The future of SGDK

[v6v]

Makes sense. I've always wondered what would account for the next update, desktop wise- but it seems the desktop implementation is pretty solid at the moment; the web implementation will receive all of the focus.

I don't want to skip too far ahead, but with SGDK2 being this powerful already (Anything is possible now without limitation), what on Earth would be in SGDK3?

[bluemonkmn]

I think there's lots of room to enhance the usability/ease of use and merge game development with game playing kind of like Minecraft.  In Minecraft you are constantly creating structures while you are playing.  What I would like to figure out is some game/engine where anybody can extend the game in an even more flexible way than Minecraft.  You would be able to submit not only alterations to the tile map, but new sprites and rules... essentially defining your own objects and recipes.  Of course there has to be some control on it.  The submitted modifications would have to go through a community review process and only the best would get into the game.  The result would be a crowd-sourced game.  You get immense worlds populated and tested by many people.  You could enjoy playing the actual game(s), playing/reviewing submitted modifications, or making modifications of your own.  But this idea is a long way from implementation.  I'm not even sure I can handle it or what kind of game the starting point should be.  I think I'll need a lot of help on this one if it's ever to get done.  It probably won't be called SGDK3, but it will essentially be the next step in game development.  Instead of just casual gaming, we'll have casual game development.  I suspect many people don't have the attention span even to get into tool/kit-based game development any more, and need to be enjoying themselves as they are creating.

[v6v]

So it would be some sort of crafter where you build the world tile by tile?

This idea is so innovative that I'm having a hard time trying to find something I can compare it to.

I guess LBP2 could work?

I wonder- the game itself would start small, but eventually gain in filesize with all of the additions- I know this idea is still barebones, but would it be an inetwork based application?
(And by that I mean would the additions become stored on a server and then sent to the player on runtime?)

I'm also wondering about the rules- are they the SGDK2 rules inside of an Open Source project, or some method of defining rules while already still in the game, through some sort of script?

But this idea is a long way from implementation.  I'm not even sure I can handle it or what kind of game the starting point should be.  I think I'll need a lot of help on this one if it's ever to get done.

You've single-handedly created a program that holds a flame to Game Maker and MMF2, (I wish more people could see that) transitioned from DirectX to OpenGL, and added an HTML5 implementation.

If you need help to make this, it has to be something insanely impressive. What would you need help with exactly?

I'm not sure if this is relatable, but in Doom Based games, they have a .wad system- which allows users to create their own maps, sprites and skins. This is one of the few ways I can comprehend how this would work. Would it be more like an ingame version of Minecraft's mod system? I'm also even thinking of the game Spore.

Sorry, again, it's just that I can't comprehend this without comparing it to something else. (Which is a good thing- it means this idea is either extremely unique, or profoundly complex)

You would be able to submit not only alterations to the tile map, but new sprites and rules... essentially defining your own objects and recipes.  Of course there has to be some control on it.  The submitted modifications would have to go through a community review process and only the best would get into the game.  The result would be a crowd-sourced game.  You get immense worlds populated and tested by many people.

Rumor has it that Notch plays all of the mods and simply puts in the ones that he remotely likes. The game in a way is being built for him.

I think I can summarize:

So the end result will be a sandbox-ish game where the players can define their world as they play, not by clicking blocks, but by creating it themselves.

[bluemonkmn]

So it would be some sort of crafter where you build the world tile by tile?

This idea is so innovative that I'm having a hard time trying to find something I can compare it to.

I guess LBP2 could work?

I don't have a firm grasp on the idea myself yet.  But one of the thoughts I had is this.  Have you ever had a moment in a game where you thought to yourself "I wish I could open this door" or "what is in this direction, and why can't I ever go this way" or "if only I could just do such and such here"?  I want to make it possible for people to fill in those wishes themselves. Then you don't have to rely on the limited creativity of one or few designers to fill in interesting seemingly infinite details required to more truly represent a full realistic world.  And because people get to choose which areas they are expanding and vote on what they like best, you're likely to get much higher quality in every detail.  I guess LBP2 is a good place to start.  Like LBP2 people would be able to submit their own content and vote on it.  But, I think unlike LBP2, the content would be more continuous and integrated yielding a larger more complete world.  Of course this is all pie in the sky, and I have a hard time believing this is actually as great as it looks in my imagination, but I think it's worth a try if I can muster up the determination/energy.  I'm half wondering if I should just post the idea in the forums (as I'm doing now) allowing someone else to steal it so I don't have to do it .  I suppose publicizing the idea is a little risky if I want to be the one to develop it.

I wonder- the game itself would start small, but eventually gain in filesize with all of the additions- I know this idea is still barebones, but would it be an inetwork based application?
(And by that I mean would the additions become stored on a server and then sent to the player on runtime?)

My thought is that there would be some "base image" (no clue what kind of format it would be in) that represents the game.  And modifications and additions could be applied to graphics, logic and/or level design in optional "packages".  When you play the game, you can choose the base image, or the base image with some modification(s) or extension(s) applied.  They would be downloaded from a server, but I think it would be nice, after downloading, if you could play and replay offline, needing only to be online when you want to download or submit packages.  Perhaps the "base image" would be periodically updated with the highest rated modifications.  It's all still very fuzzy in my head.  Of course some modifications (such as level design) may depend on other modifications (such as graphics enhancements), so I suppose there could/would be a kind of hierarchy of modifications to maintain, and when you vote on one, you're voting on everything on which it relies.

I'm also wondering about the rules- are they the SGDK2 rules inside of an Open Source project, or some method of defining rules while already still in the game, through some sort of script?

I don't know really -- it's hard to imagine SGDK2 rules as they stand today being something that can be submitted in a package that others can download and merge.  Maybe the rules would be javascript-based, and a package would just be a JS file or JS content that gets merged into the final result of a client downloading an HTML5 page/package.  It's still not user friendly enough for non-coders, but might be a good starting point.  I think we need more artwork than coding anyway.  If I have a good set of starting rules, hopefully there will be plenty of time or plenty of coders to work out enhancements to the rules that the artists would want to work with.

If you need help to make this, it has to be something insanely impressive. What would you need help with exactly?

Probably anything that anyone is really good at.  I might be able to do it all on my own, but it would just take forever.  So if, for example, there are some good PHP/MySQL coders out there, maybe they could help work out an ideal mechanism for representing these update packages... hopefully with an eye toward scalability in case this thing takes off.  At one point my idea was even more "meta"... not only would people submit content for game modifications, but also modifications for how the whole system operates -- submitting PHP files and such to improve the very framework itself.  But I think that's taking it too far.  Yes, it might mean less work to get started because we would rely on others to do practically all the work -- all we start with is the idea and a simple Wiki on which anyone can make updates , but I think we do need to have something functional out there before anyone would be interested (and there are security issues of course). I was thinking like Wikipedia except instead of the content being knowledge, it's a game framework and content.

Sorry, again, it's just that I can't comprehend this without comparing it to something else. (Which is a good thing- it means this idea is either extremely unique, or profoundly complex)

That makes two of us. I think there's something out there, and we may need a whole crowd of brains to find it... but I don't know if there's anything to motivate that crowd of brains to be put to this task.  I also don't even know if this idea is worthwhile/original and has the potential to be productive and interesting, or if it's just going to fall apart, fracture, or devolve into other already-existing projects.

Rumor has it that Notch plays all of the mods and simply puts in the ones that he remotely likes. The game in a way is being built for him.

Yeah, I was hoping SGDK2 would allow people to build games "for me", but it seems they need more of a starting point to coax them into it if I want more games.

So the end result will be a sandbox-ish game where the players can define their world as they play, not by clicking blocks, but by creating it themselves.

I do want to avoid too much of a sandbox-ish feel, though.  I want there to be definite storylines and tasks.  I just want those storylines and tasks to be extendable and expandable into a much larger scale than one sees in typical independent games.  And I want people to be able to be more productive during their "play" time.

[v6v]

Well, after reading everything, I can say a good place to start would be the CleanGame project you're working on. I had played it 2 days ago and was impressed greatly with it. If you felt like changing it to accommodate players- that wouldn't be a bad idea, but the gameplay is extremely solid.
It seems to be open enough to allow anything to be made.

Ambitious dreams need to be followed. I have no clue what I'm doing and I've reached farther than I could have imagined.

I love this idea. It would really put the HTML5 feature to good use- especially with the .js rules (Maybe some sort of *for lack of a better tern* 2nd level script that allowed people to make rules?)
(Another question, there are some things I specifically enjoyed about CleanGame *This might be off topic* are you continuing with that? I loved the HUD for items, and the general gameplay elements like putting out fire. The game could become incredibly popular if you released it.)

[v6v]

Oh, and in canon to the original topic...

http://www.youtube.com/watch?v=IWxyzzpD5dM

I've decided to test the 3 dimensional map generation with the sample project.

The only issue I have it a strange banding of textures when the viewing matrix is at an awkward angle.
That and the same spacing issue.
I can fix the spacing however, the banding I'll have to go back to the OpenTK site to look into.

I really wish someone would make one of those old timey NES RPG's, or a neat racing game with this. It would be perfect in a top down game.

[bluemonkmn]

Oh, and in canon to the original topic...

http://www.youtube.com/watch?v=IWxyzzpD5dM

I've decided to test the 3 dimensional map generation with the sample project.

The only issue I have it a strange banding of textures when the viewing matrix is at an awkward angle.
That and the same spacing issue.
I can fix the spacing however, the banding I'll have to go back to the OpenTK site to look into.

I really wish someone would make one of those old timey NES RPG's, or a neat racing game with this. It would be perfect in a top down game.

That's pretty neat.  Maybe it would be easier to take this to the HTML5 platform instead of having to go back to OpenTK forums for all your issues.  But I suppose you're not targeting the web with what you're doing right now.

No time to reply to the other message at the moment, but I do want to reply.

[bluemonkmn]

Well, after reading everything, I can say a good place to start would be the CleanGame project you're working on. I had played it 2 days ago and was impressed greatly with it. If you felt like changing it to accommodate players- that wouldn't be a bad idea, but the gameplay is extremely solid.
It seems to be open enough to allow anything to be made.

I think I kind of gave up on that because I didn't know how to balance the challenge of finding unique and rare items against the need to be forgiving when the player uses up items in the wrong way.  In my mind, the main goal of playing a game is to collect items that allow you to proceed through puzzles, and I'm having a hard time breaking out of that, which I think I need to do somewhat to make a game that is both challenging and forgiving/fun.  But I can probably build on what I have to still do that, just with different kinds of challenges.

Or maybe I just got distracted by the idea of realistic physics and haven't come back to CleanGame yet because nobody seemed particularly enthusiastic about it.

Anyway, if I am to build on this project for the new idea, the first step, then, would probably be to figure out what I need to do to make it work as an HTML5 project.  I should probably export it as HTML5, see what's broken, and implement the rules that need implementing for HTML5.  Maybe I can make something playable, then release that, then come back to the idea of a patchable, extendable game based on what we end up with.

For that matter, maybe it doesn't need to be terribly user friendly to be popularly extendable by the more developer type people among us, and there may be enough of us that the game would extend quite far without the help of *everyone*.  Do you have the SGDK2 file?  Would you be interested in playing around with the file in the IDE to see how comfortable you are in designing more levels/content based on the existing CleanGame framework?  How hard is it to figure out what's going on in that project without any help?  (And certainly, if you think it's worthwhile, I could provide some instructions to help people understand the project and make their own additions.)

Maybe we can ease into this a little bit at a time and it would be much easier.  Start out by just formalizing the CleanGame framework and accepting and merging submissions manually.  Hopefully some of the first submissions would be better artwork .

[v6v]

I'm behind you %110 percent on artwork. I know you haven't been able to see my work, (seeing as only .5 % of all of my created projects I've submitted [Let's forget about that ArtEasel/PearlShadow deal])

But I'm sure if a reasonable design motif was reached, it could form the stepping stone for this. I'm highly enthusiastic about the 2.5D maps, and I hope that I can create a similar WebGL alternative which performs the same thing- so far, it's really nice to be able to rotate the camera up and down, zoom in and out, and have one layer of the map move toward and away from the background (Creating a better view of what's ahead the player)

If anything, I think most of what I'm making can be useful in your project- (If I'm able to get it working in an HTML page.) That is, if you choose to use it.

The CleanGame seems fairly easy to understand (As opposed to my projects, where rules have completely unrelated names, and declared variables in my code are all over the .cs files)
You seem to develop in a way that anyone with somewhat of a background in SGDK2 can understand (Although I admit, Net.cs took me quite a while)

When I get the time, I'll try and submit artwork- but at the current moment (Having Highschool classes that are driving into my schedule) I'm starting to become tied up.
We could throw ideas on a separate thread. I'll help where I can.

But to start, I've never worked under a project manager- what exactly do you prefer art-wise? I know you're a fan of the Amiga/Commodore 64 games.

[v6v]

Oh, I need to ask- I never asked, the code that displays the tiles, (I've never compiled an HTML project with the newer SGDK2)

is the HTML5 Version a JavaScript file, like Display.js? Or is that automatically imbued within the HTML file?

[bluemonkmn]

is the HTML5 Version a JavaScript file, like Display.js? Or is that automatically imbued within the HTML file?

You can choose to have all the code merged into 1 file when you export.  If you do that, of course all the JavaScript code gets embedded in the HTML file.  You can also choose to have the JS file be separate from the HTML file, in which case all the JavaScript code goes into a single separate JS file.  You can also choose to generate separate files per object, in which case you will get separate JS files per sprite definition, per map, and per custom source code object (if I recall correctly).  However, there is no option that generates a separate JS file for each CS file in the project.  So there is no Display.js.  This is partly because much of the framework code is all merged into a single file, and partly because HTML5 removes the need for much of the framework.  Much less code is required for a display object because it's mostly native to the HTML5 canvas.

When I get the time, I'll try and submit artwork- but at the current moment (Having Highschool classes that are driving into my schedule) I'm starting to become tied up.
We could throw ideas on a separate thread. I'll help where I can.

But to start, I've never worked under a project manager- what exactly do you prefer art-wise? I know you're a fan of the Amiga/Commodore 64 games.

I'm not sure I'm ready to jump into this with both feet yet either.  I've never managed a project.  Although I have had artwork submitted to me (OrionPax did the artwork for the sample game).  I really like his work, but last I heard (probably at least a year ago) he was, how shall I say, "freedom impaired" due to some presumably questionable judgments in the eyes of the law.

The important thing about "CleanGame" is that the current title represents the one imperative I had in designing it.  There should be nothing messy about its appearance/design (and I fear I may already be straying into dangerous territory there with the pushable boxes - they don't work very cleanly).  But I don't was a bunch of graphics that look "close enough" when pieced together.  That's why I stuck with very simple graphics.  It's hard to go wrong when there are no details to screw up .

But I think before putting much effort into graphics I will have to work on the HTML 5 version/support... see if this thing "has legs".

[v6v]

In that case, I'll just wait until you've fully completed the HTML5 implementation to actually try to do something like that.

Good luck with the further HTML5 implementation. I'll be here whenever you need me.

SGDK3... Who could imagine? Have you determined a name for this future project?

Say... SmallLargeUniverse?

[bluemonkmn]

Heh... SmallLarge Universe.  Haven't thought about a name yet... not even sure if it's going to be a single project or a conglomeration of projects.  But there will have to be some sort of hub one way or another.  Hm... the idea is loosely based on the concept of (and trying to harvest the fruits of) the Noosphere.  Perhaps I should use that name or something related to it.  Surprised there aren't more concrete projects taking that name already.  Perhaps... I best not suggest my web site name idea lest someone domain squat on it .