Probably most of us have seen the video of the zombie-shooting game on the prototype (?) NVidia handheld, this is similar but more of an art project and much simpler technology!

I'm almost as interested in augmented reality as with virtual reality, and I hadn't seen that game- it's pretty redundant, like most AR games. They won't be worth shit until real space acts on them.

The directed laser thing is far cooler, though I wouldn't quite call it "Augmented Reality", since it doesn't have much of a reality of its own.

Incredible. Sorta reminds me of Operation Inner Space.

Playing Blendo's games is like jacking off without busting a nut. sorry

^ If your character is carrying an umbrella then it's inheriting some transformations from the character. In this case it still seems to me that you have some kind of scene graph, and you need to find a way of generalizing relationships within it. You could have nodes control the drawing order of their children nodes, such that "umbrella is behind wall" follows from "character is behind wall". When nodes are released from their parent you could also have them inherit the depth of the parent + whatever other necessary constraints.

Yakuza. Guess it's a beat-em up/adventure game with minor RPG elements. The combat is diverse and extremely fun. It's one of few games with a contemporary urban setting, which it pulls off with some of the most impressive visuals on the PS2. I've yet to pick up the sequel but it's supposed to be a minor improvement.

...is Passage really less deserving than Flower of $10 just because Passsage is only about 5 minutes long? And does that somehow make it less worthy?

Well, yes, in a way. It depends on how much enjoyment you get out of those 5 minutes. Every game you purchase you'll play for a finite amount of time, over which you'll gain some amount of some form of pleasure- which is basically what you're paying for. So if a game is really short it could be considered less worthy. Certainly not worthless, but I personally would never pay $10 for a 5 minute game. That's atrocious value compared to any book, film, audio recording, etc.

Then again, maybe the donation model differs here more profoundly than Burch indicated. Because I'd consider donating $5 or so to an artist who made a 5 minute game. Donation is completely different from purchase.

...we're willing to pay $60 for the chance that a game will be good so long as it's a big-budget, mainstream title

Some of us are. These are some of the primary reasons why I pirate games- low quality games, and the lack of availability of demos (when they are available they're seldom representative). ~98% of the time I play a pirated game for half an hour then discard that crap. If it's not even worth my time, it sure as hell isn't worth my money.

^ For me it was Myst 5. Of course whether it's exempt from a "crass profit motive" is as debateable as with Blow's upcoming game, but at least it was clearly a labor of love (even though everyone but me disliked it). TBH I don't remember a Sven-Coop mat fitting that description, and I've played it loads. What's it called?

I'm anything but experience in music, so you should probably ignore my advice, but I feel an important fact has yet to be mentioned.

I agree that learning an instrument is probably a good first step. I never tried writing music before I started learning the guitar. However, it's often forgotten that the human body is a musical instrument. Seriously, if you want to get a better feel for music, it actually helps to sing, chant, hum, scat-sing, or whistle frequently. You can improvise off songs you like, or make new versions in your style using any of these techniques. They also have the bonus side-effect of pissing other people off. Compared to any other instrument the body is highly affordable, convenient, portable, and durable.

Similarly, most surfaces are percussive instruments- including the human body. Become a hambone hero.

This works fine for fairly simple games, but it gets horrible quite quickly for complex projects. The canvas doesn't know what all these things are that it's supposed to render (assuming modularity), so when you want to add another one you don't know where to put it.

What's wrong with just exposing any necessary drawing functionality to the public interface of the canvas? There can't be that many different kinds of things to draw, surely.

The only other way around this is to serialise all your operations at rendering time. So the renderer says to each object "Excuse me, but what depths would you like to render this frame?", then processes the various rendering operations in some suitable order. But then this approach will force objects to cache information on what depths they want to render at, which is actually more difficult than just cacheing a pointer to the canvas in the general case (because you can't just say "in front of X" if object X might disappear between now and rendering).

If you have to be able to control rendering order, then I don't see the difficulty in simply storing depth as a numeric value in drawable objects. I would also keep reference to those objects sorted by depth in an addition container in the canvas. "In front of X" could be implemented by setting an object's depth to just above X, so the fate of X afterwards doesn't matter.

I'm pretty sure the original poster intended "bullet->gun->character->room->canvas" to be a hypothetical example. The fact is, long reference chains like this DO occur in perfectly well written code. The question is what to do about them...

Right, and I gave my $0.02: avoid them unless completely necessary. Just because they occur in well-written code doesn't mean every or any particular use is ideal. Neither does a solution necessarily have anything to do with globals, because as I implied, it's valid for objects to refer to instances of systems.

[Generally speaking, the more redundant pointers there are to something the more likely it is that something is going to break.]

Well as Bateleur mentioned above, "bullet->canvas" involves caching a pointer to canvas in bullet. Generally speaking, the more redundant pointers there are to something the more likely it is that something is going to break. For example, if the canvas in this case is destroyed and replaced with a new one, and bullet's cached copy isn't updated, you would either end up crashing, or, if you were using smart pointers, accessing a stale version of canvas.

To my mind the more redundant data structure traversal, the more likely it is that something will break. What if Jimbob fires his weapon then it blows up, falls in a bottomless pit, etc.? Suddenly the entire scheme breaks down, as a bullet is associated with a nonexistent gun. Or, put another way, this scheme relies on the gun, character, and room pointers being correct, which is a taller order than just the right canvas- and it imposes severe limitations anyway. If the canvas can change it should just aggregate a list of every object (which it ought to do anyway) and change their canvases.

If you like, but then "the renderer" is the thing you need the pointer to! What you can't always do is have the renderer know how to paint everything without further input, since that tends to mean very poor code modularity.

Then the canvas can just refer to every object on it, which implements a drawSelf() function, which itself just calls lower-level draw(blah blah) functions on the canvas itself. The default implementation could be used most of the time, but objects could still override it with more special stuff if they have to.

EDIT- In fact, with a small modification of the above scheme no object needs to even refer to a canvas- it can just be passed it in drawSelf(canvas&).

I still don't understand the design you have. What does a bullet have to do with a gun, apart from the fact that it created it? Similarly, a character just holds a gun, and a room just has a character within its space. Technically, a character could be standing outside of any room, and a gun can fire when noone's holding it (both these happen all the time IRL). Unless there's something you haven't said about it, I still reckon you need to flatten your structure a bit more.

a puzzle-exploration game that is philosophical, and quiet...
nuances of storytelling via environmental cues...
game [that] places a heavy emphasis on the way things look

While I have liked/loved a few games that fit this description, it's not enough to get excited about for me, yet. It's obviously not a Braid sequel anyway.

MS Visual C++ with a bunch of libraries. Though I would gladly switch to a better alternative if I found one, which isn't for lack of trying.

^ Maybe I'm being naive, but what's the justification for anything other than "bullet->canvas"? AFAICT "bullet->gun->character->room->canvas" fails to be any safer, has a maintainability issue, performs worse (which probably doesn't matter) and is less clear to my eyes. Could be wrong but I assume it's typical of every node in a scene graph to refer to its world/manager/whatever.

Saw a phat list of tests with false and true positive rates for various antivirus software, and Antivir beat all other free ones and many of the payed ones. Might try to find that again. Anyway, false positives are a fact of life today, just add cncs32.dll to the exception list.

^ Must not have been clear, but I said that a generic RTS is equivalent to an action game in terms of conscious thought. But yeah, a good turn-based strategy or tactics game rewards it.

In any case, another interesting relationship between conscious and subconscious thought in games is that the latter generally wins out, given enough time. Learning is an important part of game enjoyment, and one way it manifests is in automation of responses that the player had to ponder before.

I'm nearing the stage where I have to make an editor for my project. Pretty cool list here. Minor issues:

12. Lots of pre-computing before actually playing the level

Sorry to be pendantic, but I think you mean "lots of downtime before playing". Heavy pre-computation is surely fine as long as it's distributed through time in a way that doesn't waste it for the user.

8. Keyboard-press-only interface (e.g. press "g" for floors, "h" for gun...)

Keyboard interfaces can be much faster to use than mouse-driven ones, and no less clear. Of course placement should be driven by the mouse, but the keyboard is great for switching between modes and objects (obviously not in the way you mention), or invoking functions. Blender demonstrates that nicely. I've dicked about in editors that use the keyboard alot, and that was never a major problem with them.

@Mawy_Golomb: Ah, thanks for the clarification.

There are plenty of games that emphasize conscious thought. Any good game of strategy ir tactics does, although some tend to reward intuition (and a generic RTS on the other hand is just a quasi-action game). E.g. Shogun Total War, still my favorite strategy game after all these years. Certain tactical shooters do, such as Ghost Recon (on occasion) and Operation Flashpoint.

I only like adventure games that require conscious thought, especially the Myst series (the ones made by Cyan). So yeah, although I think we could do with more conscious thought as a requisite for games (especially since I tend to like such games more), it's not exactly hard to come by today.

Really? I thought that spatial thinking meant that you brain doesn't go towards subconscious thinking. With games where you mainly use your reflexes, doesn't that go against spatial thinking, which is all about the conscious mind? Could you explain to me the difference between the two?

You might be right, but I never heard it used in that sense. I just associated "Spatial thinking" with "spatial cognition"- thinking about spatial relationships. Google seems to agree.

I think you're talking about the difference between conscious vs unconscious, intentional vs automatic, or deliberative vs reflexive thought. Don't see how spatial thought relates.

EDIT- Sorry, didn't notice that you're the author. This part is extraneous.

@Inanimate, thank you sir!

Here's another couple loops. A low-key percussive drone, and the first hip-hop beat I've made.

^ I'm doing "fake" scrolling for this would-be awesome top-down shooter as well. Basically I subtract the delta of the midpoint of the player and crosshair from the position of every object each frame, giving me mouselook. On top of that I subtract the player movement. Turns out KnP handles events in weird ways though- it doesn't just run through the list each update; input events are triggered in non-deterministic orders. Collision works if I keep the mouse still, but by moving it around really fast, I can pass through obstacles easily. So KnP's poor consistency is the nail in the coffin for this game. Ah well.

TGF introduced groups and file I/O as well, right? And extensibility? I mean, it was kind of the first non-gimped Klik creator.

^ Haha, I dig the ending of that game. To anyone playing, it's not obvious, but you have to work for it- sort of.

Do you want me to upload anything else out of KnP retail?

Right, KnP really hates scrolling. Looks like my only choices are either to start the player at the bottom-right of the level, or make a ridiculously convoluted scheme where unique IDs for each object are associated with memory-like offscreen positions so that ghost objects storing their "real" positions can be associated with them. Yeah.