The island so far…

I figured it would be fun to post occasional screenshots of the island so people can see it gradually come together over time.  Here's the first one.

Keep in mind that this is all a rough draft for gameplay purposes, and none of this is supposed to look remotely like final game visuals.

Despite not looking fancy yet, there's already more than 5 hours of gameplay.  So that's pretty good.

(Click image for full size.)


  1. the scene is very nice, reminds me a bit the legend of zelda.

  2. I love exploration very much, since all the way back to the original Zelda. You’ve already won me over simply by being Jonathan Blow, but exploration is still a big plus. I think that it evokes a primitive and mysterious emotion that we don’t necessarily trigger in our lives normally. Can’t wait to see how the island evolves!

  3. Lookin’ good, Jon! It’s always interesting to see early work-in-progress shots from other designers. Love the blog as a whole, by the way. I’ve garnered some pretty interesting perspectives/info from it :-)

  4. You really like the idea of an island, don’t you?

  5. Were those trees generated using L-systems, by any chance? I did a 3D world project a few months ago, and those look similar to the trees I generated.

  6. Actually it’s mostly the same hand-made tree pasted all over the place. With a few other trees in there.

  7. Am I wrong if I get some Myst vibes from this?

  8. now I’m really curious how this translates into 5 hours of gameplay. there must be some kind of economy of exploration that I’m not privy to.

  9. Really cool.

    I’m really digging this development blog. Keep up the awesome work, I’m already looking forward to the final game.

  10. It’s looking quite good already!

    As far as I can tell, the shadows are still unfiltered and presumably produced by a single jittered sample. If so, what technique are you guys planning to use to reduce the aliasing?

    By the angle in which the sun is hitting, I’d expect warmer light, are you using a skylight model for this particular render?… A proper skylight lighting model can really succeed in bringing realism and deep visual association to a natural scene.

  11. Shadows are actually 4-tap PCF currently. The thing is, 4 taps is not enough. I am not convinced that non-point-sampled PCF is a good shadow technique, in general. The look of the shadows will probably change a lot before ship, but I am not sure how. What is there right now is fine for gameplay prototyping, which is what I care most about at this time.

    We are baking skylight into the precomputed lighting. However, almost everything you see in that scene has not had the precomputed lighting run on it. So, there is that.

  12. Thanks for the explanation. 4 taps is indeed far from enough for the simple PCF technique.

    I’ve had my headaches too toying with VSM, ESM and whatnot, however, there have been certain developments exposing shortcuts for use in PCF that enable faster filtering, which, if I’m not mistaken, numerically exploit the interpolators of the texture units to reduce the required samples, and, combined with the fetch4 or grab4 texture samplers, are able to provide fast, good-looking shadows.

    There’s also the smoothie method that I haven’t explored.

  13. Hello

    Looking at this island an old question arises in my mind (might seem really offtopic, brace yourself :) )

    On November 19th, 2008 at the Montreal International Game Summit you gave a speech about story and game design. You talked about story games being inherently conflicted. Most notably:
    1) Interactivity sabotages structure and delivery
    2) That conflict is unsolvable , we cannot use:
    Poetic Adjustment
    Tone Adjustment
    Vocal Emphasis
    Body Language
    with the full potency movies do

    That was an eye-opening speech for me at that time, got me thinking for long.

    Now I look at this island and I think “It is an art, it is sculpture, people used marble before, now we use Maya and 3DS, but it is sculpture nonetheless”

    But sculpture has structure and delivery technicues too. For that mater, all arts do. Visual arts have dos and donts about colours, composition, etc. Narrative, music, lyrics, concerns about rhythm and length and hormony, structure rules all around.

    And Interactivity sabotage those too !

    So basically every single tool that we use to present the system / meaning / importance of our games, inevitably leads to conflicts, that you stated are unsolvable.

    Now I may suck at drawing conclusions but … how do I avoid that in my games ? How do YOU avoid that ? And I don’t mean in problem by problem basis, I mean generally, scalable rules ?

    Is there any yet ?

    I will appreciate any additional reads on the topic btw :)

  14. I think you are making too extreme a statement.

    You have way more interactivity with a sculpture in real-life than you do with a 3D object in a game world. That does not sabotage the delivery tools of the medium of the sculpture. You as an individual, if you want to, can refuse to look at the sculpture, or smash it or something, but so what? If you want to experience it, you can do so just fine.

    This is different from the situation with games, where, even under the best of conditions, even with ideal audiences, I do not think we can come anywhere near a novel or a film insofar as storytelling is concerned.

  15. Thanks for the reply.
    I really admire what you do. Good luck with the game :)

  16. This is looking pretty good, just wanted to ask, how big will the character models be?

    Cos obviously if they are large the world might look smaller overall but if they are smaller the world will be larger to explore

    Also, i finshed Braid, and you and your team are just great, makes me wanna try my hand at indie game developing lol!

  17. I have a couple of questions. First, will the game be live-rendered, or will it be pre-rendered like the original Myst? Second, if the game is released on PC, do you forsee a delay for the PC version like there was for Braid? That was a real problem. What caused those PC Braid delays, by the way?

  18. It’s realtime 3D. And I wouldn’t get too cozy comparing it to Myst; it’s an extremely different game.

    Whereas I do think that games coming out on different platforms at different times is not ideal, that is just how it is sometimes. There are always different reasons. Why was it “a real problem”? If a game is really good, then isn’t it just as good 6-8 months later? If you aren’t interested in playing something a year after its release, then what intrinsic value can that thing really have?

    From where I am sitting, it is kind of lucky that Braid got finished at all, on any platform, so when someone comes up to me and says “such-and-such is a real problem,” I am like, what??

  19. Hmm. The more I think about this, the less simple it seems. Part (if not most) of the blame lies with the people who get impatient and turn the enthusiasm to something else, forgetting about the delayed game, basically going ‘you had your chance’. In fact, it might just be me. I have no idea how common this mindset is.

  20. I’d like to add to what I just posted. On further thought, the ‘I’ve lost interest, you had your chance, I’m not clinging on and waiting day after day any longer’ mindset is the ENTIRE problem. Again, maybe it’s just me.

  21. Ah i got no reply :( lol

  22. I forgot to mention that I’m only referring to when the game is already out on at least one other platform but is being delayed on a personally preferred platform. There’s a sense of unfairness.

  23. Sorry to multipost, but the more I think about it, the sillier the subject seems. If there’s a result of extended delays, it’s just that of loss of interest. And that’s not so much a problem as it is a fact of life – or rather, Internet.

  24. Sure, this is all true. If you delay a release on one platform beyond the initial release date, you are probably going to sell many fewer copies on that platform.

    But why act like that is a good thing? To me it just says something kind-of negative about human nature.

    All things being equal, it is best to launch on all platforms at once, but all things are never equal. There are always big issues. And ultimately, I care more about making the best possible game than I care about selling the most copies. So I prefer to put most of my thought into how to make the game, rather than what platform it is going to be on, and when, and why.

  25. Umm, the people are people-sized?

  26. Ah don’t worry, think you misunderstood the question…

    Good luck with the project anyways.

Leave a Reply

Your email address will not be published.