted a bigger challenge; and in a way, by completing Desktop Dungeons, I unlocked Crawl

What I'm drawing towards here is that Crawl is a completely different entity. I haven't yet come close to completing it with any single character or combo, so I'm focusing on mainly one build and trying to get as far as I can with that. I know that when or if I eventually do finish the game, there are a whole host of other characters and playstyles to try out. In the mean time, each bit of progress I make is essentially an unlock - by figuring out how to get further in the dungeon, I've unlocked new content.

Now, different builds suit different playstyles. Once a player gets over the bewildering "how many classes?" initial shock, they may want to try out some of the more eccentric builds to see if that suits them better, and/or to discover game features. For instance, using a Deep Dwarf's passive mapping to get a feel for hidden room locations; or playing a Kenku to explore deep water areas and discover hidden treasure; Spriggans so you can run away from everything; or a Felid to get some extra lives. Just examples. But to block off certain classes for a new player risks actually limiting their freedom to explore the game; if they just can't get far enough to unlock a certain race (and remember that at first it can be extremely difficult to even get to Lair or Orc) they may be disheartened that they're never going to unlock even one species, let alone reach Zot!

Overall: Crawl allows players to play the "unlocking metagame" at their own pace. Each new dungeon level is a little piece of unlocked content (and it's very positive reaching a new level, because at that point you know you can get there again). This is a much more organic and natural way than some kind of forced "goal = unlock" system.

Just my tuppence.

[danr]

I have half a mind to lock this thread just to save us all the time and energy of debating this further. But that would be a bit arbitrary on my part, so I won't.

There's no point continuing the discussion unless there's any interest from anyone in actually coding a patch.

[7hm]

Don't lock the thread just because you don't agree with it.

In any case - I don't like the idea. But I also tried to get my girlfriend to play the game. I don't think limiting a new player's options would be a bad thing in her case. She was undoubtedly overwhelmed by the options. And this includes going through the tutorial. She's probably not the target audience though. That fact is why I don't like the idea - were she the target audience and was she to have the same reaction, I'd be more inclined to favour locking combos, or at least making beginner combos blatently obvious. It's all well and good to say that Spriggan Enchanters and Felid Berserkers are easy. But how is a new person to know that without looking online for spoilers? There are a LOT of recommended combos, not all are beginner friendly.

[TwilightPhoenix]

Maybe there should be a new option somewhere that lists "Recommended starter combos" and sorts them by playstyle. Felid and Troll Beserkers, Spriggan Enchanters, Mountain Dwarf Fighters, etc. What exactly would constitute as a "recommended starter combo" would certainly need to be discussed. But, this could potentially solve the "HOLY CRAP WHAT DO I PICK" issue new players can have without affecting experienced players without having to introduce an unlocking mechanism.

I have half a mind to lock this thread just to save us all the time and energy of debating this further. But that would be a bit arbitrary on my part, so I won't.

What achievement would we have to earn to unlock the topic again? :p

[mumra]

I have half a mind to lock this thread just to save us all the time and energy of debating this further. But that would be a bit arbitrary on my part, so I won't.

What achievement would we have to earn to unlock the topic again? :p

The "You have thanked danr 1,000 times" achievement

[dpeg]

I was the one who brought up unlocking at some point and I still stand behind it. (Before you start screaming, note that my proposal always included an option to list all options.) We have very many players who don't get anywhere in the game (think never fetching a single rune) and are nonetheless faithful. I believe that unlocking new species would be interesting for them. However, much better would be achievements (or badges)... but also a whole lot more work.
I am not willing to talk my fellow devs into the unlocking thingie, in particular, I cannot honestly recommend Vintermann to start coding. Thanks for the offer, it is appreciated. But much better if you start with something different. On the other hand, I am willing to talk the other developers into the achievements. There's a lot of thought necessary to get the achievements right (we don't want "Die ten more times to Sigmund and pick up your badge" nonsense) but I believe it would be worth it.

[galehar]

However, much better would be achievements (or badges)... but also a whole lot more work.

How could it be more work? To unlock things, you have to achieve something first. So coding achievements is a prerequisite for unlocking stuff. I agree that achievements (and maybe unlocking) could be added as an option in the game. I doubt any dev is interested in coding it, but if a patch is submitted, we will consider it.

[mumra]

We have very many players who don't get anywhere in the game (think never fetching a single rune) and are nonetheless faithful.

Well ... I for one am still in this camp. I haven't even seen the Abyss, not even once, although I'm regularly getting characters past XL:10 and reaching Orc and Lair branch ends. I think I've just been lucky. That, and I don't let Yiuf get anywhere near me!

But yes, achievements like you describe don't really get in the way of anything and the game already has them of a sort via the morgue files. It's basically just a case of further statifying that existing database of past games. I'm sure some people like that sort of thing

[Mychaelh]

-1 for Unlocking Species in Crawl from me.

+1 for improved high-score table with additional 'best of'-statistics screen


Demise: Rise of the Ku'Tan

[jpeg]

Well, if someone submit a patch with clean code easy to maintain, there's no reason not to include it as an option. I would default it to off though.

Introducing an option that is both targeted at new players and defaults to false is kinda pointless. More so, if it involves a lot of complex code.
What I could see is a single "recommended combinations" screen covering a small amount of species background combos representative of various playing styles (and without secondary decision screens), with a key "Show all" that doubles as an option (default to false, i.e. the short list) that is remembered across games.

[pratamawirya]

Seriously, I would not expect to find a single unlockable stuff in Crawl. That sounds like something which would suit AAA games more (or flash games).

[omndra]

How about No?
Unlocking stuff is a silly gimmick implied by games in order to achieve a false sense of replay value and content.
Dungeon Crawl does not need to imply this gimmick, as it has enough content in order to provide an excellent amount of Replay.

[vintermann]

One of the dangers of a game designed by a committee of players who have already gotten very good at the game as it is, is that they (or should I say we - I've won three times too) may easily lose sight of what the game looks like for a new player.

Wow, that's not really fair.

I also said "Crawl, unlike most team-managed free software games, has an ambitious vision and a philosophy - it doesn't just exist for the benefit of its most hardcore players." That was meant as praise, in case I wasn't clear enough I know you try. Crawl is much, much better than other team-developed free software games - compare Xmoto with the game it's based on, or Enigma with Oxyd, to see where hardcore fans will take a free software game if left to themselves. Not an ugly place necessarily, but a damn inaccessible one even for people who want to like it.

But, while I praise you, I also must tell it as I see it: That the indie game scene is flowing over with original ideas, and showing us a lot of things about games which we didn't know before. Among other things, just how much a well-designed unlock system can help you get started. It's not surprising to me that Linley Henzell has found a home in that scene.

(Ironically, the idea of achievements may well have spread from Nethack's conducts, so when unlocks come back into Roguelikes, it's an old idea gone full circle and been built upon.)

[dpeg]

vintermann: To me, "achievements" are interesting because they can be used to suggest new things to players -- players of all levels. If someone has reached the Temple thrice, then a good next achievement would be to get full piety with a more obscure god. If someone has won the game thrice, a good achievement could be some kind of conduct. (One of the ideas bounced between the original contributor, jpeg and me was that instead of (or at least in addition to) a fixed laundry lists of badges, there would be a short list (think three) of suggested goals.)

[mumra]

vintermann: To me, "achievements" are interesting because they can be used to suggest new things to players -- players of all levels. If someone has reached the Temple thrice, then a good next achievement would be to get full piety with a more obscure god. If someone has won the game thrice, a good achievement could be some kind of conduct. (One of the ideas bounced between the original contributor, jpeg and me was that instead of (or at least in addition to) a fixed laundry lists of badges, there would be a short list (think three) of suggested goals.)

On this note, well worth a watch: http://www.escapistmagazine.com/videos/view/extra-credits/2758-Achievements

A very thought-provoking look at the different types of achievement system.