We are going Map Crazy.

One of my primary design goals, all through development, has been to keep the island spatially small, so that there isn't a lot of wandering through empty or pointless spaces. The idea is to take the best parts of an open-world game, involving choice, exploration and discovery, but compressing these into a maximally dense experience.

That has been going well, but there's an interesting phenomenon where, as we add more detail to the world, it starts to feel a lot bigger. When something is a mostly-featureless plain it is easy to feel like it's small, but once you are passing tiny rocks and plants that are set up very carefully, your brain has more scale reference. So there have been a couple of times during development where we stopped and said "Hey wait a minute, things are just too big."

Last week that happened again; we were looking at two different parts of the island on the same day, and both of them seemed much bigger than we needed. So, we decided to go back to the overall island plan and try to compress it further. We had a discussion about this on Thursday, then the architects talked it over; you see the result of those discussions in the image above. These maps are a handy guide to helping us figure out what we can cut and how we can scoot existing areas around to pack them a little better, and to better-define the areas between them.

This is a little painful, since we almost had the island back into playable shape, and this knocks it back out of playable shape for a little longer. But my hope is that we can just make a mess and then clean it up very quickly.

In other news, there had been a puzzle that I needed to design for two years -- in the sense that I knew I needed to build the puzzle in order for a certain area to be complete, and I knew what that puzzle vaguely had to be about, but didn't have any good ideas on how to build it. Last week I had a breakthrough on that and built it very quickly; it's now one of the best puzzles in the game. So things are coming along well!


  1. Hmm. Takeaway thought from this: size (or at least how we perceive size) is based as much on object density and detail as it is on physical geometry. Interesting.

  2. Is you work flow consist of resizing the terrain and the objects/puzzles are automatically repositioned based on their location on the terrain, or is this a completely manual process? I guess, even if it is automatic, you still need to tweak things a lot to make them look good but I’m curious how you mitigate having to reposing so many objects. I’m sure things like buildings have some kind of hierarchy but all the rocks, plants panels are what I’m wondering about.


    • It is a manual process. Automating it with good results would be equivalent to solving the hard AI problem.

      Stuff like grass automatically plants itself at ground level. Other smaller objects like rocks, bushes, etc, used to know how to root themselves to moving terrain, but as we move away from in-game terrain and toward terrain authored in Maya / 3DS Max, we start to lose that capability (but maybe I can make it work via a manual update).

      • So was the in engine terrain always intended as a flexible place holder while you were designing or will some of it survive till the end.

        • We weren’t sure at the beginning. But the in-engine terrain is height fields, which are not very good at representing vertical surfaces or fine details. The current plan is to swap all of it out for stuff authored from Maya, eventually.

  3. Coming up with an overall concept is the easy part – saying everything you need to say with as few lines, colours and movements as possible is what keeps people up at night (or it should, if their standard of quality is high enough to keep working towards an elegant solution after brainsplatting the initial concept).

    And finally understanding what you need to say such that you can say it simply AND someone else will go “oh, hey…is this what you mean?” and you’re like “OMG yes! dude, totally!” or “…you may think that, but I couldn’t possibly comment”…that is pure Art

    Everytime I talk to a local or federal “council for the arts” group and get into a discussion about whether or not game designers are doing anything as artistic as theater or film – I point them to this blog.

  4. Jonathan

    How important is leaving time between thinking about a puzzle and designing it?

    Do you ever worry that you have a significant puzzle to build, but can’t think of a way to design it?

    Any thoughts on what your brain is doing while you’re “not building” the puzzle?

    • The main time I have ever had this worry is with the one puzzle I am talking about here.

      Actually there was one other time, a couple of months ago.

      But it’s not the usual situation. Usually puzzles come from organically following some initial kernel of possibilities, so they will naturally fit together and feel complete. Sometimes, though, it’s obvious that a certain idea is needed for completeness, but I don’t see how to make a good puzzle out of it, and that is where the problems come in.

      However, just giving myself a lot of calendar time to stew on possible solutions has always solved these problems.

  5. “One of my primary design goals, all through development, has been to keep the island spatially small, so that there isn’t a lot of wandering through empty or pointless spaces. The idea is to take the best parts of an open-world game, involving choice, exploration and discovery, but compressing these into a maximally dense experience.”

    Yes! This! I wish more games would do this…

    • But not all :) Some of my fondest gaming memories are aimless nature hikes through the wilds of the Elder Scrolls universe.

    • Sometimes the pendulum just needs time to swing. For a long time, the playable area in a game was limited in a very basic technical way, so games were always just trying to make bigger worlds. Then they would brag about how big their worlds were, as a primary selling point.

      Eventually, though, people started making worlds that were too big for the game’s own good. Sure, for some kinds of games you want as large a world as possible, but for some kinds of games this is pretty bad. It’s just that most designers don’t see this yet because they are still in the old size-bragging mindset.

      The Witness is a game about focus and about things being built intentionally and carefully; the smaller the world is, the more attention we can lavish on any particular part of it.

      • Ha! I knew it! I was right all along! hahaha!

        • Yes, and GTA V is evidence of the blind “bigger is better” attitude about these things. It remains to be seen whether their overblown world size will pay off. Especially since I felt that San Andreas was already too big of a world, and yet they brag that this game is bigger still. I don’t want bigger worlds, I want more to do inside them. And I want the things I do to matter more.

          • it’s funny you brought up their “bigger than SA, Red Dead and GTA4 combined” thing, because I only just read a quote from one of the Houser brothers saying it was explicitly neccesary to make the map huge because of the extensive use of flying in GTA5.

            That rings true to me, because I know they didn’t include planes in GTA4 because the map was so condensed that you’d be able to fly across it far too quickly. In their case it does sound more like a balancing act than a “bigger is better” approach. Ofcourse, I know very little about their mindset, so I could absolutely be wrong.

            I’m sure it’s an absolute pain to traverse the GTA5 map on foot though.

          • @sebastian It is quite possible that there is more reason to have such a large world than bragging rights, although they certainly are using it as a selling point. However, I merely mention it because I think GTA is the quintessential open-world/sandbox game, and as it is a good example of mainstream thought on nonlinear world design: it is interesting to see how it contrasts with the design approach of The Witness.

            Of course, the games have radically different gameplay and target audiences. But the mainstream idea is clearly that a larger open world is better. Which is fundamentally true, up until a certain threshold is reached, at which point it is either unnecessary or actively damaging to the quality of the game. I believe the Witness is seeking the complete opposite result. It intends to be as compact as it can possibly be while still containing the experience that Jon hopes to convey. (Not that he is exactly an experiential game designer, but I can’t think of a better thing to say. Substitute gameplay or content if you want, but I think that is disrespectful to the intrinsic value of the intellectual muddling that is involved come jotting and exploring one of Jon’s games.)

            Either way, I am excited for both games and very happy to see some diversity in approach to game design. Even if it is most often achieved by simply taking the diapositive design approach to the mainstream. We all have to push the envelope of game design in multiple directions at once, with as many orthogonal approaches as we can muster, if we ever want to even approach the level of expression afforded to more accomplished mediums.

      • Do you worry this might make the game less immersive and believable? Also, I understand that, in puzzle games especially, you don’t want extraneous elements distracting/misleading the player, but if a game is “too tight” it can feel overly scripted and, in a way, monotonous. You’ve talked before about the importance of breaking format once in a while, and making everything “maximally dense” seems to cut against that.

        • What does “immersive” mean? What does “believable” mean? We are not trying to trick the player into thinking they are walking around the real world or something. That is just not what kind of game this is…

          … all this has to do with what I was saying about knowing what a game is about.

          Since this game is not scripted at all (it is nonlinear!) it is not going to feel overly-scripted. The thing you are talking about (a game feeling too-tight) is not the real problem. The real problem is that the game is linear in the first place and is trying to fool the player. If you stop needing to fool the player, then you no longer are subject to the constraints you had imposed in order to fool the player.

        • Billy – its important to remember that believability and immersion are two very different things. Sure sometimes making a more believable world can make the player be more immersed in the game. But I would argue that consistency of the world and interesting game dynamics are much more important to achieving immersion.

          • You’re right, I used the word immersion carelessly, my apologies. What I meant was believability.

            Jonathan said something interesting though: that, in this game, there is no need to fool the player. Certainly, unless I am very off-base about this game, there is some need to do this: you want to give players the experience of walking around this beautiful deserted island. In reality they are looking at models made in Maya and listening to .wav files and so on, but you don’t want players thinking about these things.

            I’m not trying to play “gotcha” here, I bring that up because at a higher level something similar happens. If a player exploring the island stumbles upon, say, an empty peninsula, what goes through their mind? What (I think) you want is for the player’s thought process to be along the lines of “… oh, this is interesting. There probably isn’t a building here because it is too narrow. Maybe whoever lived on this island fished or swam here…”. You want their imagination to sort of fill in the gaps. What you don’t want is for the player to think “… well, so far everything else on this island has been critical, so I bet the game designers put this peninsula here for some reason. Now I gotta figure out why.”

            I’m sure you all have thought this through and have taken deliberate steps to account for this. And I’m certainly not accusing you all of taking all of the “life” out of the game based on one blog post. My concern is that, if there is this deliberate attempt to tighten things up as much as possible, your margin of error goes down. You can still create this organic experience and render it on to the player, you just have to rely that much more on your own skills to do it.

  6. hyrulefields in oot was spacious and enjoyable (novelty of the horse – distance had some meaning to it)… now though, i hate having to walk more than 7s in the same static environment, particularly elder scroll games. as nice as the environs look, one gets acclimatized to it quickly and tires of seeing those procedural generated terrain. points of interests are too apart and the movement speed is frustrating. (insta travel anywhere on the map also makes me feel detached from the game as well)

    side question: please tell me there will be support for the oculusvr. i imagine your game will be out before the consumer version of the hmd will be out. devkit/sdk i think will be out in the new year though (i have no associations with it – i just imagine this type of game would look spectacular with it). also, don’t spend time implementing it until you have released the game — i want the game asap :D

  7. Braid was rare for me in that when I recognized what it was, I felt a need to drag it out or to understand completely what was offered at each world. I played it in small, 30 minute increments for weeks. For me it made me rethink the importance of the end of games, despite the incredibly surprising ending it still delivered. I imagine the witness might unfold similarly.
    I would like to think that the Witness will be completed because it is done, and not because of financial concerns. To that end, why not start taking our money now? There has to be a legion of people waiting to pay for it regardless of time frame and if that buys all the time and resources that you need, are there any drawback to raking in dough? Considering how quickly indie games can do this, especially credible ones, I don’t know why you wouldn’t.


    • I agree with AL! Please don’t get a publisher, they’ll just end tripping up progress. If you are developing for longer and need more money just get it from the people. We will never tell you what to do or when to release

    • I had the exact same experience with Braid. I played it over a long time for very short periods at a time, and consistently found that 20 – 30 minutes spent in that game was fulfilling enough for me. It was better than a whole day spent playing most other games.

      I never understood the people who complained that braid could be completed in 6 – 8 hours. Why on earth would you do that? It took me at least an hour just to take in the artwork and music of that game.

      • I completed Braid in a few hours. It wasn’t a challenging game. Beating it quickly isn’t something to brag about.

        I haven’t played it since. To me, it’s a “reference game” for game designers; something you put on your shelf and take out when you need it.

        It does so many things right that many problems can be solved simply by asking, “How did Braid do it?”

        In my opinion, that’s also one of the things that keeps Braid from being fun after a few plays. It’s a closed loop; so theoretically correct from a game designer’s standpoint that there are no flaws to be inspired by.

        That seems to be a flaw carried by artistic game designers. We create a product so “perfect” in our eyes that we dare not taint it after completion. I can nitpick at any game someone else develops, but I put so much heart and soul into my own creations that I have difficulty touching them even after encountering legitimate criticism.

        Games like Braid are a vision. A vision in the flesh.

        • That is precisely the approach that confuses me. You said you completed it in a few hours because it was not a challenging game, and I’m wondering why somebody would push through a game as fast as they possibly can. What I mean is, if the puzzles were hypothetically twice as easy as you found them, then you could conceivably have completed the game in only an hour and a half. I was only trying to say it took me about that long just to take in all the books, the soundtrack and the art style.

          Everything else you said, though, has really given me pause. The concept of a reference game, and whether Braid is inspiring. I respect your thoughtfulness on the subject Pritchard, I really do.

          I do wonder how accurate it is to say Braid is flawless though. To me, the rules of videogame best practices seems to be changing all the time. Jonathan Blow himself said in a talk once (paraphrasing) that a puzzle in the first world troubled him, and he may not have done it that way if he had more time. I believe it was related to showing the player it’s ok not to get every single piece right away.

          Those are exactly the kind of things that make Braid so interesting to me. Who can say for sure whether the game in its current form was right? I actually found that very piece to be one of the most memorable experiences of the game.

          All this does seem to confirm your point that Braid may be a game to refer to, rather than one you simply just play. But is that any worse than a painting that inspires discussion? I don’t know that I’d want to play a game that didn’t give me something.

    • At some point it has been “long enough” and you really ought to ship the game. Even if finances didn’t exist as a concept, we would be nearing this “long enough” point pretty soon. So I am motivated to get the game done. (But I won’t release it if I feel that it is missing something.)

  8. Literally cannot wait for this game to near completion. Glad to hear the progress is excelling. Snafus happen – projects are built from the messiest circumstances. Please keep it up, your fans are eager!

  9. Jonathan:

    I ran into you at a coffee shop in Oakland last week, and though I tried to act nonchalant about it, it was quite an honor to meet you. I believe I left you with wishes of good luck on The Witness. Braid became that “novel I read every day” or that “film I watch every couple of years.” It has inspired me to seek deeper experiences in my own discipline. Your insights into the creative process have empowered me to scrap large pieces of work that did not perform to my expectations. And you managed to do that in competition with a remarkable number of works that I cherish and study (it’s sort of my job).

    I’m not sure if this gushing fandom seems like little more than a very satisfied consumer who got his money’s worth from a product you’ve released – but I suppose it’s still better than saying nothing.

    Best wishes. And thanks.

  10. Did I see a pokeball ?!! OO”

  11. Interesting; I guess walking a block through Las Vegas would feel like covering more distance than walking through the desert that used to be there.

    Re: that new puzzle, was curious – I been making a fair bit of music lately, and I’m finding that a great way to structure my time is to have “tools” sessions and “action” sessions – the tools sessions are pretty left brain, learning new music software, a scale on a guitar, setting up sounds I like in a synthesizer, etc. The “action” sessions are much more right brain, just creating as quickly and intuitively as possible from the tools I’ve set up for myself.

    Been wondering lately if/how I can apply that to game programming – seems there are always so many implementation details for any given problem, that it is hard to create rapidly and freely. Having a good codebase and tools can help, but seems there’s a certain amount of “essential complexity” in game programming you’ll always have to reckon with, to where it’s hard to get out of engineer mode.

    From your descriptions of your process (i.e. the puzzle you built last week, or the early prototyping in Braid) sounds like you might have something akin to right-brain “action sessions” – curious if that’s true, and if you had any quick thoughts/advice on staying creative as a programmer?

  12. Does this reduction at least spare you some time, work and money which it would cost to add all those details like rocks to the old, bigger version of the island?

    Or would that not have been a problem and this is absolutely only about that it would have felt too big?

    (I feel a bit sad about the stuff that has to go now, but I trust you of course that it will be a much better, focused result. When I play games, I usually take my time to admire all the levels, art and props and look for easter eggs, etc. Must be because I have always been fascinated with 3d graphics from the demoscene. The world in Brütal Legend was so great to explore)

    • It probably does save us some time later. Not a huge amount, though, because percentage-wise we are probably removing less than 10% of the island’s area.

  13. Wow!! The beauty of game’ distance lies on the amount of physical elements placed (in some order). Braid gives that feeling of satisfaction regularly whenever I play it. The music, the paintery style background, and the game logic all are so fulfilling and beautiful to the eyes, ears, and the mind that it starts feeling very good at heart.

    My friend, Paras, and I are developing a game called MirrorD (pronounced as Mirrored) (http://www.igf.com/php-bin/entry2013.php?id=348). It took more than a year to come to what exactly has to be shown in the game. It’s inspired by Braid, Fez, Edge and many more. He designed the levels small and yet each level tells you something new. We leave the understanding completely in the hands of the user.

    A 9th grade fellow completed the present eight levels and realized whatever each level told. He enjoyed the logic of the game completely. After completing he told us that he has never/rarely player computer games and dislikes puzzles but he liked this game’s logic very much.

    The feeling that fills a developer when someone tells that they enjoyed one of the arts of a game is so thrilling. :)

  14. This is one out of only two (!! that’s how shitty the gaming industry has gone!) games I seriously look forward to. Braid blew my mind many times over. I am extremely grateful to read you’re putting the same kind of though, dedication and love into this game. I’m sure that however long the wait, I’ll love every millisecond of it, and I’ll be flabbergasted and stuck on the soundtrack for months on end again, after :).

    • I loved the soundtrack to Braid as well, it was beautiful. I noticed you said you were looking forward to being stuck on the soundtrack for months, and I just wanted to say I’m almost positive I once saw on this blog that Jonathan was considering using all environmental sounds for The Witness. I could have misinterpreted that though.

  15. I just played through Riven again for the first time in a decade (the iPhone version). Whole strings of ancillary winding hallways and corridors connect some of the spaces, without puzzles. One example is a “blue cave” between the village and the stone steps on the main island—it appears to have no real in-game purpose—but as you pass through it, the music changes into a hushed/mysterious ambient piece. It’s nice. Maybe I was a bit frustrated, eventually, that such a memorable place in the game doesn’t have “purpose” in the larger scheme of things. There’s no essential or even optional puzzle; just some mysterious cave art. But I’m not sure the game would be “better” without these kinds of atmospheric pit stops. I don’t know.

    But then again, I think I remember just wanting to get to the damn puzzles when traversing the island meant actually physically swapping CDs :p

  16. I am so incredibly hyped for this game. Knowing the placement of every smallish stone is considered only heightenss that.

    The ideas of density and how that effects the feel of scale of a world are really interesting. I wonder if this is the game that will have the largest density of design compared to area of map ever – at least in terms of effort and thought (e.g. all the architects) it very well could be. I love exploring large open environments such as in Far Cry 2, but I would love to see more devs designing in the direction of smaller denser worlds. The upcoming game Gone Home might be trying something similar with going for that level of detail – there’s not that many others though, unless I’m just blanking on some clutch of games.

  17. I recently played some of Beyond Good and Evil and felt like the Pedestrian District demonstrated this. It was a small area that felt big because of how it was layed out and I enjoyed exploring it.

  18. This game is still looking amazing. Other than the island shots, I haven’t seen anything of this game since a demo on youtube a long while ago.

    Slated release still claims to be before then end of this year, and I was just wondering if that was still a reasonable expectation. I obviously want you guys to take as much time as you need to make the game as great as it can be, so I just wanted to know if I should adjust my expectations about release timeframe according to development realities, rather than a number I saw somewhere.

    Thanks for the updates, and keep up the great work.

  19. The concept of “maximally dense” in a game is intriguing to me. I from a players perspective, I do understand that some developers have erred on the “bigger is better” approach to game design, with sprawling levels that go on and on….

    However I do like having areas in some games that have no real purpose other than to allow the gamer to immerse themselves in the game and ‘feel’ the world that the developer has created, ala Dear Esther or Myst.

    So my question is, how big is the island/ how long does it take to traverse it? And how have you been able to gauge the ideal physical size of it? I’m guessing that you have gotten some sense of what is ‘just right,’ but what led you to determine that?

  20. The map looks an awful lot like Africa’s eastern shore has collided with Europe’s north. There’s some crazy Pangeac break-up happening here.

  21. Does this set back the estimated release date by much?

  22. with 400+ puzzles and some pictures i remember posted of a few puzzles that look the same(monitor with dots or something like that) i have to ask: how many of the puzzles are of the same kind? how many unique puzzles (like the shrine up on the mountain or the roofed bay\ruins) are there? of course if those locations have indeed unique puzzles..
    no disrespect meant and i love reading here every once in a while.

  23. I knew it! I know you get what he means. You know exactly what he’s talking about. Why are you being mean?

    They’re all mazes, some are on blue screens, some on computer like monitors, and some on see through crystal. Some are about avoiding shadows, matching colors, going above or below some dots on the maze and others are two lines that do the exact same movement left and right symmetrically and others go by sounds or shapes in the environment!

    There! That’s the answer the fan Invincible Man was looking for.
    Capcha: Not employed

  24. Indeed. I don’t believe his intent was to be critical of the design of the game, he is simply asking a reasonable question about the nature of the puzzles. (Even if you have answered it dozens of times over the years.)

    Perhaps you are being sarcastic, Jon? That stuff never works on the internet, doncha know?

    Yes, Justin is right, at least insofar as Jon has let on. All the puzzles have the same input method, drawing lines on a screen. This is actually a strength because, unlike many adventure games, it provides an input method for puzzle-solving that is clean and recognizable. Although there are several mechanics within the screens themselves that create puzzles. The variety seems to mostly come from reflecting upon the context given to the current screen. The mazes are used as a tool to reflect upon interesting aspects of the universe.

    Or at least that’s my hope. I have not played it yet, but I am always excited to hear Jon’s thoughts. Braid was essentially a slice of his brain, but The Witness seems like its the whole proverbial pie-brain. (or enchilada, to reference Jeff&Casey) It is a dangerously personal work, from what I have heard, and that is pretty unprecedented.

    Jon’s hand seems to be hovering over the artistic self-destruct button.

Leave a Reply to Matthew VanDevander Cancel reply

Your email address will not be published. Required fields are marked *