Welcome, Guest. Please login or register.

Login with username, password and session length

 
Advanced search

1411283 Posts in 69325 Topics- by 58380 Members - Latest Member: bob1029

March 29, 2024, 12:50:33 AM

Need hosting? Check out Digital Ocean
(more details in this thread)
TIGSource ForumsDeveloperDesignVisualizing kinesthetics
Pages: [1]
Print
Author Topic: Visualizing kinesthetics  (Read 893 times)
FNKVSSL
Level 0
***



View Profile WWW
« on: September 26, 2015, 07:00:56 PM »

It occurred to me a few years ago that the sensation of motion is the facet of gaming I cherish above all else, so I'm drawn to anything that involves jumping and dodging shit. I'm really into shmups in particular, as they represent the purest translation of movement to the screen- it's abstracted dance choreography, all joy is the product of rhythm, reaction and improvisation. I'd argue that shmup quality is largely dependent on bullet pattern novelty (i.e. variety of movement). I wanna talk about that.

Case study: Radiant Silvergun prioritized novel movement with 1) combat that occupies every region of the screen, 2) A variety of weapons that each target a particular quadrant of space (front, sides and back of the ship at varying ranges), 3) a horizontal aspect ratio coupled with an unusually slow player speed to maximize the possible range of movement, 4) bullets with varying properties that move in nonlinear fashion (homing shots, bullets that bounce, gravity wells that suck the player in etc). The developers understood that movement was the core of the game, so every combat scenario is built around herding the player into a particular pattern of movement. Smart.

Compare with just about any Cave shmup where 1) all action is constrained to the bottom of the screen because enemies spawn exclusively from the top, and 2) all shots are identical and move linearly from 1-3 points of origin. The developers prioritized spectacle with oodles of visual Juice, while the movement is characterized by short left and right taps because dense, porous waves of projectiles encourage the player to move as little as possible. So the emphasis is on deciphering and reacting to patterns in visual noise, which is fun, but it's only a subset of Radiant Silvergun's flavor. This method defines the majority of games in the genre, it hasn't progressed much since Batsugun (1993). And then people wonder why the genre's dead.

So here's the rub: I've written whole tracts about this over the years, and every time I do these goddamned nerds just blow me off because Radiant Silvergun's "overrated/too slow/too long/the scoring sucks because it only wants me to shoot 1/3 of the enemies in each stage". So what's my thesis missing? Cold, hard data, for starters. I need to actually show the discrepancy in player movement somehow before I can get into why it matters. I've thought about doing some motion tracking over game footage to try and plot the player's movement, but that seems impractical and time consuming and I'm lazy. Then it hit me earlier today, why not automate the process by recording player inputs?


SO, any bored coders out there want to burn an hour or two? Here's the pitch: A tool that logs every press of the arrow keys to a text file with HTML color codes (ROYGBIV) assigned to each direction. Merge simultaneous button presses into a single character for diagonals. Here's your symbols. Needs a hotkey to start and stop recording with each session saved incrementally. For extra credit, maybe try increasing point size according to how long the key is held. Maybe get Ascii Draw guy in on this?

Once I've got that, I can play through some games with joytokey and try generating some images to get a visual representation of the player's movement. Hopefully the payoff will be some images akin to condensed frame samplings of feature films that showcase the color and mood over time. I can even fill in the holes for a cleaner visual. It doesn't need to be shmups either, there are a handful of games I'd like to scrutinize to gauge the complexity of inputs demanded- though obviously most of them are gonna be and . The data could seed some more sophisticated visualizations down the line, like graphs and shit. I bet some really interesting patterns are going to arise from some of these boss encounters- or most likely we'll just wind up with a bunch of meaningless noise. We could possibly learn something, anyone else on board? No? Shucks

In which case you can just treat this thread as Kinesthetics General because I want to see some discussion on codifying Game Feel. I want Labanotation for games.
« Last Edit: October 06, 2015, 12:18:01 PM by Ossiferous » Logged

diegzumillo
Level 10
*****


This avatar is so old I still have a some hair


View Profile WWW
« Reply #1 on: September 26, 2015, 11:08:17 PM »

Sounds like you need a key logger. There are a bunch of softwares that do that, you'll probably find one that spits out exactly the kind of file you want or allows you to customize it.
Logged

Si13
Level 0
*


View Profile
« Reply #2 on: October 05, 2015, 06:25:47 PM »

Why is no one replying to this thread??? Kinesthetics is, like, 99% of any good action game.

I don't think input complexity matters very much -- or at least the kind of complexity you could measure and analyze. Like Super Mario Bros., for example. Maybe Mario isn't really what you had in mind when you made this thread, but that game is millisecond by millisecond psychology. All you're really doing is holding down b, pressing right on the d-pad, and pressing a from time to time. occasionally you'll pull back on the d-pad. You'll almost never let go of b. But because of the inertia, because of the momentum, you don't have true control over Mario. Moving in that game is a struggle through time and physics.

In Mario you can

A) Run,

and,

B) Jump

Running is subject to friction. If you let go of the d-pad you stop. If you press the opposite direction it takes less than a second to turn around. You hear a little screech and watch Mario grind to a stop. Turning around is exciting because, usually, if you're turning around it means something a few pixels ahead of you is standing around and ready to kill you. You have to make extremely precise calculations in your head when turning around -- turning around takes time and it takes space. If you're off you die. You have to replay the level. If you run out of lives you have to replay the game. That's pretty high stakes.

But when you jump there's no friction at all! It's also a lot harder to turn around. That is, momentum in the air is pretty crazy -- you lose the control that you had on the ground. Jumping is a commitment -- you can adjust your trajectory, but only slightly. Just like how turning around is a commitment. The game turns into choosing what to commit to, and every 300 milliseconds you're making new commitments, adjusting the magnitude of your commitments, and watching them play out. The watching them play out part is more important than the level of input complexity, I say.

So, uhh, I think you'll learn more just by playing Radiant Silvergun and thinking about what the game is forcing you to do, forcing you to struggle with, and thinking about how it shows you that struggle than you will by mapping out all the inputs. But man! I haven't played that game yet. By god I need to. Have you played Bangai-O Spirits? That's probably my favorite game. Anything that Treasure's put out I need to give a try.
Logged
diegzumillo
Level 10
*****


This avatar is so old I still have a some hair


View Profile WWW
« Reply #3 on: October 05, 2015, 06:27:34 PM »

Why is no one replying to this thread??? Kinesthetics is, like, 99% of any good action game.

Because he is not starting a discussion on kinesthetics. He is asking a technical question of how to record someone playing so he can study kinesthetics.
Logged

Si13
Level 0
*


View Profile
« Reply #4 on: October 05, 2015, 06:32:13 PM »

In which case you can just treat this thread as Kinesthetics General because I want to see some discussion on codifying Game Feel. I want Labanotation for games.

Though I guess I didn't say much about the whole Labanotation in games thing.
Logged
FNKVSSL
Level 0
***



View Profile WWW
« Reply #5 on: October 06, 2015, 11:47:26 AM »

So, uhh, I think you'll learn more just by playing Radiant Silvergun and thinking about what the game is forcing you to do, forcing you to struggle with, and thinking about how it shows you that struggle than you will by mapping out all the inputs.

True that. I'm not going to find any substitute for physically playing a game, I'm just thinking about how I might communicate this action to someone who's unwilling or unable to play it. For instance, now that I'm working on a game with a team, if I have a boss fight in mind, what would be the most efficient way to relay that to the programmer? The only way to get 1:1 results would be for me to animate the action by hand with some basic shapes, but that's time consuming. I tried sketching out the motion- game's basically just Castlevania with 8-way whipping and a grappling hook- but there's no telling how much is going to be lost in translation. I had no shorthand for velocity, so anything relating to how an object's speed changes over time would need to be explained in text, like how the boss smoothly slides from side to side in puddle form.

I started looking into motion graphics for a solution (I was convinced my infatuation with Radiant Silvergun was telling me mograph was my true calling anyway). The way After Effects controls velocity is with a graph, where a change in height indicates a change in speed. It isn't the simplest thing to read, but then I went looking for an introduction to the speed graph to post here and saw that

had broken it down really well in the first 3 minutes. Maybe I could just scribble tiny curves like those everywhere? If there's any possibility of a simple catch-all notation for game movement, the solution's probably in here somewhere.

But yeah, I sort of cooled off to the idea after posting the thread. I'm still curious about seeing a visual translation of the movement in this particular game, but it wouldn't be useful, just neat. Inertia is super important, glad you brought it up. Love that you used three paragraphs to break down three seconds of Mario gameplay. That's the resolution of scrutiny I want to see, because we really are making a lot of complex decisions on a millisecond basis. Hate to bring it back to shmups, but it just occurred to me that the genre thrives on bullet hell with minimal input because every movement is swamped in exciting psychological noise: I find many deaths are the result of botched inputs because the visuals sparked a left in my thumb when I really wanted right. Hardcore players swear by joysticks, I guess because having to use your entire wrist to actuate movement minimizes the likelihood of twitch error. A shmup with ice physics would be anathema to those players, but man, wouldn't that be neat? What if the player was allowed to move freely about an area larger than the screen? What if the screen rotated? It would take so little to put a spin on the formula. Ice Shmup could be really fun.

Changing it up now because I like the idea of deconstructing games on a case-by-case basis. Take the original Tomb Raiders and Croc: Legend of the Gobboes- games I love that were well received at the time, but the tank controls wouldn't go over well today. The disconnect of movement not matching the character is a powerful turnoff, as if Lara's a car that needs to go into reverse to line up every jump. But I like that weight. I like how you have to steer Croc like you're banking an aircraft, that the cobblestone path textures project and anticipate the movements you're meant to make. That the simplest act of grabbing a row of crystals requires intense focus on my part makes it the most engaging collectathon of all time to me. I think there's still a place for that type of control if we could contextualize it properly, contrast is part of appeal.

Oh and yeah, Bangai-O's a good time! I think. I played the N64 version years ago, it's not fresh in my head. Actually meant to start playing the Dreamcast version recently but got sidetracked, gotta get back on that.
« Last Edit: October 06, 2015, 11:58:43 AM by Ossiferous » Logged

Pages: [1]
Print
Jump to:  

Theme orange-lt created by panic