Saturday, September 21, 2013

DarkSpyre: Mazes and Dead-Ends

A level whose title encapsulates the game.

I've been thinking about it a lot--you get a lot of time to think when you play DarkSpyre--and it occurs to me that games with navigation puzzles are little more than conceptual mazes. Let's say you start in a room with a pressure plate, a lever, and a teleporter to exit. If you do things in the right combination, the teleporter will take you to the next level. If you do things in the wrong combination, you get kicked back to the beginning, and everything resets.

Your options are thus:

  • Start > Teleporter
  • Start > Pressure Plate > Teleporter
  • Start > Pressure Plate > Lever > Teleporter
  • Start > Lever > Pressure Plate > Teleporter
  • Start > Lever > Teleporter
 
You are thus in a conceptual maze with five potential paths, four of which will lead to dead ends. There's no particular intelligence involved in figuring out the right order; you just have to try them all until you find the right path.

Sigh. I suppose I'm going to have to test all of these.

Now add a few other variables into the mix...

  • The lever might require multiple pushes, and it might be intertwined with the pressure plate. For all you know, the right combination is Start > Lever > Pressure Plate > Lever > Lever > Pressure Plate > Lever > Teleporter.
  • Crossing the pressure plate may not be enough. You may have to put something on it and keep it down.
  • Entering the teleporter doesn't "reset" anything; things remain where they are, so the only way to start over is to reload.
  • Even if you're right, entering the teleporter might not take you to an exit. It might take you to a room that looks exactly the same as the first room, with a different combination of "right" moves, only since you think you're back in the first room, you don't bother to try the ones you've already tried.
  • Instead of one lever and one pressure plate, there are eight levers and nine pressure plates, each located a five-minute walk from the others, so trying even a handful of combinations requires the better part of an afternoon.
  • You may reach the final room only to find a keyhole that requires an item you used to weigh down the pressure plate in the first room.

...and you'll have an idea of what it's like to play DarkSpyre.

Another dead end with no way out.

I get that there's a certain appeal to this sort of game. People who like Sudoku might like it. Sudoku is another type of conceptual maze. You look at a square and figure that either a 1 or a 9 can go there. If a 9 goes there, then either a 4 or a 7 goes in the square next to it. If a 7 goes there, then the 4 must go here, but the 4 can't go there--we've hit a dead end--so we have to return to the previous branch of the maze. Some people can work out the combinations faster than others, but that doesn't change the nature of the puzzle.

Helpful tip...or reverse psychology?
 
I don't like Sudoku. I like crossword puzzles. I like puzzles that require knowledge, cleverness, and lateral thinking. If DarkSpyre consisted of 127 consecutive sphinx riddles, I'd think it was great. As it is, I find the levels tedious and exhausting. I've only played five of them since the last post, and they collectively took me more than six hours.

The half-hour video below shows some of the puzzles that I had to solve through rote repetition. It's not even one of the harder levels. You also get to see inventory, attributes, skills, combat, and other elements of the game. Even though I've already played once at this point, and have taken careful notes, I still get screwed up in a few places.


The enemies have gotten a little harder, but I've only died once. Dying, incidentally, treats you to a graphic of the three gods' heads floating around the screen while they note that the "last champion" has failed and the world will be destroyed.

"We will create a race of beings more talented at repeatedly banging their heads against walls."

The only time I've really been in a tough situation is when all my weapons shattered except for a large "troll club" that left me exhausted from swinging it, and most of my spell points were used up. It's near-impossible to flee from an enemy, especially when your endurance is low (it makes you walk slower), so I ended up feebly punching him until he killed me. On a reload, I just spaced more time between enemies, allowing my spell points and hit points to regenerate, and I did fine. The experience really drove home the importance of having backup weapons--not just one, but three or four. A long sword called "Whirlwind" was great for a while, but it ultimately shattered just like all the others.

About to die.

At one point, I found an invisibility spell and cast it to see how it worked. It predictably turned my character invisible, which made it nearly impossible to fight, because I couldn't see where I was or which way I was facing. Despite the handicap, I painfully navigated an entire level waiting for it to wear off before I realized the way to deactivate it was to cast it again.

An ironic message to get when I can't even see myself.

Overall, despite deliberately creating a spellcasting character (with high "power" and "talent"), I haven't been using many spells beyond healing and the "Knock" and "Zap-Away" spells you need to solve certain puzzles. They take too much power and the points regenerate too slowly.

I just thought this was a cool shot. Fortunately, there's an option to avoid all these enemies.

At this point, I've found three of the five runes I need before the end game, so I guess I'm about halfway done. But it's going to be slow-going from here. The levels are taking longer and longer, and I don't have the stamina to complete more than one at a time without a long break. I can envision DarkSpyre extending for several more weeks. Let's see if Dragon Lord offers any kind of interesting contrast.

81 comments:

  1. When you get to Might & Magic's Xeen series, you will get your crossword puzzle in one of the dungeons. Of course, I forget if it is in Clouds or Darkside at the moment :) When you complete it, be ready for an "Interesting" reward.

    ReplyDelete
    Replies
    1. he already encountered one in Dark Heart of Uukrul. Is it similar with this?

      Delete
    2. As I recall, the crossword puzzle kicks in with World of Xeen, which means you need to have MM4 and MM5 installed together.

      Delete
    3. Speaking of Might & Magic, anyone saw the 10th installment at PAX?

      Was it any good?

      Delete
  2. Sounds like those puzzles are a little too long and a little too evil.

    ReplyDelete
  3. I started doing crosswords recently (The normal ones, not the cryptic ones that you and my Dad like), and they are a lot of fun now that my spelling is good enough (I last tried them when I was younger, and having more trouble with my learning disability).

    ReplyDelete
    Replies
    1. I don't like cryptic crosswords very much. I was tickled enough that The Dark Heart of Uukrul included a crossword at all that I was very positive about it, but I vastly prefer American-style crosswords and The New York Times in particular.

      Delete
  4. Very nice video, enjoyed it.


    p.s. The reason i read your blog is to avoid such moments like walking dead scenarios or some disappointment but "feel" game as whole )

    -K

    ReplyDelete
  5. Congratulations for making a good blog post out of this game, and navigation puzzles in general. That is the greatest accomplishment you can take from this I think, making good posts out of not much!

    ReplyDelete
  6. This seems to be a game for me. Maze puzzles without dying often. But then I also enjoy sudokus.

    ReplyDelete
    Replies
    1. So do I. But I like crossword puzzles and Hangman/Wheel Of Fortune type of games too.

      Delete
    2. I like sudoku as well, but this type of game would aggravate me.

      Delete
    3. The comparison was imperfect. Among other things, Sudokus can be solved in a few minutes. I can deal with any style of puzzle for a few minutes.

      Delete
    4. Heck, there are some crossword puzzles that had me stumped for days during the pre-Internet days.

      Delete
  7. From the first ten minutes of your video:
    1) The skull clearly told you not to go into the teleporter. Throwing something in it is the only other option, which makes it kinda obvious (and should be your default behaviour with teleporters anyway). It's not exactly fair to complain that the game makes you a walking dead after you did something you were explicitly told not to (especially when it happens for the second time).
    2) IIRC you can Zap Away balls too. And you DO know there should be an object at the end of the arrow, the skull once again clearly told you so. So you just roll it to avoid fireballs and then zap it away - kinda obvious too.

    I didn't watch the rest of it (not really into gameplay videos, esp. narrated), but if for some reason you're deliberately ignoring the hints you get, it's no wonder it looks like a dumb maze to you. The same goes for Sudoku - you usually get enough clues to at least minimize the number of combinations you have to try out, it's never that tedious if done right.

    On a completely unrelated note, why on earth would you want to fight while invisible? Isn't the whole point of invisibility to avoid fighting?

    ReplyDelete
    Replies
    1. (Also, the door hanging between two bushes looks delightfully insane ;))

      Delete
    2. >>On a completely unrelated note, why on earth would you want to fight while invisible? Isn't the whole point of invisibility to avoid fighting?

      I think it was more that the game actually does make you - invisible (no fuzzy or ghost outline to show where you are)? I imagine this makes moving about VERY difficult as of course you can't see where your character is... :)

      Delete
    3. That may be true, but Chet has clearly written that invisibility "made it nearly impossible to fight", which is puzzling.

      Delete
    4. VK, did your dad make this game or something? You're being oddly belligerent in your comments.

      1. Regardless of what hints a game provides, it's an odd game that puts you in so many dead-ends in which you have to re-load. If you listen to the video, I wasn't bitching about it, just noting it.

      2. On the invisibility bit, I did mention that I couldn't find a way to turn it off. Is it so hard to imagine that I cast the spell to avoid one particular enemy but didn't want to avoid EVERY enemy? Fighting enemies is the only way to increase skill levels with the weapons, and it would be a bad idea to avoid all of them.

      Delete
    5. Sorry if I sounded agressive, never meant to. Not being a native speaker, I can be imprecise with my use of language at times.
      I didn't even finish DarkSpyre, got bored around level 10 or something. I do have a soft spot for navigational puzzles (if they don't involve manual mapping, that is ;)) and puzzle-heavy RPGs in general, so there's that ;)

      Delete
    6. I thought the entire idea of being invisible is being able to strike at your enemy without being seen. It's the ultimate combat advantage.

      Delete
    7. ...when the game mechanics allow for it. That isn't really the case here.

      Delete
  8. If you are solving sodukus by exhaustively trying every combination you are doing it wrong in the same way that if you write random words in a crossword you are doing it wrong. Soduku is a logic puzzle, not a try-and-error-puzzle. There is an element of searching for the square(s) where you can know what is the correct number however, in difference to crosswords where you (at least in theory) can guess any word correctly from the start.

    ReplyDelete
    Replies
    1. That's not how I solve Sudoku, and I never said that. Naturally, a skilled player studies the board and hones in on the squares that have a limited number of possibilities first. Whether you're playing it dumbly by trying every possible combination or intelligently filtering down to only the most promising combinations, the mechanic is still the same. It is absolutely a trial-and-error puzzle; it's just that a skilled player can minimize the number of trials.

      Delete
    2. That's not how I solve sodukus. I select a square (I admit this can be a bit of trial-and-error to find interesting squares). Then I check all constraints and mark which numbers are possible through the direct constraints affecting this square. When a square only has one possible number, mark it. Repeat until you have filled in the whole puzzle.

      Note that it sometimes is not possible to do this easily, making you having to guess one of two possibilities.

      Anyway, I don't know why I defend Sudoku so much since I think it too repetitious.

      Delete
    3. That's not how I solve them either. I usually start not with a square, but with a number, that is prominent on the board, and mark all the squares where it cannot apear. Then check the rest against where this number is the only possible one. Rinse and repeat until finished. With easier puzzles that eliminates the trial-and-error approach completely, with harder ones I only need to resort to it a couple of times throughout.

      Delete
    4. I'm pretty sure both approaches yall stated are still "Trial and error" by definition.

      Regardless give'm a break. He's playing this so we don't have to.

      Delete
    5. Not really, it's more about combining several logical clauses with several "true" states into one with one "true" state. It doesn't involve any more trial than a traditional crossword, where you can also have several possible solutions for a given word.

      It's not that I'm accusing the Addict of doing it wrong or complaining about anything, just offering an alternative view. I'm sorry if it sounds like that. Maybe I've been reading too much RPGCodex lately and it skewed my views on how much antagonism is acceptable in a friendly discussion, but I don't mean any offence, honest.

      Delete
    6. I don't see how you possibly solve Sudoku without trial and error. Just because you write down the possible values for each square doesn't mean you don't have to test possibilities. Unless you're doing absurdly easy puzzles, you'll always have multiple possibilities for each square. So you go down a "path" where you test one possibility, note its effects on other squares, ultimately decide it doesn't fit (i.e., reach a metaphorical dead end), and try another path.

      Sure, the same thing happens to some extent with crossword puzzles. If I get a clue like "Halloween animal" that's three letters, and I already have the second two as AT, I know it can be CAT or BAT. I then go down a "path" of mentally putting the "C" in the first square, note its effect on the crossing square, and see if it makes sense. What separates crosswords from Sudoku is the mental process that narrows down the possibilities to CAT and BAT in the first place, or that interprets an even more obscure clue like "October critter?"

      But whatever. It was a minor point in my overall post, and I'm perfectly find with resigning ourselves to regard each other across a gulf of mutual incomprehension.

      Delete
    7. Er... so, Sudoku is a crossword puzzle with numbers and a crossword puzzle is Sudoku with letters, right?

      I think both are actually solved by the method of exclusion with Crossword Puzzles having an extra element of providing hints for each word as there are 26 different possibilities for each box and way more boxes than a Sudoku puzzle.

      When I'm playing Sudoku, I don't think, "hmm... let's try using this number... Uh oh! Wrong! Next Number!". I think "This box cannot be a 1, 2, 3, 4, 5, 6 or 9. So, it can only be a 7 or 8. Let's see if I can solve the other boxes first to see if I can logically deduce which one is it."

      Crossword Puzzles, on the other, requires you to guess the hint correctly. If it doesn't fit or if one of the letters does not overlap the clue letter (printed or answered), it means you're wrong and you'll have to think harder.

      Trial-&-Error should have no place in logic puzzles because, if it does, even a hen (can't say monkey because apparently they're very good at logic puzzles) could do it if given enough time.

      Delete
    8. Chiming in, in defend of the Addict (finally someone, who doesn't like Sudokus!). Ultimately the difference comes down to this. You can write a program that solves all Sudokus in some finite time - just make it go through all possible numbers for all empty squares and see if the result is valid or not. Sure, it can be done by deduction, but it can also be done by trial and error, and if the computer the program in running on is fast enough, it will do it quicker than a human being.

      Doing a similar program for cross-word puzzles is at least far more complex task, if not truly impossible. Sure, we perhaps might have a program with all the known words of the relevant language and it could check all the possible combinations of the words, but it would still come across with cases where it cannot distinguish between the right and the wrong answer, if it cannot somehow see what is the connection between the hint and the solution - and this would require making the program somehow seem to understand the meaning of the words, which is a pretty mighty task.

      Delete
    9. What happens to people like me who enjoys both? (T_T)

      Also, Sudoku has its roots in Pythagorean mysticism.

      Surely that counts for something... right? RIGHT?

      http://www.themasonictrowel.com/books/symbolism_of_freemasonry/files/illus05.png

      Delete
    10. > What happens to people like me who enjoys both?
      It's like mixing matter with antimatter: BOOM!!!

      Delete
    11. @Ilmari: I don't think the program does has to understand the meaning of the words.

      The problem with the crossword solver is that the question is given in a language that does not fit a computer well. If you store all possible answers in a data base and give all questions in a corresponding language, say SQL, solving the crossword becomes much easier.

      Delete
    12. I don't find the "but a computer can do it" argument a very compelling reason to prefer a crossword over sudoku. It reminds me of 10 years ago or so when computers could play chess very well but go not so well so I heard the argument that go was superior; now that go seems to be closer to falling to computers, is it suddenly a worse game?

      Similarly, suppose Watson (the Jeopardy-playing computer) were adapted for crosswords; would that make them suddenly no longer fun to play? Why do we care that a computer can have perfect an instant knowledge of a crossword dictionary? Similarly, why do we care that it can check through millions of possibilities extremely fast in sudoku?

      (Incidentally, Andrew Stuart's sudoku solver has 35 strategies, and doesn't get to "trial and error" until #35.)

      Delete
    13. Another perspective: I usually have to use an eraser while solving crosswords, but (almost) never while solving soduku. I guess the Addict has it the other way around since he has better crossword solving strategies than I.

      Delete
    14. Well, I was not trying to say that Sudokus are essentially worse than crossword puzzles (even if I personally don’t enjoy former). Instead, I was just trying to point out that there’s a substantial difference between the two: a difference I think Addict was trying to convey and which some commenters above contradicted by saying that in playing crossword puzzles we use same sort of reasoning as in playing Sudokus.

      Same goes for my use of programs as an illustration: I did not want to say that computer beating us in a game would make that game no fun (I do enjoy chess, even if quite simple programs beat me). It was just more of an illustration of the difference in complexity between Sudokus and crossword puzzles – you would need far more intricate program for latter. It's still rather easy to find a decision procedure for a sudoku solver. The elements of the game can be described in a simple formal language, so there's no problem for making the Sudoku Solver to recognize what a correct solution is like. Then even the worst Sudoku Solver -program, which for Sudoku with n empty squares goes through all 9^n possibilities of filling the empty squares in a trial and error -fashion, will unerringly and infallibly find the right solution: the answer must be one of those possibilities, so all the Sudoku Solver needs is time.

      The case is far different with Watson and its supposed crossword modification. Watson cannot make do with just numbers, rows and columns, but it must deal with all the vagueness and openness of natural language. Just look at what Wikipedia says about Watson: “more than 100 different techniques are used to analyze natural language, identify sources, find and generate hypotheses, find and score evidence, and merge and rank hypotheses” and it also uses “databases, taxonomies, and ontologies”, apparently just to make sense of what the clue presented to it is trying to convey and to formulate a question corresponding to that clue. All in all, its sources include “200 million pages of structured and unstructured content consuming four terabytes of disk storage”: I am guessing this goes far beyond what is required for Sudoku Solver. Still, if I understood correctly, armed with all this artillery, Watson might still fail to give the right answer (even though it is more successful than human contesters), thus making its decision procedure far from infallible.

      As I am no expert on either Jeopardy or crossword puzzles, I don't really know whether Watson would have to be greatly augmented to make it the Cross Wordcracker. All the examples of Jeopardy clues I've seen appear to have rather clear relation to the question that should be the answer (e.g. it's easy to see why ”Who is George Washington?” is the correct answer for the clue ”The Father of Our Country, he didn't really chop down a cherry tree”) and thus seem still far from the more expert crossword puzzles, where it's often difficult for us non-aficinados to see the link between the clue and the answer (why on Earth is ”nail” a correct answer for ”nothing to hold a spike”?). I suppose the dependence on wordplays and metaphors might make crossword puzzle solving a bit more difficult to simulate than playing Jeopardy.

      Delete
    15. Okay, my last words on the subject:

      1. I didn't mean the term "trial-and-error" as literally as some of you are interpreting it. I wasn't trying to suggest a rote, unthinking process, but rather a systematic testing of potential "paths" to the solution. A skilled solver can quickly limit the number of viable paths through deductive logic, and perhaps if he's skilled enough, he eliminates all but one path right at the outset. I still maintain that some trial-and-error is usually required except in the easiest puzzles. Jason, Andrew's "Sudoku solver" lists "trial and error" as the last step not because it's least important but becuase it's ALL THAT'S LEFT when other deductive paths have been eliminated, and I would maintain in most puzzles, you almost always reach this step.

      2. Sudoku may be a puzzle of deductive reasoning, but it's a puzzle of MECHANICAL deductive reasoning. There is a clear right way and wrong way to solve a Sudoku puzzle. You don't solve it through intuition, or lateral thinking, or creative deduction. You solve it through a very mechanical process. It was the mechanical nature of this process that I was trying to highlight in my comparison of Sudoku to DarkSpyre.

      3. Hence, Ilmari is 100% correct. You write a program--not a terribly complex program--that will solve a Sudoku puzzle almost instantaneously with 100% accuracy. There is no way to write a program to solve crossword puzzles with the same speed and accuracy. The Sudouku bank has 9 potential values; a crossword puzzle bank has millions, when you consider compound phrases, proper names, quotes, abbreviations, foreign words, puns, and so on. Moreover, a Sudoku puzzle can be solved without any reference to any external clues, whereas a crossword grid could easily be filled with words that "fit" but have no reference to the clues.

      4. The argument that a "computer can do it" by itself is not an argument to prefer crosswords over Sudokus. The reason to prefer either over the other is that they require an entirely different skill set, and that the computer can do one and not the other is a reflection of that skill set.

      5. "So, Sudoku is a crossword puzzle with numbers and a crossword puzzle is Sudoku with letters, right?" Kenny, you're a frequent and valuable contributor to my blog, and I like you, so please accept that it is with the greatest affection that I tell you to go %#*@ yourself.

      Delete
    16. "I would maintain in most puzzles, you almost always reach this step."

      Actually, it means the puzzle is badly written. Most puzzle aficionados would call it broken.

      But that segues nicely into your point #2, which puzzle aficionados would also agree with. The hardcore World Puzzle Championship types actually find vanilla Sudoku uninteresting because it has been "solved" so to speak. The various puzzle championships ensure there are lots of original puzzles (or at least variants) in the mix that require fresh logic.

      (If I had to pick one logic puzzle type I'd have to stick with forever, I'd probably go for Slitherlink, which I find much more satisfying than Sudoku and also makes your "concept maze" idea almost literal.)

      Delete
    17. None taken, Chet! I wish I could do that, though... I'd be an A-lister in adult films! Stand back, Ron Jeremy! You might be able to suck yourself but you ain't seen NOTHING yet! XD

      But seriously, people, it's like comparing apples and bananas. Not everyone likes apples and/or bananas. Some don't even like both. Right, I'm going to play Minesweeper from now on.

      Delete
    18. @Addict: Point #2 I definitely agree with. I find it repetitive (which I said in my second post).

      And I also agree that it's the term "trial-and-error" that is causing issues. I still don't understand exactly what you mean by it in this context.

      Delete
    19. I personally dislike crosswords because they're not self-contained; if I don't know the name of an actor or film or whatever that the puzzle asks me for, that's that--game over. I don't like Sudoku specifically--they bore me--but I prefer puzzles *like* Sudoku, where outside knowledge isn't required. It all comes down to personal taste, I guess.

      Regarding the video game, it would drive me nuts to have to reload the game because I got in an impossible situation; some kind of in-game "reset level" option is clearly called for.

      Delete
    20. @bd- Good point, there. I was also thinking about this point too. Sudoku reaches out to anyone with the ability to read Arabic numerals. Crossword puzzles require you to, not only have an extensive vocabulary, but also have an in-depth knowledge akin to that of the puzzle's creator.

      Delete
  9. I suck at both crosswords and Sudoku. I've never been a terribly logical person, unfortunately.

    I passed by Darkspyre ages ago, because it didn't seem like my kind of game. The posts about it as a whole so far confirm that, so by all means, take a look at Dragon Lord now and then.

    ReplyDelete
  10. Very interesting post! Game (and level) designers have a difficult challenge - Trying to make puzzles challenging to players without making them unfair. Likewise with trying to make an intuitive interface vs. providing a tutorial.

    Many game puzzles seem to be as arbitrary and "brute force" as the ones in DarkSpyre. But if they can be solved without trial and error, they are frequently too easy for players. I have a problem with many Sphinx-style riddles in that there are frequently alternate answers that make sense, but games (and usually human judges, for that matter) only accept the "intended" answer. I remember, for example, typing "xavtug bs qvnzbaqf" in Wizardry II and getting frustrated. The only answer it would accept was "gur xavtug bs qvnzbaqf". Grrr! (Or "gur" in rot13.)

    I never did get past one or two of the riddles in Heroes of Might and Magic. I would come up with four or five answers that all met the conditions of the riddle, but none was acceptable to the game.

    This is actually a variation on the general problem of bringing the feel of a tabletop RPG to the computer. One game master friend of mine came up with four different solutions to a "locked room" puzzle to make sure the players could solve it. They did - with a fifth approach. That only works on a computer if all of the solutions are emergent from game mechanics (e.g. Portal).

    Puzzles looking for the player to match a particular solution are usually arbitrary and can require trial and error to solve. Personally, I try one or two things, then look up the answer on the Internet. Every once in a while I read the solution and say, "Oh! I could have figured that out! I should have tried more things." But more often, the solution leaves me scratching my head and glad I didn't spend hours trying to guess the unguessable.

    ReplyDelete
    Replies
    1. Or the most common problem in RPG:s and Adventure games (at least in my experience): Oh, I had missed to pick up an item 3 gameplay hours ago.

      Delete
    2. The Quest for Glory series of games doesn't hit you with that often, but King's Quest is famous for making games unknowingly un-winnable if you failed to do something hours ago.

      The early Space Quest games were even worse than King's Quest. I think it's the second Space Quest game where you somehow end up or start in a swamp on a planet teeming with enemies.

      Did you remember to "search" the swamp in the beginning of the game, despite no indication that you should have? No? Well, too bad you'll be stuck at the end of the game, lacking the one item that was in the swamp.

      Ugh! I think the Quest for Glory games avoided all those issues, except for the third one, where you could sort of bug the game out if you had a certain combination of skills. Something about making one of the tribe's initiation rites impossible or something.

      Delete
    3. Corey, it's nice to have your perspective as a developer, but I think perhaps you're setting up a false dilemma. There's no reason that a game can't offer riddles and alternate solutions. Just as your text parser accounted for every possible variation of verb--LEAP, JUMP, VAULT, and so forth--there's no reason that a riddle couldn't account for every possible variation of the answer--THE MOON, MOON, MOONLIGHT, and so on.

      Oddly, I find walking death slightly more forgivable in adventure games simply because they tend to be relatively short once you know the puzzles. It took me 17 hours to figure out all the puzzles in Beyond Zork and 1 hour to win with the final character.

      Delete
    4. I was thinking, you could do that with OR statements the old fashioned way, but Regular expressions would do it even better. I wonder if you could use some of those weird twister mathematics to make it even more flexible?

      Delete
    5. Addict - Of course we can handle simple synonyms and variations. My point is that a computer game only handles actions the designer and programmers considered. For a riddle, perhaps "Night" and "Death" might both be reasonable answers, but maybe the riddler didn't think of one of them.

      We went out of our way to provide reasonable synonyms and alternate phrasing in Hero's Quest and Quest for Glory 2 (and the parser handled minor variations such as plurals), but I'm sure many reasonable phrases were not handled. Point and click adventures solve this by only allowing responses the game is prepared to handle, but in return they take away the "Ah ha!" of guessing a challenging riddle because the answer is right there in the dialogue menu.

      Lori and I try to keep our point-and-click games challenging by tracking a large number of game actions. If you haven't gotten the secret password by listening in on a conversation or intercepting a message, we don't allow you to guess it. That's actually good game design discipline as well. It forces the designer to include the information needed to solve each puzzle in the game, rather than relying on outside knowledge. (There's a puzzle in Star Wars: the Old Republic that requires knowing a piece of Star Wars trivia that pretty much nobody would know. Real players will solve it either by reading a walkthrough or by trying every possible answer brute force while their characters slowly burn to death.)

      Delete
    6. Thanks for the reply, and I get what you're saying. Perhaps I went down the wrong path when talking about sphinx riddles. But there are other ways to introduce creative and intuitive puzzles that have less opportunity to screw the reader.

      Skyrim has a few decent examples. The first time you encounter one of those turn-the-pillar puzzles, you have to figure out the solution by intuiting that a collapsed section of the wall held the middle pillar in the puzzle. I'm not suggesting this is Einstein-level creativity, but I still though it was fun. There were a few others of this nature, where you had to figure out the sequence from something you read in a book. If you got frustrated, you could always just try all possible combinations--there are only 27 with three pillars of three choices each, after all.

      Other games have dealt with this by reducing experience rewards, or providing combat alternatives to the logic puzzles--The Dark Heart of Uukrul and Wasteland come to mind.

      Delete
    7. This reminds me of unit testing in software development. If you write your own tests, you can prove that your code works the way you envision it. When someone does something you didn't anticipate, your code turns out to be less robust than you thought. Same idea; you can always account for what you think of. You can't account for everything your users think of.

      Delete
    8. Not to sound too antagonistic once again, but personally I found Skyrim's puzzles isultingly easy. They always have a solutions explicitly written somewhere nearby, often in the very same room. The only decent one I remember was the Oculory in College questline.
      But that has generally been the trend with modern mainstream games: they either have none or have them so easy, that a 4-year old could solve them. Surprisingly enough, the only AAA game of late that offered somewhat interesting puzzles was Dragon Age: Origins (namely, the shapeshifting one in the Fade). If you could get to them through hordes of identical trash mobs, that is.

      Delete
    9. I may be completely wrong, it is only an opinion.

      But games like Skyrim have 500 puzzles. Simple, yes, mostly. Older games had 10 or maybe 50. perhaps they tried to provide the same amount of game time or difficulty regardless of the amount.
      There is probably no perfect median. I, however kind of like a small puzzle over and over with a few hard ones mixed in (in between death duals with dragons and secret meetings with under-folk) rather than 10 or so super hard, almost arbitrary puzzles relying on prior knowledge of the canon specific to the game.

      perhaps a mix is best. Better than best, however, is allowing multiple paths through them. a HARD puzzle or a HARD fight. I know it may be difficult to produce, and possibly impossible but therein, somewhere, lies a middle-ground.

      I'm only 1 person though and can imagine how difficult it must be to program fro millions. (I have written a few very short BASIC games and can imagine how a 400 line code could develop into a 4000 line code just to fit in all the "maybe's".

      Anyway it's GREAT to see such thoughts floated about and I've come to expect nothing less from this blog.
      CHEERS!

      Delete
    10. I will admit; two puzzles in Skyrim stumped me. One I never figured out, I just brute forced it from the one clue (A bunch of pillars in the plains west of Whiterun.) I found one of the right symbols based on the one hint I found, then brute forced the other two.

      The other one had a book there, but I didn't notice the theme around the switches, which was very duhhhhhh once I looked it up.

      Delete
    11. @Ryan It's totally possible. Look at Dark Disciples 2 (first one too, but to a lesser extent). It's a freeware game written by a single person in his spare time, yet it's of similar scope if not to Skyrim then certainly to Arcanum. And it's filled to the brim with tons of interesting puzzles, many of which offer alternative, more combat-heavy solutions and no two look the same. Of course it's easier to do in a top-down engine than in 3D, but if such a design is possible for a single gamemaker, large teams should have no problems with it.

      Delete
    12. Cool! Another person who played the game! XD

      I was also rather active in his forum last time as well... but real life beckoned back then and I had to stop my gaming lifestyle for a while.

      Delete
    13. It's a shame it isn't better known. The first one was a kinda niche hit back in the day, so it's really puzzling why it didn't work out for the much better sequel. Guess there are too few people left who can stomach top-down graphics.

      Delete
    14. When I said that Skyrim had good examples, I didn't mean they were particularly challenging, just that they required a bit of creativity and didn't trap the user.

      Delete
  11. You have far more stamina than I do. I bought this game and tried to play back in 1993; because, it was in the bargain bin and because I was looking for something to do. It didn't take me long to realize that this game is terrible.

    I had hoped that your review of it might change my mind, and 20 years later I would dig it out and load it up in DOSBox and give it another run.

    Instead, you are confirming what I already remembered about the game. It was terrible then, it's terrible now and while it's not a forgotten classic, it should definitely be forgotten.

    ReplyDelete
    Replies
    1. I'm not prepared to go quite so far--it has its moments--but I'm not surprised there aren't any passionate defenders of the game on my blog.

      Delete
    2. When the sequel comes along, I'll defend that one. Vehemently too, I might add. XD

      Delete
    3. The sequel is great (just started a replay, as I never finished it back in the day) but there's one thing in it that really needs defending: controls. That combat facing... Ugh.

      Delete
    4. Yeeeaarrgghh!!! Thou doth not talk sh!t about The Summoning! Defend thyself, knave!

      Anyway, what's wrong with the combat? I had no trouble with it. Veil of Darkness (built on the same engine), on the other hand, had really small characters and objects that are pretty hard to click on.

      Delete
    5. Facing. At (most) times it's extremely hard to determine from which side the enemy is attacking and turn in that direction (controls aren't very responsive in that regards). And that's in melee. Ranged combat is even worse, which makes it completely useless.

      Delete
    6. Oh, that! Okay, the trick here is to time your movements so that you don't move your character the same time as the AI. It somehow eats away your input when it is moving at the same time. Once you get that movement down, melee becomes much easier. As for ranged, the AI is actually pretty smart by not moving in a straight line towards you. Use narrow corridors to your advantage. That, or you'll have to anticipate the AI's movements.

      Delete
    7. Moving is the easier part of this, determining where to move is harder. I seems like enemies only have sprites for 4 main directions, but can attack diagonally. I ususally end up taking a couple of hits while trying to tell whether the enemy is to the right of me or to the lower-right.

      Delete
    8. Hmm... Have you tried the DMR (Dungeon-Master-Run) tactic?

      You start off at least 1 square distance away, close in, hack the bugger and step back before it retaliates. And if it closes in when you step back, you immediately attack it again with your off-hand weapon like a free attack of opportunity.

      http://www.giantitp.com/comics/oots0216.html

      Delete
    9. Oh, I'm managing quite fine (usually through "rest in a safe place for half an hour" tactics, but that's for the better since I get to get some work done inbetween), I was just saying that the controls in the game aren't exactly an epitome of convenience ;)

      Delete
  12. Alright, since everyone is talking about puzzles here: Back when I was in grade 5 I did a series of logic puzzles where you got five hints (Janus has red hair. No one with red hair is left handed. No one has both blue eyes and blond hair) and then a grind of circumstances and you had to find the killer or whatever.

    Anyone know what this game is called, since you all seem to be game snobs?

    ReplyDelete
    Replies
    1. You mean logic grid puzzles?
      Like those:
      http://www.braingle.com/Logic-Grid.html

      You're welcome.

      hauns, Game Snob. ;)

      Delete
    2. That reminds me of a shareware logic puzzle game I played: http://www.kaser.com/sherwin.html

      Hard to believe they're still selling it.

      Delete
    3. Thanks dahauns! I just finished one, that is exactly what I was talking about.

      Delete
  13. They're called Zebra Puzzle or Einstein's Riddle: http://en.wikipedia.org/wiki/Zebra_Puzzle

    ReplyDelete
    Replies
    1. I actually really like this kind of puzzle. Even though they're both called "logic puzzles," I don't see it as the same thing as Sudoku at all. The Zebra Puzzle involves creative thinking and inference, whereas Sudoku is just rote deduction.

      Delete
  14. Ok, I know it doesn't really make sense anyway that some gods or evil lords construct giant dungeons/mazes/spires, but... isn't it illogical that you have to use a lever twice or more? I mean, the ordinary lever only has two states with one being the default. So you should never have to use it more than once in a row. Yeah, that bothers me...

    ReplyDelete
    Replies
    1. Not always. Some levers have three states: Neutral (Straight up), positive, and negative.

      Delete
    2. Ok, that's true, so it should take at most 2 tries.

      Delete
    3. That depends, with N+2 you could have each position change a state. So if you start in state 1 + N, then set it to +1 position, you go to state 2, while if you go -1, then you go to state 3.

      Say for example that your lever adds a set amount of pressure to a device that unlocks the door when you hit 3 PSI. You start at 2 PSI, each up pull adds 2 PSI, each down pull removes 1 PSI of pressure.

      Delete
  15. I always find it amusing that casually mentioning something so benign, your Sudoku bit for example, can set off a firestorm of posts.

    ReplyDelete

I welcome all comments about the material in this blog, and I generally do not censor them. However, please follow these rules:

1. Do not link to any commercial entities, including Kickstarter campaigns, unless they're directly relevant to the material in the associated blog posting. (For instance, that GOG is selling the particular game I'm playing is relevant; that Steam is having a sale this week on other games is not.) THIS ALSO INCLUDES USER NAMES THAT LINK TO ADVERTISING.

2. Please avoid profanity and vulgar language. I don't want my blog flagged by too many filters.

3. Please don't comment anonymously. It makes it impossible to tell who's who in a thread. Choose the "Name/URL" option, pick a name for yourself, and just leave the URL blank.

Also, Blogger has a way of "eating" comments, so I highly recommend that you copy your words to the clipboard before submitting, just in case.

NOTE: Spam has gotten so bad lately that I've had to turn on comment moderation for posts older than 10 days. I apologize if it takes a little while for your comment to appear.