From a design point of view, the whole story is actually much more interesting than "
some mechanics are fiddly"
and "
I like/don't like this element"
.
Fundamentally, the whole everlasting conflict from: serpent mechanics to movement order, unlimited undo to room designs, etc... has to do with one particular idea:
static versus
dynamic nature of the game, puzzles, and puzzle elements. It's hard to give a precise definition of it, but roughly, static puzzles are, well, static, which mostly involves sitting down and thinking through the state of the game, and most importantly, your actions (input) is mostly decoupled with the response of the puzzle; Meanwhile, in dynamic puzzles, the puzzle itself gives complex responses to your actions that the best you can do is to make out a general strategy, and then deal with the details as you go. A classic laser ray game (Chromatron, Aargon, etc) is a definite example of a static puzzle, while Chess and Tabletop RPGs in general are very dynamic there is no way to surely tell what'd happen a few turns ahead (except for chess endgame stuff).
(Or, in DROD terms, a singleplayer
Snakes to the Death is static, while multiplayer is incredibly dynamic.)
Now, why does it matter? You see, back in the days of AE and JtRH, it's pretty obvious that the game's supposed to be incredibly dynamic from ground, and everyone, from players to devs, are
expecting the game to be so. For example, let's take a look at the first room of
Contest Hold back in 2004:
Contest Hold : Contest Level : Entrance. It is a glaring example of a very dynamic puzzle: a hack-and-slash room with a huge brained horde that requires very delicate movements to hold your ground enough before you hit a wall on your back.
KDD and JtRH are literally filled with these kind of puzzles, and the dynamic nature is even present in the meta-game, in the form of undos: still remember
this post back in AE era? Undo functionality is greatly frowned upon in dynamic puzzles because
the whole point of dynamic puzzles is to be a great tactician and choose the best moves from your great experience of handling similar situations; having undos would undermine this whole premise. It's just like how you don't have undos in chess or tabletop RPGs. I suppose this might be a result of Erik working out the original DROD with a big influence from tabletop RPGs, but instead of relying on RNGs to give the "
be the best tactician"
kind of difficulty, very simplistic AIs were involved instead, and they move according to an intrinsic order just like how you take turns.
However, during the course of TCB, the overall direction of DROD has clearly drifted: it has become more static (from extremely dynamic to slightly more dynamic than static). The new elements are a mixture between static (e.g arrow rotator tokens) to dynamic (e.g waterskipper nests), but the one most notable example is, of course, pressure plates, which fundamentally changed the notion that only the player can interact with room mechanisms, and it gives sooooooo much more possibility on the static puzzles one can make. It's also reflected in the room designs: there are relatively more lynchpin rooms. Of course, TCB didn't quite get it right, and kind of fell short on the static side of things; GatEB got a much better hang of this, and TSS has a incredible mixture of both sides of the spectrum.
Also, as a result, undos are being less and less frowned upon, and now we have UU(unlimited undos) in TSS. The whole argument of UU is
still around the premise of static vs dynamic puzzles, but since the metagame's more inclined to being static now, so is the amount of undos players can legitimately have in game.
(As a side note: TSS elements design is... very interesting. Instead of trying to be either static or dynamic, it instead gives lots of elements that are either suitable for both types, or are more like an extension to the tools available for the player/architect. It makes designing puzzles way more versatile.)
would make snakes work as follows:
1) If you are in the same row or column as a snake it turns to face you (same as it is now).
2) A snake otherwise continues to head in the direction it is facing unless it can't in which case it tries turn clockwise, and failing that anti-clockwise.
3) Brained serpents work as they do now, heading straight for you via the shortest path.
So, serpents. While many people have argued about the peculiar H/V and NESW preference and/or wished for a type of serpent that works less chaotically, I'd steer away from it, and instead argue that a serpent that doesn't have such preference
would have been impossible to have existed back then because back in the days, every element has been designed to be dynamic, and intended to be used in a dynamic way. You're supposed to adapt to the snakes' preferences, and then play with and dance with it. A snake which walks along a perfectly predictable path for dozens of turns before it kills itself at the perfect spot is... unthinkable of back at the era.
Well, until larrymurk appeared and made rooms everyone banging their heads at. larrymurk's architectual design requires lots of pre-thinking and going through possible states of the game, and it's not something the people used to dynamic nature of designs have prepared for. Honestly, it's a truly amazing feat that he made so many such rooms back in JtRH era.
(Also, larrymurk is... kinda well known for making horrendous horde rooms, so there's that.
)
Similarly, goblins, wraithwings and brained roach queens, with their notoriously fiddly AIs, are something which can only happen during AE/JtRH era. Can they be tamed? Mostly: Rheb is famous for lots of goblin manipulation rooms (and precise sequence moves in general), while larrymurk's contribution, Mother of All Roaches in TSS, has rooms that turn brained roach queen movement quirks into lynchpin puzzles. (Meanwhile, wraithwing hordes seem to still resist being tamed into a well under control puzzle.)
Incidentally, there is a similar idea made into a famous custom element, and that is bent-tail lemmings (featured in TSS and Red-XII's hold Rodents of Unusual Size (ROUS)). And a vast majority of the rooms are very static, with the exception of a few rooms that involves baby lemmings and/or blue lemmings. Lemmings and bent-tail lemmings are intrinsically very static.
Now, onto tarstuff. Oh man, this is truly a classic ironic example. KDD tarstuff rooms have all been "
it's tar all over the place and you have to cut through it so you can do specific things"
, and sometimes it's worse as there are things hidden under all the tar. JtRH added mud, and tarstuff doors which only toggles with all tarstuff is gone. The usage of tarstuff back at that point is mostly dynamic - tar and mud wrestling is commonplace, and the research of tarstuff theory, and the static aspect of tarstuff, is pretty underutilized:
this is pretty much the only key thread talking about it around that time, and the knowledge around tar is mostly "
2x2 is unclearable, 4x2 is unclearable too, so is any rectanglar tar with dimensions of even length"
... and then gel appears in TCB. Gel is so annoying and hard to deal with, that some theoretical knowledge on what is possible is necessary so as to not end up trying to cut up gel to achieve an actually impossible state. And then we know the rest - lots of and lots of brutal static tarstuff rooms have been made, and the most famous of them being, of course, Zch and Kallor, with Gigantic Jewel Lost. (There are even exploratory secrets with mathematical notes inside GJL!) Tarstuff is a perfect example of a
subversion of an element that was intended to be dynamic. And we can all agree that tarstuff is one of the greatest invention ever.
(Incidentally, golems are also among this type of element, which was intended to be dynamic but, in the end, turns out to be suitable for both static and dynamic puzzles.)
(Also incidentally, GJL has some mathematical puzzles using serpents and complex pressure plate/door mechanisms.)
Well... You could have some kind of simple, geometric movementorder. Like the northernmost row moves first, then the second row, and so on. Just like reading 
That way the player would always know what would happen.
Now, onto movement order. Movement order being intrinsic to the entity itself is a mechanic with an incredibly dynamic nature. If movement order is changed to coordinate-based instead, which is incredibly static in nature, two very obvious things will tend to happen - horde rooms become much less viable (because every pile of horde always behave the same) and would've seen less use as a result, and also that it becomes something that only a few would ever
try to utilize as a lynchpin. It's much less versatile too, of course. So...
Movement order: I'm not sure if sworded monsters still have to move before everyone else? I'd remove that restriction.
...Yeah, there's probably no reasons to keep this either.
Being too dynamic/static has its problems. Being too dynamic makes things incredibly fiddly and resorts to magic sequence rooms, a notorious example being
Journey to Rooted Hold : Nineteenth Level : 2 West; Meanwhile, being too static greatly limits the potential of a mechanic or an element (for example, just *ahem* look at 90% of the custom elements). And you can't just "
add in"
dynamicity by arbitrarily fiddling with the rules, which is about equivalent to doing a random walk, that makes things very messy and decoherent. Good dynamic stuff are predictable, easily under control and mostly coherent, and this can be achieved by, like, architects intentionally making one type of monster move before the other, and informing the player about this.
(Of course, I do agree that there should be a better way of adjusting monster movement order in the editor. Currently you can only change their order by replacing the monsters.)
Pathfinding. Pathfinding is also something that is incredibly dynamic in nature, that could only be conceived back in the early period.
Of course, while dynamic pathfinding is good, obscure or outright unpredictable (read: bugged) pathfinding is not. (Halph, Slayer, citizens, builders and soldiers/stalwarts. Also, arbitrary brain-invisible obstacles.) They are the polar opposite of good dynamic mechanics/elements - for being actually uncontrollable, they become untameable.
(Side note: Gentryii chain straightening is borderline unpredictable)
I can go on and on about this whole idea of static versus dynamic nature, but I guess it's more than enough to illistrate my point.
So, is this change of design choice of DROD from dynamic to static good or bad? I won't even try to answer this question (as this would just open the cans of worms everybody's familiar of), and no matter what, the change has been made, and once a change is done, it cannot be undone, which effects will echo through the stream of time to bring an irreversible impact on how everyone think DROD is and treat DROD as. I think it's apparent that DROD players are now at a position more inclined to staticity, in that we prefer banging our head against tough rooms that requires lots of thought rather than tough rooms that requires lots of efficiency and tactical movements. And personally, I'm incredibly static when it comes to conceiving puzzles - I'm literally incapable to come up with any horde rooms at all. My mind is just unsuitable for that.
However, this is also what makes DROD unique and special - it has gone through both phases of being dynamic and static, with elements scattering all over the spectrum of both sides. And TSS provides much more freedom to roam around this spectrum. This is, I think, the biggest reason why DROD is so great. The fact that the conflict exists simply means DROD is capable of achieving two seemingly distinct types of puzzles, and is it not beautiful to witness such a crossbreed which we would all proudly refer to as "
the best puzzle game ever in the world"
?
So, what does all this mean? It means two things:
1. For what it's worth, the evolution of DROD's design choice, while being strange and has caused lots of dividing opinions, is actually what makes DROD truly great as it is right now. Instead of arguing about whether we should retcon earlier elements to the thinking of nowadays, it's perhaps more beneficial to analyze the change of the nature of puzzles DROD has gone through, and design the next batch of elements in the next DROD with this in mind, so as to achieve a even greater DROD.
2. If you're scripting custom elements, instead of relying on intuition and asking "
Does this
seem interesting?"
, following by a bunch of trial and error before figuring out the element actually doesn't have much potential, maybe you can think in terms of how static/dynamic it is, and how to twist or subvert it, or mix up other elements with varing degrees of staticity/mobility to make interesting puzzles. It's always better when you have some design vocabulary of puzzle elements, since that's how you communicate with others (and yourself) about your puzzle style!
And I hope that somewhere in the world, a drodfolk would be enlightened by this wall of text I wrote in 3 hours. I hope.
[Last edited by uncopy2002 at 12-19-2016 07:44 PM]
