Desktop Dungeons

[dislekcia]

Cool thanks for clearing that up :)

What does one have to do to be part of the beta?

Right now we're thinking all pre-orders should get beta access ;)

Two days ago, I redownloaded it 10 times, so its actually <24hrs old.

Oh damn. I'm going to have to go through the zipfile release with a fine-toothed comb then. Thanks for letting me know!

[Ezzer]

First up, have a medal! Totally hadn't played a Transmuter like that before, nice.

I kinda feel like it should be an achievement.

STRAIGHT TO THE POINT: Slew the dungeon boss without having killed any other monsters

[BlackShipsFillt]

Transmuter worshipping Binlor. Killing things is not a necessity!

Genius! I have to say, I'm kind of in awe of you Ezzer. Either you're incredibly lucky, or the finest Rogue-like player I've ever heard of.

Also I'm in awe of Desktop Dungeons for having a ruleset that allow this kind of interesting thing to occur.

Pre-orders?!

[Drew]

Where can I preorder?

[Ezzer]

I'm kind of surprised myself no-one seems to have thought of it before. It seemed like a rather obvious strategy!

[dislekcia]

I'm kind of surprised myself no-one seems to have thought of it before. It seemed like a rather obvious strategy!

I think a lot of people dismiss the Transmuter class because it's so different to play.

Those huge class differences are what we're trying to get our heads around for scoring at the moment ;)

[Drew]

The way you describe the new main menu sounds over whelming D:

Will it be easy to adjust to?

[dislekcia]

The way you describe the new main menu sounds over whelming D:

Will it be easy to adjust to?

We hope so! The choices that players have to make have been streamlined a lot: Instead of having a horde of buttons that you can click on at any time (even though at the moment it sorta looks similar) we've made each step along the way highlight specific buildings that you can interact with. Some buildings serve multiple purposes, depending on the decision the player needs to make. So the Adventurers Guild might provide a class choice to the player when it's time to select a class, as well as being highlighted during preparation selection if the player has upgraded the adventuring lockers. The idea is not to overwhelm anyone, but make more room for choices at the same time. It's a tricky balancing act and one of the major reasons we need to run a beta.

[Catwalk]

About level scumming and gold hoarding

Gold feels pointless at the moment because you can simply restock before a real run. While players can impose house rules on themselves saying not to do that, I think most players would have more fun with a different system. It's a little like allowing for reloading, players can simply refrain from using it but it might be too much of a temptation for many players and would end up making the game less interesting. This can easily be tied in with level scumming. I propose the following:
1) Starting gold for a new game is determined by how many successive runs you've made
2) Only one run per class at a dungeon will be counted, so you get no benefit from completing the Crypt multiple twice with a Transmuter (for example)
3) For each unique successive run, you get 5 + x/3 gold added to your starting gold total in new games, where is the number of classes you've completed the Crypt with (replaces current bonus to max gold)
3) No gold is carried over otherwise
4) Whenever you die or fail to complete a run, the progress of one class at this dungeon (for purposes of gold hoarding only), chosen at random, is wiped from your records
5) This system is only implemented for the four challenge dungeons
6) In Lothlorien you always start with 100 gold in the first mission, and you keep whatever is left over
7) In the Gauntlet the system is implemented with three differences: Progress is tracked independently of the challenge dungeons, you're allowed to use the same class multiple times and the gold transferred is twice of what is transferred for challenge dungeons
8) When mousing over a dungeon, the game will show you both which classes you've completed the dungeon with and which characters you've completed it with in the current run

This way the normal dungeon is used mainly for practice and unlocking stuff without having to worry about dying, and the challenge dungeons become a bit more challenging. You'll really feel the pressure to squeeze out a win from a difficult map, even if it seems impossible. On that note, "impossible" maps should be removed. You should not be stuck in a room where all the exits are blocked by high level monsters.

It would also be possible to simply wipe the progress each time you die, but that would be rather heartless and probably frustrating to many players. A setback is a more interesting challenge than a complete reset.

[Catwalk]

About regeneration and exploration

One thing that both intrigues and annoys me about DD is the regeneration system. I really like the concept of non-replenishable resources that you have to make the most of, but I absolutely hate having to ration my exploration. It makes luck a much bigger factor, especially for purposes of finding altars and glyphs early on. If you have to walk through half the factory in search of a fireball glyph and a high level thingy-that-is-vulnerable-to-FB (forgot the name), your run is all but over. Whereas if you find both in the first handful of explored tiles, your run is all but won. And with most altars giving significant bonuses for signing up early, the problem is compounded. At the same time, it's usually way too easy to explore most of the dungeon killing just a few level 1 monsters. I propose the following:
1) Monsters will be specifically set up to block your progress during map generation
2) The map generator will be more likely to put low level monsters near your starting position and high level monsters further away
3) All dungeon features (items, altars, shops) will be guarded by a monster standing on top of it
4) Thieves will be able to steal items from monsters and rogues will be able to sneak by monsters (not sure how)
5) Automatic regeneration is replaced by food rations which will be scattered around the dungeon
6) There will be a rest button which lets you consume food to regain HP and MP, in the same way regeneration functions now

This will encourage exploration while at the same time making it more difficult, and will keep the core regeneration system essentially unchanged. There will still be a finite amount of regeneration available in the dungeon, and monsters will regenerate when you rest. Monsters will be of more interest because they're no longer mere XP farms waiting to be harvested, you'll much more often need to kill them for a specific purpose other than XP.

[Catwalk]

Miscellaneous ideas

1) Dodge is changed so that it takes effect every x blows, completely predictable
2) Potions will allow you to exceed normal HP and MP, all surplus points after reaching your max will be halved before they're added to your total
3) Gods changed to make it less important to sign up early, there are already measures in place to do that but they aren't really working
4) All altars will be guarded by a high level monster (7-9)
5) All altars will be present on every map! Or maybe increase it to 5 altars per map.

[BlackShipsFillt]

About regeneration and explorationI really like the concept of non-replenishable resources that you have to make the most of, but I absolutely hate having to ration my exploration. It makes luck a much bigger factor, especially for purposes of finding altars and glyphs early on.

I do like a lot of your ideas Catwalk (and I think QCF probably really appreciates that you are providing creative feedback), but I just wanted to comment on this one.

I can't exactly speak for Disleksia and Nandrew, but I'm pretty certain that having exploration a resource is Desktop Dungeons.

I hear what you're saying though, it does beg a bit of extra suspension of disbelief. But imagine this: If Desktop Dungeons were to be modified like you suggest then players would always explore every square available, in fact as soon as you kill a monster everything the monster was blocking should become revealed automatically.

So the only choice players need to make is which monster to kill next and whether or not to consume food/use alters/buy from shops.

This admittedly is what the player goes through in other rogue-likes, but in Desktop Dungeons, the way it is, every tile of movement is an important tactical decision. Although what you have suggested could be just as difficult as Desktop Dungeons, the tension that Desktop Dungeons creates, that forces players to take it seriously and makes the reward for completing a dungeon that much sweeter, would be lifted.

Sorry about the lecture... on the other hand I really like the idea of guarding shrines... It has got a nice flavour to it. Also I like the idea of pickpocketing monsters or finding sneaky paths.

I still marvel at the brilliance of turning exploration into a resource in a rogue-like game. I guess I feel passionate enough about it to defend it.

[Catwalk]

You make valid points, and I concede that I'm probably in the minority on this one. I've debated the same point for many games countless times and found myself on the losing side more often than not. I'll stick to my guns, though :)

While I agree that this is an integral feature of DD, I think that my proposal maintains the spirit of it. Regeneration is still a limited resource that requires you to explore, and monster regeneration will function exactly the same as now. I feel strongly about luck in games, and find that this particular aspect of DD adds way too much luck. In that way, it's highly detrimental to my game experience. Many aspects of the game serve to minimize the luck aspect (such as combat and even distribution of items on every map), so I find this to be a great shame.

Yes, the game would be a little bit simpler with this proposal. To me, the puzzle part of DD is a big part of its appeal. It has a rogue-like feel to it, but it actually breaks with a lot of rogue-like dogma and succesfully so. I think this proposal would help stress the puzzle part of the game greatly. There'll be more known information and you'll be able to make more qualified decisions rather than having to guess. And for me personally, I'd be able to finish a game a little faster. I play to excel, and I can much more quickly make a decision based on known information than on unknown information. You can still plan around unknown and random factors, but it's really difficult. Usually you'll just forego planning and see what the roll of the dice brings you, then take it from there. As I explained, these guesses can make or break a game in a difficult dungeon. In an easier dungeon with a strong class you can probably overcome bad luck, Transmuter always wins for example :)

I don't mind at all adding a bit to the complexity of the game btw, and if that is done I feel it is further justified to make this change. It looks like a bunch of features will be added in Unity, throw in pick pocketing and sneaking on top of that and you have a deliciously complex (yet simple) game that I'd happily pay for.

I doubt we'll ever agree on this point, which is okay with me. I hope you're up for discussing the other points too ;)

[BlackShipsFillt]

I don't think DD is going to change its core at this late point... especially after winning the IGF prize for design...

But a game with your suggestions would indeed make a good rogue-like puzzler, albeit a different experience to DD (like you say, with less constant guessing and a lot less luck).

I would think that in that case, without the exploration as a resource, a few more complexities should be added. I mean DD is quite complicated... but if all the available monsters are known then there may need to be a few other factors to make predicting the correct monster to fight a more nuanced decision.

Ie. If which monster to pick on next is the only problem the player is facing, then I think that factors like chaining monsters and more effects upon leveling up or on killing certain monsters could be emphasized.

Actually, let me say here, I am 100% against any puzzle game that involves no luck at all, if every single choice can be made correctly then I may as well be solving sudokus. What I enjoy is finding optimal solutions that can be applied to partial information about a set of goals and then testing my predictions against the results.

I'm certain that you, Catwalk, can't be against luck entirely, otherwise I can't imagine you enjoying DD at all... but what I'm guessing you want is more information in order to make better predictions that don't feel as lucky.

I've actually been thinking about making a Desktop Dungeons inspired game ever since DD was first made public... though the thing I would change about it would be to make it real-time and time itself a resource (a lot like half minute hero in fact). Because in truth I'm a secret fan of real-time gameplay.

[Catwalk]

Because in truth I'm a secret fan of real-time gameplay.

Seems like you're a very different sort of gamer than I am :) And nothing wrong with that, of course. I generally enjoy turn-based gameplay, being able to play at my leisure and pondering decisions for as long as I like. I think a RT based game would have to be dumbed down even further in order for it to be fun to play (Diablo as an example, way dumbed down but very popular), else you'll just be stressed out.

I don't think DD is going to change its core at this late point... especially after winning the IGF prize for design...

I actually do think that the core functionality of the game stays the same with my proposal, and it upholds the good aspects (IMO) of the concept of having a fixed amount of resources in the game.

I would think that in that case, without the exploration as a resource, a few more complexities should be added. I mean DD is quite complicated... but if all the available monsters are known then there may need to be a few other factors to make predicting the correct monster to fight a more nuanced decision.

I think that having monsters guard stuff (along with multiple ways of dealing with monsters, teleporting monsters away would also become a lot more interesting and I notice that this is already being emphasized in Unity) would go a long towards accomplishing that, adds another major factor to decision making. And to me, DD is very much a leisurely tabletop game. Not having to think as much as in other strategy games (but still more than normal rogue-likes) is greatly appealing to me.

Actually, let me say here, I am 100% against any puzzle game that involves no luck at all, if every single choice can be made correctly then I may as well be solving sudokus. What I enjoy is finding optimal solutions that can be applied to partial information about a set of goals and then testing my predictions against the results.

I'm certain that you, Catwalk, can't be against luck entirely, otherwise I can't imagine you enjoying DD at all... but what I'm guessing you want is more information in order to make better predictions that don't feel as lucky.

Defining luck is tricky, people often mean different things when they use the word. I'm not opposed to randomness in games, for starters. And used properly, I generally prefer some element of randomness to none at all. Chess never appealed to me, for example. I'll give you two examples of randomness, one that I like and one that I dislike:
1) Keywords. This is a Danish game which plays little like charades. The basic concept is that you roll four 12-sided dice with letters on them, and using those four letters as starting letters you make up to four words that will help your partner guess the word on your card. Some rolls are better than others, but it is always possible to come up with good keywords if you're creative enough. One of the plays I'm particularly proud of was using the words "Gives Wheels Nourishment" to help my partner guess the word Gas Station (category was place). This is an excellent use of randomness and I enjoy that game greatly.
2) Kahootz. This is also a Danish board game which is quite similar to the above. Each player has a card in front of him (which he can't see) with a single word to guess. The player left of him rolls a pair of dice which determines how many keywords he gets. He then uses any number of words to help that player guess the word. If the player at the bat has more words to spare after he guesses it, he may use those words on the next player's word. This is a bad use of randomness IMO. Rolling a 12 is infinitely better than rolling a 2. A player can get a decisive advantage through good rolls that can't be negated through skill. No matter what, rolling higher is better than rolling lower.

These examples demonstrate well how I feel about luck, but they're a little difficult to transfer to a single player rogue-like puzzle game. Arguably, luck is much less of an issue in a single player game since you're only competing against yourself. If it was just a question of some maps being harder than others, I wouldn't have a problem with that. My problem is when you have to resort to guessing and can't really formulate a proper plan. Changing the regeneration system isn't the only way to remedy what I see as a critical flaw in DD. If gods are changed to make it much less crucial to sign up early, a major luck component has been removed. I also find myself enjoying enhanced visibility a lot as it gives you a much better basis for making decisions. In fact, Transmuters + Half Dragons + Vampires are much less luck oriented than the other characters and I enjoy these the most. I suppose it's half a solution to remedy it through character selection, but I'd much prefer not to be stuck with 3 classes to have fun. If glyphs are balanced better (which they will be) and spread out more evenly (which seems to be happening already), that will also mitigate luck somewhat.

[BlackShipsFillt]

Seems like you're a very different sort of gamer than I am :)

:) I like all sorts of games... but I like making certain kinds. I guess I'm part animator at heart so real-time speaks more to me.

I think a RT based game would have to be dumbed down even further in order for it to be fun to play (Diablo as an example, way dumbed down but very popular), else you'll just be stressed out.

Although I don't think it'd appeal to you, Half Minute hero isn't exactly dumbed down... it is slightly more forgiving, but even so it is in fact a lot more stressful, and strangely that is part of the appeal (to a certain audience).

But speaking of audience I've debated with Nandrew about randomness before, about his Onslaught of the Electric Zombies, which I thought had too much randomness (of the kind you don't like that makes winning/losing sometimes something outside of the players control).

While Desktop Dungeons has much less of that sort of randomness, it still retains some at it's core. (as you have pointed out)

What I find very interesting is that audiences don't seem as adverse to this sort of randomness as myself (and certainly you). What you term a "critical flaw" in fact seems to drive some of the success of the game.

It really helps DD that uncontrollable randomness is a choice, in that players can choose to receive more information by choosing classes that have scouting abilities, or choose to have some other advantage but be at the whim of the game's layout.

I expect that without the choice to be able to reveal more players would not forgive DD for it's randomness. But as it stands entering a Desktop Dungeon is often a great gamble between more power or more information, which is something unique to DD, I don't myself like more gambling heavy play styles, but I think I can see the appeal for others.

Studying DD for me has been a bit of a schooling in randomness, it does a lot of things beautifully in ways that I wouldn't have done them and I'm always very interested to see how people respond.

By the way Catwalk... I don't suppose you build games? It would be cool to see you give your idea a shot (and I'm sure the QCF team would be interested as well, as they have been in the past with DD inspired games, although I cannot speak for them of course).

Maybe a nice compromise could be a few more classes that have scouting type abilities that remove the randomness of the layout as a factor driving the player's success? Maybe also a little more predictability of the level design, I mean more defined sets of rules/themes to the generation of the maps that could be learnt by players?

But like you said Catwalk, the desire for less uncontrollable randomness is probably a fringe desire. Don't know... I'll probably keep thinking about ways to make it work in real-time :)

[dislekcia]

Wow, hardcore discussion ;)

First off, I really don't think there's any good reason for us to change the core mechanic that DD is built around. Exploration = regeneration is the major factor that drives pretty much everything in the game. It's not an accident that all the hardest classes to balance are the ones that have differences in their regeneration strategies. The food+rest suggestion has a few big drawbacks, particularly late game when a player has what essentially amounts to regeneration stored up for free. The granularity of regeneration is what makes so many of the choices in the game work. Reducing that granularity means that there are far fewer points where players have to reconsider what they're currently trying to do: Compare knowing that you can get a full regen off a specific "amount" of space vs not being sure a passage will be blocked or not. It's this granularity that lets us experiment with different dungeon layouts and generators, seeing what sorts of different experiences those create is really enlightening.

I think the real issue that you're responding to is actually more about the strategies that you've discovered for success in the game so far: Relying on specific overpowered combos is something we wanted to remove in the full. We've mainly done that via upping the usefulness of all the spells, there are no more instant conversions. Information load is not the major problem in the situations you describe, inflexibility is... Try playing a Wizard with no fireballs or not worshipping a god at all ;) (Seriously, gods are there to change play-styles and add extra hurdles, not as crutches to create powerful combos - the rebalanced gods reflect this much better)

Just the other day I did a run through the freeware, seeing how many games it would take me to get all the classes and races. Managed every class at first try, even with a couple of blocked-in runs where I had to use potions early to get out (and one memorable teleport - blocked runs are much less of a problem in the full BTW). Only started having to retry with the Monk in the Snake Pit and the Wizard through the Factory (couldn't deal enough damage to the SMM the first time around, shouldn't have taken Mystera). Then again, I know the game inside out, but even then I'm not making "perfect" decisions because of the random nature of the game, I'm simply responding in the way that removes the least number of options for eventual success.

Yes, level scumming and gold buildup are things that we need to deal with, but we've chosen a completely different set of strategies to do so: The Kingdom handles all of that really well, whilst giving the player more agency in their own progression at the same time. Choosing which preparations you want access to is rewardingly nuanced, deciding which specific ones you want to pay for before each individual run is even more interesting. Some of the preparations even deal with dungeon information: There are quite a few ways to learn more about the dungeon layout, for example; Or you might be tempted with the various Glyph control options from the Wizard's Tower... It's also not a great idea to make things like dodging deterministic: That's the one random immediate effect in the game and because of that it DOES change play styles drastically. If it were completely predictable, why not simply amortise the damage you'd "not take" out over 5 hits and give the player 20% resists instead of allowing them to scum for dodge hits off small monsters and save the sure dodges for a boss? It's not a viable solution. Then again, I happen to think the Witch's dodge potion is ****ing awesome ;)

The thing is, we're aware of individual differences in play-style more than anything else. We have to keep the game balanced for everyone that plays it. As such, changing something as fundamental as the regeneration doesn't make sense. Although I would be pretty keen to see what you come up with if you do build your own game that follows your ideas. I think you'd find it'd be very different to DD in the end. You'd probably have to include die-based random damage to create situations that are open to player response, for instance.

[Catwalk]

:) I like all sorts of games... but I like making certain kinds. I guess I'm part animator at heart so real-time speaks more to me.
...
By the way Catwalk... I don't suppose you build games? It would be cool to see you give your idea a shot (and I'm sure the QCF team would be interested as well, as they have been in the past with DD inspired games, although I cannot speak for them of course).

I love "making" games, although I've never done anything other than mods for existing games. Unless you count my pool variant, aptly named CatBall :P I just went to shoot some pool today and found a few players who liked it, so it's been a good day. I've made (meaning I've started out, made very decent progress and then couldn't be arsed to complete it) mods for several strategy games: Civilization II, Master of Orion, Master of Magic, X-Com: Terror From the Deep and Colonization. I used to have a dream about actually making computer games, but I'm currently 31 and haven't really gotten far with anything :) So those dreams are shelved for the time being, maybe I'll have my own little game some day when I get old. I love providing feedback though, especially if the game designer shares some of my philosophies about good game design. I spent a good part of the last 10 or so years playing an online game named Utopia, an MMO strategy game whose prime virtue was that the game ran in rounds of 2-3 months after which the game was reset and everybody started over. Between rounds the rules were changed, sometimes a little and sometimes a lot, and the players had a big say in these changes. The game designer did a pretty good of listening to feedback while still pursuing his own vision of the game. If anyone is interested in having me as an unpaid volunteer for an interesting game project, I might well be interested in putting a lot of time into it. Alas, I have no coding skills and no money :)

Although I don't think it'd appeal to you, Half Minute hero isn't exactly dumbed down... it is slightly more forgiving, but even so it is in fact a lot more stressful, and strangely that is part of the appeal (to a certain audience).

I will make sure to try it out!

But speaking of audience
...
too much randomness (of the kind you don't like that makes winning/losing sometimes something outside of the players control).

While Desktop Dungeons has much less of that sort of randomness, it still retains some at it's core. (as you have pointed out)

What I find very interesting is that audiences don't seem as adverse to this sort of randomness as myself (and certainly you). What you term a "critical flaw" in fact seems to drive some of the success of the game.

I agree with this general conflict of interests, and I can't really argue that my ideas would make the game more appealing to everybody, or even the majority. But I do think that good compromises can be made in most cases. Much of the time, what less hardcore gamers object to is micro managing and excessive complexity. Micro managing and excessive complexity often happen as a result of trying to balance games and make them less luck dependent, but it is not necessary. If the objectives of a hardcore gamer can be achieved without it leading to micro managing and excessive complexity, I think chances are good that both camps will be happy. I think that it can often be more constructive to start out with ideals in order to reach a practical compromise later on, rather than curbing your creativity too soon.

It really helps DD that uncontrollable randomness is a choice, in that players can choose to receive more information by choosing classes that have scouting abilities, or choose to have some other advantage but be at the whim of the game's layout.
...
Maybe a nice compromise could be a few more classes that have scouting type abilities that remove the randomness of the layout as a factor driving the player's success? Maybe also a little more predictability of the level design, I mean more defined sets of rules/themes to the generation of the maps that could be learnt by players?

I'd like to optimistically turn this notion upside down: Rather than balancing the game around the interests of casual players and then making exceptions to account for the taste of hardcore players, I actually think games benefit from being balanced around the taste of hardcore players and then making exceptions to account for the taste of casual players. I think Starcraft is a good example of this. It's a complex and demanding (and incredibly well designed) RTS which has a large level of complexity when played multiplayer against good opponents. When there's a skill difference, you don't just lose you get completely floored. But casual players can have immense fun with it in single player mode, or against opponents their own level. Blizzard put a lot of effort into making it a great game for players competing at the highest level, and it has attained a huge hardcore following as a consequence. The same can be said for NetHack, it's a fairly complex rogue-like which definitely caters to their hardcore fans. Maybe they forgot their casual audience a bit, a tutorial or a newbie class would be nice. I'd like to see the same for DD, basing classes around hardcore gamers but making a few classes which are easier to handle for beginners and casual players yet subpar for hardcore players. As long as casual players can have some positive initial experiences with the game, I think they'll consider it appealing that there are areas of the game that are outside of their grasp and which will require a fair bit of mastery to fully enjoy.

That turned into major rambling, so I think I'll end this post here :)

[BlackShipsFillt]

Wow, hardcore discussion ;)

lol. Catwalk has some very engaging thoughts.

Although I would be pretty keen to see what you come up with if you do build your own game that follows your ideas. I think you'd find it'd be very different to DD in the end. You'd probably have to include die-based random damage to create situations that are open to player response, for instance.

I think that things that lend themselves to losing/winning by chance are a nono for the Catwalk-type audience. But I think something would have to be added to make it more granular. (I'm liking this term)

Even without the exploration == restoration feature there would still be some unknowns, like players would not know what is behind the next monster they fight. But this is relatively trivial compared to the amount of unknowns and their importance in the current Desktop Dungeons.

And like you say Disleksia, the ability to restore oneself whenever one wants makes the strategy far simpler.

But other things could be added in to a Catwalk version (like Catwalk suggests)... maybe things like factionalism and more alters and story type events that make the strategy more flavourful. Maybe some of the elements of Nandrew's Fiendash could become strategic features (like chaining and spending godly favour). Another idea is to include some kind of crafting that allows the player to collect items and resources that fall from monsters or are scattered about the level... or have somewhat random item drops like Diablo, or set drops like a Rogue-like, and have a character customization screen. Another idea is to make every movement and attack cost a turn and score the player based on how many turns he/she takes.

Anyway... there are lots of ways to do it, but it won't be DD.

[Catwalk]

And I missed the second response before posting, my apologies.

The food+rest suggestion has a few big drawbacks, particularly late game when a player has what essentially amounts to regeneration stored up for free.

Is this much different from how the game is usually played? You'll always want to store some regeneration for the late game, and exploration matters a lot less once you've discovered enough important stuff (wide monster selection, bunch of shops to choose from, god selected) and it's a very minor sacrifice to leave 30-50 tiles unexplored.

The granularity of regeneration is what makes so many of the choices in the game work. Reducing that granularity means that there are far fewer points where players have to reconsider what they're currently trying to do: Compare knowing that you can get a full regen off a specific "amount" of space vs not being sure a passage will be blocked or not. It's this granularity that lets us experiment with different dungeon layouts and generators, seeing what sorts of different experiences those create is really enlightening.

I didn't quite follow this part, sorry. What do you mean by granularity?

I think the real issue that you're responding to is actually more about the strategies that you've discovered for success in the game so far: Relying on specific overpowered combos is something we wanted to remove in the full.

To a large extent yes, and I agree with your notions about flexibility. The more viable options the player has available to him, the less crucial early luck in exploration becomes.

Just the other day I did a run through the freeware, seeing how many games it would take me to get all the classes and races. Managed every class at first try

I agree that blocked runs are not a major concern, was mostly used as an example of things to avoid. That said, you got me going to do you one better on that one :)

Yes, level scumming and gold buildup are things that we need to deal with, but we've chosen a completely different set of strategies to do so: The Kingdom handles all of that really well, whilst giving the player more agency in their own progression at the same time. Choosing which preparations you want access to is rewardingly nuanced, deciding which specific ones you want to pay for before each individual run is even more interesting. Some of the preparations even deal with dungeon information: There are quite a few ways to learn more about the dungeon layout, for example; Or you might be tempted with the various Glyph control options from the Wizard's Tower.

Contributing surplus gold from succesful runs towards kingdom improvements sounds like a really good concept. Will you be able to use gold from unsuccesful runs, though? If so, the whole point is somewhat ruined IMO.

It's also not a great idea to make things like dodging deterministic: That's the one random immediate effect in the game and because of that it DOES change play styles drastically. If it were completely predictable, why not simply amortise the damage you'd "not take" out over 5 hits and give the player 20% resists instead of allowing them to scum for dodge hits off small monsters and save the sure dodges for a boss? It's not a viable solution. Then again, I happen to think the Witch's dodge potion is ****ing awesome ;)

I'd actually consider using small monsters for safe hits a pretty interesting strategy. You only get to use it if they actually hit you, which often won't be the case when you're ready to fight the boss. And any monsters who do hit you may well do enough damage to not make it worthwhile. I think it'll only be useful on occasion, and it'll require a bit of strategizing to work.

I would be pretty keen to see what you come up with if you do build your own game that follows your ideas. I think you'd find it'd be very different to DD in the end. You'd probably have to include die-based random damage to create situations that are open to player response, for instance.

I'm surprised that you think I'd end up having to add more dice tossing. However, actually making a game is way outside the scope of my capabilities, I just like "improving" on existing games :) And I really don't have much I'd want to change about DD, I think it's really well designed.