next »

[Bezzy]

I wish I had just a quater of the knowledge required to pull off some of this stuff:





I mean, come on! 64k!

Man, I really wish I could learn to rock it in the demo scene. After my current game, I think I'm going to have to try and learn from these guys. Quite a lot of the stuff I can sorta deconstruct and figure out a rough way to do it (albeit inefficiently), but a lot of what makes this stuff great is just really good art direction - might be much harder to learn than the coding techniques (for me, atleast).

I discussed this a bit more in a recent blog post.

[Zaphos]

If that were a game, I can envision how the TIGSource review would look ...

"I’d probably have a bit more to say about the game if it would run on my (shoddy) laptop. But it doesn’t. The screenshots and trailers look sweet, though!"
 - BMcC

The 64k restriction doesn't really do much for me, but it does look pretty sweet!

[PoV]

I wish I had just a quater of the knowledge required to pull off some of this stuff:

Outside the per pixel lighting (I'm not even sure there were any shadows), you have HDR color space effects (overblowing colors, with the idea overblown colors glow), fullscreen blur or depth of field blur (probably just be a fullscreen blur that doesn't toy in the Z buffer), Subtle focal length tweaking in the camera for the dirty pans, and a slew of procedurally generated things (clouds, all or most of the geometry).

Not to mention the metrics of the fact that they don't need interactivity.  So like film editing, they can cheat and only show what's cool.

Don't get me wrong, it's a brilliant display in the tradition of the scene.  Just depending on your 3D knowledge, you're just shaders, full screen effects (with shaders), fancy use of 3D camera, and some effective experiments with procedural content away.

[BMcC]

If that were a game, I can envision how the TIGSource review would look ...

"I’d probably have a bit more to say about the game if it would run on my (shoddy) laptop. But it doesn’t. The screenshots and trailers look sweet, though!"
 - BMcC

EDIT:  It is a valid point, though...  I hate it when games only run on the latest-and-greatest machines.  (And I'm disappointed when they don't run on my Average Joe machine.)  Do we really want to see that graphics pissing match in the indie world?

Also, I doubt a game that looked like that would be compelling.  The coolness of that video depends on the camera movements, the scene changes, the abstract effects...  I don't know how you'd be able to fit gameplay in there without reducing the graphics to a visualization.

But for what it is -- very cool!  Any more links?  Is there an established King of the Demoscene Videos?  I'd like to see that.

[Terry]

This one really wowed me when I first saw it, though it's a bit old now. Should run on pretty much anything too:

[Bezzy]

I wish I had just a quater of the knowledge required to pull off some of this stuff:

Outside the per pixel lighting (I'm not even sure there were any shadows), you have HDR color space effects (overblowing colors, with the idea overblown colors glow), fullscreen blur or depth of field blur (probably just be a fullscreen blur that doesn't toy in the Z buffer), Subtle focal length tweaking in the camera for the dirty pans, and a slew of procedurally generated things (clouds, all or most of the geometry).

Yep! Any of those individual things, I think I could do. It's just applying them artfully which I'm a bit scared about.

Not to mention the metrics of the fact that they don't need interactivity.  So like film editing, they can cheat and only show what's cool.

My latest blog entry says basically that, so yeah, i agree

Still, very inspiring stuff for me.

[Bezzy]

EDIT:  It is a valid point, though...  I hate it when games only run on the latest-and-greatest machines.  (And I'm disappointed when they don't run on my Average Joe machine.)  Do we really want to see that graphics pissing match in the indie world?

Leave no path unexplored! Besides, today's funky chicken graphics are tomorrow's acceptable standard.

Also, I doubt a game that looked like that would be compelling.  The coolness of that video depends on the camera movements, the scene changes, the abstract effects...  I don't know how you'd be able to fit gameplay in there without reducing the graphics to a visualization.

In the wrong hands, sure. Certainly it harkens back to the long standing general failure of developers to marry fixed narrative and interactivity*... but funky camera stuff in gears of war etc. don't necessarily hurt the game play, so I think there may be a middle ground, even if it's tiny.

I'm not saying it'd be easy, but games have a big space to explore... a solution to this conundrum might be out there, somewhere.

But for what it is -- very cool!  Any more links?  Is there an established King of the Demoscene Videos?  I'd like to see that.

Lots more if you look around http://www.pouet.net/ and http://www.scene.org/.

*even though i believe the L vs. N is solveable through lots of masterful use of emergence - but it'll be a while before we're well enough versed in complex systems to have the kind of intentionality which sees simple systems giving rise to generally interesting narratives.

[Anthony Flack]

I think it's even more than that. I think the difference in visual quality also relates to the added difficulty of making a good, playable game. The endless tweaks and reworkings; the constant juggling of different elements; trying to balance what looks good with what works well. It all adds up to a ton of extra work, and you only have so much enthusiasm and energy to push a project through. I think half the reason these demos look good is because they are such small things.

Trying to apply that level of polish to a full game without losing sight of the gameplay along the way is the sort of thing you generally need a bigger team for, if you're going to make it through the process alive.

Also, I think that these 64k demos do often rely on high-spec machines as it's all about the fancy rendering - since they don't have the space to make it look cool with old-fashioned texture and model work.

But I do think that cool camera angles could be put to use in games more often, and not just for scripted sequences. I quite like the look of locked-down or on-rails cameras. And sometimes it's a good thing to restrict freedom of movement within the 3d environment.

[AdamAtomic]

I think the points about camera movement and direction totally stand (As well as skimping on AI and interactivity), but if you assume these demos slam your machine because they look so dang good, then you are at least 95% wrong!  The amazing thing about these demos is not just how good they look period, or how little space they occupy or require for download, but they are generally VERY fast and efficient programs (outside of some massive loading times for procedural textures in some).  indie games can't look like this because there are only a few people in the whole world that are good enough programmers to make this stuff, and they're busy making demos

EDIT - for examples of games that WERE made my demoscene guys check out:
http://www.riddickgame.com/
http://www.housemarque.com/
Pretty, huh!

[Impossible]

I agreed with Bezzy, there should definitely be more games as stylish and attractive as scene stuff, but all of the problems brought up in this thread are major issues that would have to be overcome. Another trick is that a lot of what makes demos appealing, beyond the graphics and even the camera work, is music synchronization.

If you wanted to make a game that looked and played like a demo, I think the best bet might be to go with some kind of rhythm action game.  A rhythm rail shooter along the lines of Rez is perfect for a demo.  You have a relatively fixed camera path, and music sync as part of the game.  Just go crazy with the effects and game play and you could have something golden.

[MattC]

If you wanted to make a game that looked and played like a demo, I think the best bet might be to go with some kind of rhythm action game.  A rhythm rail shooter along the lines of Rez is perfect for a demo.  You have a relatively fixed camera path, and music sync as part of the game.  Just go crazy with the effects and game play and you could have something golden.

The first thing I thought of when I read this was Jeff Minter's

which was made with the same Neon platform he used to make the Xbox 360 music visualizer. Though, this is somewhat opposite from a demo scene video as it is reactionary to the music instead of being highly scripted around it.

Also: Space Giraffe has phrases like "NAD GET" and "GIRAFFETASTIC" over the screen instead of depressingly ponderous mantras..

[Akhel]

EDIT:  It is a valid point, though...  I hate it when games only run on the latest-and-greatest machines.  (And I'm disappointed when they don't run on my Average Joe machine.)  Do we really want to see that graphics pissing match in the indie world?

Don't complain. My computer is a Duron 950mhz with overheating issues.

[BMcC]

[...] Leave no path unexplored! [...]

Heck yes.  I like your style, kid! 

[...] I'm not saying it'd be easy, but games have a big space to explore... a solution to this conundrum might be out there, somewhere. [...]

Also very true.  But Anthony 'Flak Cannon' Flack articulated some the difficulties that would arise quite well, I think.

[...] but if you assume these demos slam your machine because they look so dang good, then you are at least 95% wrong!  The amazing thing about these demos is not just how good they look period, or how little space they occupy or require for download, but they are generally VERY fast and efficient programs [...]

Is this... true?

[...] indie games can't look like this because there are only a few people in the whole world that are good enough programmers to make this stuff, and they're busy making demos [...]

Pfft, BAN-worthy. 

Seriously, gimme a break.

EDIT - for examples of games that WERE made my demoscene guys check out:
http://www.riddickgame.com/
http://www.housemarque.com/
Pretty, huh!

LOOKS LIKE PLASTIC TO ME 

Don't complain. My computer is a Duron 950mhz with overheating issues.

I'm on a Speak & Spell... with Windows ME. 

[Steve Swink]

The first thing I thought of when I read this was Jeff Minter's

which was made with the same Neon platform he used to make the Xbox 360 music visualizer. Though, this is somewhat opposite from a demo scene video as it is reactionary to the music instead of being highly scripted around it.

I think a lot of the demo scene stuff is amazing, beautiful, and inspiring, so I agree with Bezzy and Impossible. Whatever the constraints, it's an avenue to explore.  As indies we can't do [whichever recent game you think is most graphically pretty but which relies on a ton of art production muscle] but we sure as hell can do stuff that's cheap, stylized, different; demo scene aesthetics, essentially. Also, I think that something as beautiful and polished as the stuff Bezzy's linking above would be that much better if it reacted to user input. The interactivity/scripted tradeoff can favor interactivity in the right hands, Rez proved that in its cool but limited way. 

Space Giraffe is funky and interesting but lacks the slick coolness of a lot of the demo scene stuff IMO. Minter is a beast and I dig his beasties, but I found his demo of Space Giraffe at GDC to be somewhat nauseating (and I can see how its obvious psychedelia visuals could be offputting to some.) It points to a great venue for this kind of thing, though, one which addresses a lot of the constraints. Xbox and PS3 have graphics muscle to spare and appropriate delivery platforms for niche content. In fact, when I worked for Neversoft one of our graphics programmers won a Sony sponsored 64k competition on the PS2.

@BMcC - bravo! And, lest we forget, Spore. Nuff said.

[mike]

One demo group has made a game http://kk.kema.at/files/kkrieger-beta.zip using the same techniques they use to produce their demos.
This game weighs in at 96kb. Pretty impressive once you've played it.

Of course, the game plays rather poorly... Not very challenging or very good in terms of games... But it is an example of what Bezzy is talking about making... Even if it's not a good example...

-Mike

[shinygerbil]

The first thing this brought to mind for me was Darwinia by Introversion Software. I'm sure you've all heard of it, whether you like Introversion or not.

The intro always reminds me of something from the "demoscene" - both from the visual style and the music, which I really enjoy; this feeling carries over into the game. As well as this, Introversion use a lot of procedurally-generated content in Darwinia, and if you read their blog, they are experimenting a lot with such content for their next project(s).

For me, Bezzy's original point still stands, but it's obvious that the game is first and foremost; if you can design an awesome game and then enhance it with some crotch-grabbingly gorgeous visuals without disturbing the gameplay too much (or better, at all) then you can only be on to a winner. Well, maybe that's not true - of course, it depends on the game's style in the first place - I can't see Balding's Quest having much use for raytracing, for example ;P

If we look at a game like Darwinia, though, it was clear that the visuals were intended to be stunning, and up to a certain point, making the graphics prettier would definitely improve the game (so long as the style was not disturbed). Another game that has been mentioned in this thread - Rez - I have never played, or even seen much about, but from what I am given to understand, the same principle applies; added graphics and prettiness would not really detract from the game.

OK, so perhaps I'm overstating myself a little, but I definitely feel that there is something to be gained from tapping into these kinds of things. The wonderful thing about indie gaming (from a player's point of view) is the diversity of the games and their various influences. Inspiration can come from anywhere, so why not the demo scene?

[Chris Whitman]

indie games can't look like this because there are only a few people in the whole world that are good enough programmers to make this stuff, and they're busy making demos

Because... they know how to generate Perlin noise? They can write fragment programs? They have grasped the vast subtleties of the marching cubes algorithm?

None of the things done in these demos are beyond an undergraduate level of understanding. All the procedural stuff uses widely known algorithms which have been around for years. Beyond that, they get their cheezy demo-scene look by lifting colour schemes and fonts out of late nineties shooters.


I'm not saying that any joe off the street could just sit down and whip up one of these programs, just that it does not require 'world class abilities.' Time, effort and a bit of research would be sufficient.

[Xion]

It's not the knowledge that makes things pretty. It's how you apply it.


...I dunno. I've actually been wondering the same question for a while, too.

[siiseli]

I was actually at the party where this was released :D It was awsome

[ஒழுக்கின்மை (Paul Eres)]

Okay, a better question: even if it'd be hard to make independent games that look this good considering that demos are not games and are not restricted by the restrictions of gameplay: couldn't we at least make *intros* and *endings* to our games that looked this good?