Okay, I played the new alpha build on a more powerful computer. Not sure if this is the preferred place to put my feedback, but I'll put it here anyway.

Here are my thoughts, as I played the game a few times:

The space station setting was not obvious to me, despite the "traveling through space" and "looking at cryo-pods" scenes. Some sort of establishing shot that ties these together and provides more of a sense of place would help tie these together, I think.

I experienced the "framerate hiccups" that were mentioned as a known issue (at the highest quality setting).

I found the turning by moving the cursor to the edges of the screen to be an awkward and annoying experience. I would like to see if it feels better with a "turning hot zone" that is bigger (so I don't have to move the mouse so far), but also slower, so the turning is not so fast and abrupt (the closer the cursor is to the edge of the screen, the faster it would be). And without a delay between moving the mouse to the side and the camera turning.

The way the (male) avatar's head leans as it turns (combined with the effect of the wind on the hood) looks disturbingly unhuman, like the avatar has a rubber neck and no head (turns too far?). The way the shoulders move with the turn also looked very unnatural.

The way the avatar side-steps (when turning?) looks very awkward and unnatural, as the legs twist into each other.

I liked it when the surf came up to my avatar's feet. I wanted to wade further in, and was disappointed when my character could not walk past an invisible boundary on the beach.

This may be intentional, but the white text of the messages was usually obscured by the white of the background (so I could rarely read more than half the letters of the text).

The messages also came up and disappeared very fast, so by the time I realized one was there it was already gone and I was unable to read it.

The seagulls shoot off like rockets when I approach them, instantly snapping into a full-speed flight away from me, which looks very odd to me. There is no gradual waddling away and gradual take-off like I would expect.

When I turned to look back at my avatar as he faced the building, the camera was inside the building and could see stray pieces of polygons.

While waiting for a partner at the table, I don't like seeing the back of my avatar's head (low-res texture and polygons) filling the screen. I would like to see my avatar at the table, far enough away so his fully body is visible (at a comfortable distance if I were a person watching him, not shoved up into the back of his head).

The top-down view of the chessboard with a partner feels claustrophobic, feeling like someone is holding my head down (since I want to look at my partner but can't). I would really like to see my partner as a first-person view of what my avatar would be seeing.

The lag between moving my mouse cursor and the chess piece following feels annoying - I would prefer the movement to be one-to-one. Or at least, the hover text could be based on the mouse cursor position rather than the chess piece (that would feel more immediate). You could have a highlight on the circle that the mouse is over, in that case (use a white circle instead of black).

The transition to text (instant cut to black screen with text) feels jarring, and then it goes away too fast, before I've registered what has happened. I'd like a more gradual transition, and also for the text to stay up longer, so I can feel it (instead of remember it fleetingly). And I'd like a second or two of delay between releasing the chess piece and seeing/hearing the text.

I would like to be able to move the objects on the table, not just click on them (wineglass, ashtray, cigarette packets). The transition to black screen for drinking/smoking is also too jarring and brief for me (just like the text).

Holding down the right mouse button while moving the mouse and then clicking the left mouse button without letting go of the right mouse button is uncomfortable for me, and I imagine it would be even more so for people who are used to only using the left mouse button (casual gamers). That may not matter to you.

The hard edge on my "eyelids" as I close my eyes feels weird (doesn't feel like closing my eyes). I'd prefer to see a blurred edge (and rounded, concave) to feel more like eyelids.

The intake of breath when smoking a cigarette sounded like a woman's even when I was playing as a man.

I was disappointed in how similar the woman's clothing is to the man - they seemed interchangeable, when I was expecting a different style, and maybe dark colors, for the woman.

I often found myself clicking on an object to walk toward it, only to find that since the thing I clicked on wasn't walkable, the game ignored my click. I would prefer if then my avatar would walk as close as possible to the object instead of ignoring the click completely.

Pressing the scroll wheel only closed my eyes - I was unable to run. Maybe a double-click could be a run action?

By the way, I bought and downloaded the alpha a while back, but my ancient laptop is too weak to play the game, it seems.

At some point, I'll see if I can try it on a more powerful computer.

I finally got around to skimming through Bret Victor's Magic Ink, and it actually talks about the idea of "expressing general logic through specific examples" that I'd mentioned earlier.

There's even a name for it - it's called programming by demonstration. I'm imagining demonstration mostly by diagrams, but there could be other ways to do it.

We're getting closer...

Ian Bogost's new project Game-O-Matic is a very interesting take on this problem. You make a concept map of things and verbs that connect them, and it randomly generates little games based on them.

You can see a video here:
http://www.youtube.com/watch?v=Y7KacgqA8cI

Finally got around to reading (skimming) through the paper.

Mostly interesting for its description of this visual programming language, like a mix of Reactable and Haskell:
http://vimeo.com/19384095

Described further here:
http://yaxu.org/writing/texture-icmc-preprint.pdf

I wasn't complaining about the aesthetics as much as I was about the whole clunky process.

The easier / more cost-efficient alternative is to use a post-processing effect that mimics some of the aesthetics you're after.

I was hoping that I could pair this with a totally different creation method - if you're not building with triangles, you'd have to build with something else. If you can tolerate a probabilistic rendering method, maybe you could tolerate a more probalistic data representation and sketchy creation process. That's how I meant to address the original topic of discussion.

I have some vague ideas of how you might go about this, but nothing specific.

I agree. It's a similar problem to the problem of programming, though maybe not as bad.

I would love to see a probabilistic rendering system based on something other than triangles, that produces something like this:



It's something I've been thinking about in the back of my mind for a long time, but no grand insights (or experiments) yet. Still, it is a dream of mine.

The closest thing I've seen is maybe the game Love, by Eskil Steenberg.

A possible solution, or at least a way to work towards more satisfying results, for us has been to reduce the importance of concept art in favor of working in the game engine as much and as quickly as possible. Imagination is your worst enemy sometimes. It's better to start dealing with how you work will be presented as early as possible. And then you work with everything that's at your disposal.

I think this is a great partial solution, in the meantime. Kind of like rapid prototyping with game programming instead of just imagining grand design visions in your head.

For me, suggestion in videogames involves triggering things from the viewer's less than conscious memory. The memory of how the sun feels on one's skin, the memory of the smell of wet wood, the memory of the sound of the forest. We do not need to actually show these things, or recreate them. We only need to make the viewer imagine them, often on a less than active level. People do this spontaneously when stimulated in the right way.

Videogames are about being, not seeing. So what we are suggesting is existence itself, not, e.g. the logic of narrative or chains of causes and consequences.

Very nice answer, thank you.

If you want to look at film as an example, the solution was a better use of editing through a transition from depiction to suggestion. What does suggestion look like in videogames?

This is a good question. I don't know the answer, but maybe making more short games like The Love Letter will help...

I'm definitely familiar with Knytt, Knytt Stories, Seiklus, and Small Worlds.

Knytt is definitely a big inspiration. I guess I could describe my thought as taking Knytt and giving it more of a story, rather than simply exploring the beauty of colors and sounds and moods in different platformer spaces.

Knytt Stories gets away from this - there's more story, but the gameplay diverges from it even more. Seiklus lacks the kinesthetic joy and expressiveness of Knytt, and does not try to tell a story. Small Worlds is closer, but again, it does not try to do too much.

And hopefully, there could be a way to make it more accessible than Knytt, even, for people like Michaël.

I just played dys4dia, found it interesting and worthwhile and inspiring - and crude and primitive - but it's a start. I have no problem with pixels, but I didn't like the visual presentation in this game.

I like this idea of giving gamers their gameplay fix at a subconscious level while focusing on meaning and content at the conscious level.

Along those lines, I've thought about doing a story-focused platformer, where there is a continual low-level gameplay experience of jumping and platforming, but the focus is on exploring a story and environment rather than trying to make the platforming challenging. The platforming would be like eating popcorn while watching a movie.

Like what Dear Esther might be if instead of boring FPS controls it used something more like the controls in Assassin's Creed - much more tactile and kinesthetically fun.

Always the same emotion.

I agree that that's a problem.
And definitely a point where such games differ from art.

To draw upon Robert McKee's ideas about story (as I described briefly here), what these games provide is not "emotion" - it is a bunch of emotionally neutral changes in value, positive (with success) and negative (with failure).

In order to be "emotion" (by this definition), the neutral value changes would have to be tied to an emotionally resonant "mood" that provides context. Kind of like a wind instrument - the value change is the vibration created in the mouthpiece, and the mood is the resonating chamber that locks and amplifies the initial vibration into a pitched note. The better aligned the gameplay (neutral value changes), mood, story, and setting, the more likely that the player will experience emotion beyond the bland sweetness of virtual coins and levels.

This is not easy, but I believe it is possible. Lately I've been encouraged in this direction by the apparent success of my little (not)game The Love Letter. This article has a great discussion of how these elements align in The Love Letter to create an experience that is more than the typical neutral pleasure of gameplay:
http://www.slowdown.vg/2012/02/25/on-the-love-letter/

Nice find! I'm looking forward to reading this...

Nice! I liked it - some interesting points about nets, linearity, exploration and rebellion.

Anyway, something caught my eye skimming over it, she claims every girl winds up dead with their broken body shown at the end. I don't remember any broken bodies. It's implied that a few of the girls die but it seems far from certain for all of them.

I'm pretty sure the article is referring to the 2D images that flash briefly at the end of the grandmother's house sequence for each girl.

Thanks for the link. I'm reading through it now.