Fan made projects related to MLP

Search /collab/ threads

Password  (for post and file deletion)

File 136059220335.png - (1.22MB , 2683x3810 , drawing.png )
44047 No. 44047
Awesome Pony Fighting Game.
This will be an open source 2d pony fighter game with original characters.


By contributing to this project (Awesome Pony Fighting Game) you agree that your contribution is considered to be released under the Creative Commons Attribution 3.0 Unported (CC BY 3.0) License(SEE

All code will be released under GPL.

Do not use copyrighted characters,locations or anything else that is copyrighted. Use only assets that are under Creative Commons Attribution CC BY.

Step 1

The Art
Here something I commissioned. The SVG files are broken but someone is working on fixing it.
We will create original characters using this.Use Inkscape to make vector art.
After you have created the art in Inkscape port it using the SIF exporter extension for Inkscape

Step 2
How is this game going to made?

Sprite animation

We will be using the Linux version of Synfig a free and open source software.Here is the latest version . There is a windows version available at but its dated.If anyone willing to port the Linux version to windows it would be awesome.Here is a tutorial on how to make sprite animations using Inkscape and Synfig

On a side note it is also important that this person's Indiegogo campaign become successful
He is responsible for creating the Stickman template we will be using for this project. Here is tutorial series for the Stickman

Also here is a tutorial for creating good sprites

The Engine

For the game engine we will be using the the free and open source Fabula Game Engine.
We will need to port the code from an open source fighting game engine under GPL.


Animate a Walk Cycle Using Synfig
Make Pegasus Fly(Annoying.)
1 Magical Attack. Fireballs from Unicorn head.
Idle animation.
Jump Attack.
Animate a trot and run cycle.

What we need

Vector Artists
Thats it.

Current Time-line.

We currently have to wait for the next stable version of Synfig.You can make the process faster by helping them.Go here to help
They need Developers, packagers, bug testers, translators, artists, web developers, wiki writers... you can contribute!

We also have to wait for the SVGS file to be fixed and for a new improved version of the model sheets.
The Stickman improved version with character template is not released so we have to wait for it.Here is how it looks

After the improved Stickman is released we have to make a Stickman for 4 legged creatures.

Things to do right now.

Someone has to port the code from zero2d to the Fabula Game Engine. Any takers.

The ability to dock un-dock Synfig windows is important.We need someone to program it.

A pony creator like thing for Synfig to store all the assets is also important. .We also need someone to program something like that.

Whats the gameplay.

You will chose to be either earth pony, Pegasus or unicorn.
You are on an isometric square(by square I mean the shape square).

There is an NPC you can talk to and an earth pony.

Go to the earth pony talk to her then their will be a scene transition like Pokemon.

The game will now switch to fighting game graphics like street fighter which are Side-scrolling

.You fight her.

If you lose there will be different dialogue then if you win.

Thats the entire game.

I commissioned the model sheets. I now have to wait for the SVG files to be fixed and the commission someone to make a more show accurate version.

Suggestions and Questions are recommended since I am lazy.
Unspoiler all text  • Expand all images  • Reveal spoilers
>> No. 44048
Another "This is what I want, now do it" thread?

Please read the pinned post on top of this board regarding how to make a proper request as right now this "project" is just a long wishlist where you guys ask for everything and want to take control of it once you receive it.
Also, are you the same from this locked thread >>44041 ?
Because recreating this thread after a Mod locked the other is not a very good move.
>> No. 44050

I have usable vector assets for this game which I am getting fixed. I am going to be doing some work in this.

Last edited at Mon, Feb 11th, 2013 09:01

>> No. 44075
I have some good news the person responsible for fixing the vectors will try to have that out this weekend.
>> No. 44112
How do I spread the word about this project?

Last edited at Fri, Feb 15th, 2013 14:30

>> No. 44220
Have to do extremely boring layering in Inkscape
To complete stage 1 of this project

Need motivation to continue. Give me your strenght.
>> No. 44222
I think rather than spreading the news, your difficulty lies in making the news spreadable. Don't get me wrong, I love the idea of working on a pony fighting game, but I had to read through your initial post three times to figure out some of what I think you're trying to achieve here.

For instance, there is an NPC you can talk to AND the earth pony, or the NPC you can talk to IS the earth pony?

Why does the isometric scene exist? It seems a lot of effort to go to, making an entire play mode for what sounds like about three seconds of gameplay.

I won't go into what I think about making a strictly single-player fighting game, but I will comment that it seems like it would be much, much harder to make a worthwile AI opponent than to make a networked game.

And why are you specifying all the tools you want people to use? All it seems like that will achieve is weeding out most of the people who would be able/willing to help you. Why should an artist who knows how to use flash learn how to use inkscape and/or synfig? Why should a programmer use learn to use fabula? It seems like you'd save time and effort; and attract more people if you described the problem and then let professionals and those with experience in their field figure out how to solve it for you.

I hope I haven't come across as rude, old horse. It's just that I need (medically NEED) a pony fighting game, and i'm quite willing to get my hands dirty to see it happen. The title of this thread got me all excited, but this project makes little sense to me...
>> No. 44224
Well this is suppose t be a free and opensource GPL game. Synfig is super important because the next stable version has all the requirements to make sprites.Its lightweight,free and open source and is a vector animation software. See here
Fabula is a python game engine under GPL. Its python so it will be easy to code.
>> No. 44228
File 136187149037.jpg - (96.66KB , 400x302 , that_is_why_you_fail.jpg )
>"This is what I want you to do" project
>Obscure as fuck software, engines and shit

There aren't that many programmers and even less free/open source software programmers in the fandom, and if you find someone, most likely they would be more comfortable programing in something like PyGame (turns out Fabula is Python), Ruby/SDL, Gosu, etc.

Why don't you get started on some of the work intead of just asking other people to do stuff for you?

And why are you being so restrictive about the tools for this project when you aren't doing shit?

It feels like you're recruiting help for Synfig instead of the game.
>> No. 44229
Could someone at least compile the linux version of Fabula game engine as a deb so that I can use it.
>> No. 44230

A point Yoda has.
>> No. 44240
>> No. 44244
File 136195583918.gif - (2.58MB , 427x240 , mlfw9203-Final2.gif )
Buck it. I'm a certified programmer, i've studied game development and i'm currently studying AI. I can make this work.

I will fix this project, or see it burn to ashes around me. :)

So, speaking as your lead developer; some things need to change. You don't make code decisions. I do. You don't tell me what tools to use or what language to write in. You've posed a problem, I will solve it with tools that I am familiar with and that I judge to be suitable for the task. Anyone else who joins this project, from whatever discipline, should do the same.

Secondly, Fabula is out. I'm not wasting a few days learning a new language, only to waste more time modifying an engine that was designed to make a completely different kind of game, especially if it provides zero benefit to the project.

Thirdly, the isometric scene/game mode. I asked you why this existed and you didn't give me an answer. It seems to be a waste of time on my part, on the artists part and, most importantly, on the part of the player. Fortunately, it has nothing to do with the rest of the engine, so I can ignore it and just tack it on later if it does turn out to have a point.

Fourth, the pony creator. I'm still not 100% on this, but I assume this is so the players can create their own characters that are then tacked on to the generic moveset? This is less a programming problem than a gameplay one. Letting players create their own characters is great in an rpg and can be fine in a lot of other games. But it's terrible in a fighter. Making dynamic character generation will not give you an infinite number of characters. It will give you a single character that looks infinitely horrible. Even if I do end up programming it (and it sounds like a bit of a headache, technologically) it won't be until we have a pretty good idea what the engine is doing, so we have plenty of time to gather suggestions, ideas and opinions.

*whew!* /text wall

I'll make another post today or tomorrow with my goals for the engine, maybe a vague timeline and anything else anyone asks about. I'll post again in a couple of days with my actual progress on the engine.
>> No. 44245
Zero2D is a engine designed for fighting games under GPL how does that sound?If you create a custom engine it must be under GPL
>> No. 44246
Which languages can code in?Do you know C++?
>> No. 44250
File 136203296739.png - (28.07KB , 618x445 , Untitled.png )
Roight. Not much to look at, but the basics are done. Player input, simple animations, and some collisions. Fighters can push each other around a little.

I'm currently working out the kinks of a prototype combo system. Might not get that done tonight, I have assignments and junk, but I should have something to show for it sometime tomorrow. By the weekend I hope to start on hitboxes and setting up a second moveset for testing purposes.

Zero2D seems pretty cool, actually. I'll probably lurk their forums a little, hopefully pick up some advice about making a fighter. I kinda just want to get on with it though, so imma go ahead and make my own. :D

The only problem I have with making my code GPL is a slight twinge of emarrassment over how poorly documented my classes tend to be. :P Eh, maybe this'll break some of my bad habits.

Oh, almost forgot! For testing purposes, i'm using sprites from the very talented Urimas, downloaded from:

>> No. 44257
File 136214263198.png - (61.89KB , 618x445 , Untitled.png )
Combo/cancel system is done. Moves are determined by input combinations, and specified moves can cancel into one another.

Added Twi, along with her own little moveset.

Next stop, making a little horse punch another little horse in the face.
Or hitboxes, if you prefer.

I know a bunch of languages. Yeah, I speak C++. I've never used it outside a classroom though. Everyone i've ever worked with has either hated it, or been afraid of it. :)
>> No. 44261
I was wondering if you could help with the coding of synfig its a free and open source vector animation software. The Linux version is awesome. There is a windows version but its dated here is the windows version and here is the state of the art Linux version .To see what its capable of go here .
Currently they are busy creating the next stable version which will be available for Linux,Mac and Windows.
I was wondering if you could help with coding the next stable version.
Here is a tutorial on how to use synfig for 2d games
>> No. 44264
File 136219260926.gif - (243.82KB , 350x417 , mlfw7922-ed5.gif )
Uh. No. I'm not going to help code Synfig, mostly because i'm a bit busy coding a fighting game. It's also out of scope, an artificial dependency and, if i'm honest with myself, the maths behind vector image manipulation are probably beyond me.

Is Synfig really that vital? Is it really the only piece of software that can get the job done? What makes it so much more infinitely valuable than any of the other tools artists might have to hand, open-source or otherwise? What makes it so impossible to carry on without it, that you'd move your only developer off the ACTUAL project?

Compare it to, say; GIMP. An open source, animation-capable piece of software that, and this is important, already exists.
>> No. 44276
Well there are 2 things important about synfig. One create an infinitably scaleble PNG sprite sheet to be used in the game. And the second thing is using the stickman.By saving the stickman rig people would be able to use any 4 legged creature with a side view SVG that walks like a pony in the game kinda like the cookie flex rig of 2d projects
For now you are right we should focus on the open source GPL game engine as the next stable version of Synfig will be released in under 2 months.
>> No. 44301
File 136233854247.jpg - (23.00KB , 678x442 , Sugarr Starr.jpg )

You are using made up charries?

I will apply for character design :)

Heres an example: Sugarr Starr:

A frisky young pegasi trying to be number one. Yet to discover her cutie mark.

Last edited at Sun, Mar 3rd, 2013 12:33

>> No. 44302
The Ponies in don't have cutie marks they have skill marks or as I prefer skill sheet. A skill mark is something the ponies can create themselves. Its on their flank and they can change its design. They shows skills like this see the skills section of this guy. Their skills can be self rated or certified.
>> No. 44312
File 136239419029.jpg - (15.46KB , 320x236 , mlfw4489_small.jpg )

I think he means "yes, we're using made up characters." ^^

I never thought i'd say we need more OC's... XD

My plan is to program in movesets that are... *ahem* "Inspired" by characters from the show. (not our fault if some Dastardly Knave messes with the sprites. ;) ) So, we'll be needing a whole lot of OC's of each of the three main races, at very least.

Last edited at Mon, Mar 4th, 2013 03:51

>> No. 44314
I want characters to have stats like this
Here is how the races are balanced.

Earth Ponies are master of all ground elements.They have strenght.If they use non-elemental attacks aka Hand-to-hand combat there stamina doesn't deplete.

Pegasus are master of all air elements.They have agility..If they use non-elemental attacks aka Hand-to-hand combat there stamina doesn't deplete.

Unicorns can master only 2 elements. But because they have magic there stamina for using the elements is 5 times more than other species.If they use non elemental attacks aka Hand-to-hand combat there stamina depletes.
Unicorns can do status ailment attacks like confusion,sleep etc

Last edited at Mon, Mar 4th, 2013 06:48

>> No. 44360
So how is the team progress going?Currently I have commisoned someone to draw water melon pony and fix the SVG files.
>> No. 44361
The engine now has hitboxes, launches and grabs, i've added health bars (so the 'game' such as it is can actually finish) and i've started on wallbounces and a simple camera with parallax.

Once all that and projectiles are finished it'll be ready to start adding characters for testing. In about a week i'll start to make up and build in characters and get a few testers to run the engine through its paces, see what breaks.

Last edited at Thu, Mar 7th, 2013 05:43

>> No. 44400
Can you put your engine on github
>> No. 44407

If you want to actually compile it though, using this will be a pain. I really don't recommend going through that.

The engine as it stands really isn't ready for scrutiny, but if you want to test it or whatever just let me know and i'll make a proper build for whatever OS you're using.
>> No. 44408
I use ubuntu 12.10 64 bit.

I want it to support all major OS
Windows XP+ (x86)
Mac OS X 10.6+ (x86_64)
Linux (x86, x86_64)
>> No. 44413
Oh dear, looks like this took off quite quickly... I don't suppose there's room for another programmer in here.
>> No. 44418
Absolutely! :D Provided you can put up with my commentless code... I should really go and fix that. <_<

Cool, I haven't had an opportunity to test it on linux yet. :) The graphical backend should work happily on windows, linux and mac. I've also seen projects using it with android, but i'm not entirely sure how that works. A fighting game on a touchscreen would be weird anyway...

I'll have a linux build up and running and soon as I finish messing with the camera. :) May as well toss a windows build up as well...
>> No. 44420
Also can you put a license of GPL Version 3 in your github
It suppose to be called COPYING like this
>> No. 44424

Windows build:
Linux build:

The camera is still a little wonky, but it'll do for now.

Controls are:
Player 1: arrow keys to move, A,S,Z,X for attacks.
Player 2: 2,4,6,8 on the num pad to move, O,P,K,L for attacks. (Don't ask me why, it seemed like a good idea at the time.) >.> Quarter-circle forward or back followed by an attack button will all do different special moves.

Oh, and obligatory "This is a (pre)alpha build, everything is ugly is sin and will be made good later" message. :)

Last edited at Wed, Mar 13th, 2013 01:26

>> No. 44426
File 136317712417.png - (65.54KB , 832x960 , noname.png )

I guess I'll try to catch up! Oh dear.

BTW, the only systems I use are Linux-based, so you can be darn sure I'll make sure it stays compatible.
>> No. 44427
[email protected]:~/Downloads/EXPORT$ ls
Assets Fighting

[email protected]:~/Downloads/EXPORT$ sh Fighting*
Unable to access jarfile /home/john/Downloads/EXPORT/Fighting

There seems to be some problem with the linux version.
>> No. 44429
File 136319890799.png - (9.02KB , 1280x720 , mhapes.png )

Disclaimer: I have no idea what I'm doing but this is as far as I got.

# Extract jar file
tail -n +4 Fighting > fg.jar
# Extract jar file :)
unzip fg.jar
# Assets are included twice it seems.
# Fix capitalization and replace one system incompatibility with another
sed -i.bak svb/ -e 's/assets\\sprites\\/Assets/Sprites//g'
# Replace one system incompatibility with another
sed -i.bak svb/ -e 's/Assets\\Moves\\/Assets/Moves//g'
# Recompile
javac svb/
javac svb/
# Run (at least that's what MANIFEST.MF is saying we should do)
java -Djava.library.path=. svb/SimpleTest

Were those images a lie?

>>44250 >>44257

The included .png don't look anything like that.
>> No. 44430
File 136319930092.txt - (575B , script.txt )

The backslashes should have been doubled here.
>> No. 44431
File 136320191379.png - (153.13KB , 900x861 , noname.png )

Thank you very much! I was trying to figure out why `Fighting" was 4 lines of bash followed by a blob of binary. It's been a while since I've been in Java.

I am crying delicious uncommented tears parsing the source code, although part of it is just probably just because of foreign conventions and design decisions. I hope Jubilee can clarify when he returns.
>> No. 44437
File 136324138273.jpg - (20.23KB , 218x320 , mlfw5009_small.jpg )
Doesn't really surprise me. :( I'm a total linux noob, i'd have been pretty surprised if it had just worked. I just used JarSplice to build the .sh file for me, in exactly the same way it made the .exe. >_>

They aren't a lie, they're just old. I stopped using Urimas' sprites about a week ago. They were useful at the time, but they're a bit too small, and at the frame rate I wanted they moved like a blur. I made a grey square and warped it in flash so I could get a more accurate representation of how our actual sprites will behave once we start using them. It's not pretty i'll admit, but if that's a screenshot, it's exactly what it should look like. Though 49 FPS is a worry. :/

As I said, i'm really useless when it comes to linux.
If i'm reading that correctly, you're replacing the double blackslash file seperators with a single frontslash?
A) Linux is amazing, B) The windows version seems totally comfortable with that. Shouldn't be a problem anymore.

And, apparently JarSplice also includes all the assets for the linux version. *blush* Good to know for the future. :D

I'm sorry! D: I have been adding comments all over the place, and every class should have at least a brief description now. I'm hoping to write a more holistic explanation of how the whole thing fits together by today or tommorow.

EDIT: Actually, scratch that. I have two assignments due tomorrow, and four more and an exam next week. I'll try to get something up on the weekend, but i'm going to be pretty useless for a few days. :(

Last edited at Thu, Mar 14th, 2013 01:40

>> No. 44438
File 136327033768.png - (203.23KB , 1280x953 , noname.png )
Yay~ I'll be waiting! Watching. Staring at your repository. O_O It would be impolite to jump in until you were at a point at which you wished to divide the work. I have a bunch of actual work I could be doing anyway.

( I hit the 100fps constantly, but I have dedicated graphics and a quad-core processor. >_> )
>> No. 44443
File 136328549706.txt - (11.40KB , Manager.txt )
> Thank you very much! I was trying to figure out why `Fighting" was 4 lines of bash followed by a blob of binary. It's been a while since I've been in Java.

You're welcome. Most of my java experience comes from trying to make games work on my machine though (which is why I put up a disclaimer). I'm looking at you Patapony! (If the game wasn't so good I wouldn't have tried so hard until it worked!)

> I just used JarSplice to build the .sh file for me

I think just giving us the .jar file would be better (with some instructions like "Run java -jar jarfile.jar"). I don't remember how to build those though.

> They aren't a lie, they're just old.

Don't get me wrong. I'd love to play a fighting game where everybody's a (literal) hitbox. I mean, why not? You can still add effects and such. If you wanted them to be "pony", you could just pick thematic move sets.

I just wanted to make sure I got thing right (since there were no screenshots of this version).

> Though 49 FPS is a worry. :/

Its claiming 200-300 FPS most of the time. Although I don't really believe that since I don't have any particular hardware that should warrant this. It doesn't feel choppy except at the very start (see comment below).

> If I'm reading that correctly, you're replacing the double blackslash file seperators with a single frontslash?

Eeyup. (I can never remember which slash is back and which one's front or forward.) I've posted a copy of to confirm.

> B) The windows version seems totally comfortable with that. Shouldn't be a problem anymore.

I never would have expected this to work!


Oh and you can host compiled files and archives on github too (either in the same branch or a separate one). Mediafire has ads and it feels unnecessary to use them here.

Here's a comment dump.

Of course I know this is an alpha version and also some of these may be intentional by design.

- Jumping forward while close to the opponent feels strange. Sometimes, your foot hits their face and you are stopped in front of them. But if you back up just a bit, you can jump into them.
- Air moves don't stop once you land.
- Near the very start, pressing UP (jump) makes the right player disappear and drop from the sky. It stops doing that (and jumps normally) after one or two jumps.
- There's no blocking. And the gameplay still feels a bit too button-mashy even with the delays on attacks.
>> No. 44445
>Jumping in
You actually showed up at the perfect time for that. Before I got distracted scrawling over everthing with my comment-crayon, I was trying to decide between about five different things to do next. As soon as I can explain what they actually *are* you can jump right in. :)

>Just the .jar file
I s'pose I could. ^^ I want to make the player have to do as little work as possible to get this working, and for a lot of people the command line is a scary place. An .exe is convenient, but it sounds like a .sh is not. I assume there's a linux equivalent of a .bat file, so there's still just something to double-click?

>200-300 FPS
That definitely shouldn't be out of range of any machine made in the last six years or so. I'd expect a framerate in the 1000's at least if there wasn't a limit of 100 fps enforced in the code. I dunno how you're getting 300. :/ It shouldn't hurt anything, but it's a bit weird. XD

>Hosting on github.
That... Did not occur to me. XD I will do that.
(Now what am I going to do with this dumb mediafire account? :/ )

Yeah, the collision system needs to be burnt to the ground and rebuilt from scratch. ^^ I ripped it out of a different project I was working on from a while ago, and cut out all the bits that didn't do what I wanted. I can fix the 'passing through each other' thing easily enough, but the other stuff still makes it unacceptable for the finished game. We'll experiment with better collision systems somewhere down the line, but they're gonna be derpy for a while. :(

>Air moves don't stop when you land.
That problem should magically disappear in a build or two. ;)

Oooh, well spotted! I did some digging, and i'm pretty sure I know what causes this. Unfortunately, I don't really know how to fix it directly. Fortunately, there'll be a menu and the whole "Ready? FIGHT!" thing seperating program startup and the actual game. Unfortunately, there's another way to make this bug happen. I do have one way to avoid it, but it's a little ham-hoofed. :/ I'll hunt for a more elegant solution.

>No blocking, button-mashy.
The engine does support blocking, (lil' Twilight could block) but I haven't given Harold (what i'm calling the grey square :P) a way to do it yet. It's a five-minute job, hopefully, but i'm lazy and distractible.^^ Once he can block, I think the button-mashiness will go to the opposite extreme. Almost any combo will be instantly interruptible by blocking. It's gonna take a lot of fine-tuning and a couple of other game mechanics to get made, before any of it actually meshes together well.

...Really, ponychan? Really? XD

Last edited at Fri, Mar 15th, 2013 06:49

>> No. 44446
File 136336238559.png - (342.17KB , 3000x3000 , noname.png )
Oh yes, words are always the hardest part, aren't they? ^^; Perhaps utilizing the issue tracker and milestones on Github would help with that?

.sh IS the Unix .bat. .sh is perfectly fine, as far as an end-user is concerned it's just an executable file that will give them a desired result. Jars can be made executable, too, by specifying the main() class. I believe Minecraft does this on Linux-based systems.

I would like to mention one teensy-tiny strange thing I noticed... It strikes me as very very strange that the Entity class extends the Rectangle class. I mean, I understand that an entity is bounded by a rectangle, but it isn't really a rectangle-type object itself. It just doesn't seem to fit the OOP spirit.

Last edited at Fri, Mar 15th, 2013 08:49

>> No. 44447
File 136340013462.jpg - (13.32KB , 320x180 , mlfw1142_small.jpg )
That... Um... That should not exist. XD My bad.
Apparently I only deleted it from my working directory, and not the repository. The git should be fixed now.
>> No. 44456

There is now a page in the gitwiki explaining all the big deal classes and how they interact. Fair warning, some of it may have been written at 3AM, so apologies if it's a bit strange. :P Hopefully it's readable, and worth reading. If not, just let me know and i'll try to clear it up more. I might not be able to get to it for a few days though. :(

I've also added a handful of stuff to the gits issue tracker, so if my explanation does work feel free to dive straight in. ^^
>> No. 44459
[email protected]:~/Downloads/APFG-master$ ant build
Unable to locate tools.jar. Expected to find it in /usr/lib/jvm/java-6-openjdk-amd64/lib/tools.jar
Buildfile: /home/john/Downloads/APFG-master/build.xml



[echo] APFG: /home/john/Downloads/APFG-master/build.xml

/home/john/Downloads/APFG-master/build.xml:33: Unable to find a javac compiler; is not on the classpath.
Perhaps JAVA_HOME does not point to the JDK.
It is currently set to "/usr/lib/jvm/java-6-openjdk-amd64/jre"

Total time: 0 seconds
[email protected]:~/Downloads/APFG-master$

Last edited at Mon, Mar 18th, 2013 08:53

>> No. 44462
File 136363656310.txt - (1.24KB , broad-explanation-summary_md.txt )
I've read the Wiki and here's what I understood from it (remove the .txt extension and run it through markdown). Can you also give us a drawing of the current state and "cancels"? Or just the most important states if there are too many (or have some states grouped together).

> I assume there's a linux equivalent of a .bat file

Yes, that's right. Including a .sh file isn't a bad idea. I suppose this comes back to Tanaki's note.

> `Fighting" was 4 lines of bash followed by a blob of binary.

This is pretty weird (and makes things harder to fix once they break). Including the two files separately would be good, I think. (I know if was mostly JarSplice's doing and not yours.)

Also, I think linux user tend to expect to have to do more manually anyway.

> limit of 100

The new version seem to enforce this limit. But now the game feels much choppier.

> Sky-jump

Looks like this is fixed in the new version! But the game still feels slower at the start than later on (with lots of abrupt pauses).


It looks like this build.xml was auto-generate and um, not that great for compatibility. :|
>> No. 44466
Sorry, it's midnight, i've been mathing all day and i'm obviously having some kind of brain-fart, because I can not make that thing do. I've marked it down as hard as I can, i'm still just getting the 403 at the bottom. :/

>States & cancels
IIRC Harold has 35 states at the moment, and all but two or three are connected in a big ol' cancel tree. Most states just have one or two cancels though, I can draw up an example or two of how they connect tomorrow.

>FPS now 100, but choppier.
That's really weird. I don't remember doing anything to change framerate. :/ How exactly is it choppy? Does it look like the framerate is lower than it's claiming?

*sigh* I thought i'd found the perfect solution. If it's borked the framerate though, I might have to find another one. :(

>Pausing at startup.
It's loading quite a lot of stuff right at the beginning. Unfortunately, this is probably going to get worse before it gets better. There'll be more and more assets to load up as backgrounds and sound effects and so on start getting tested. Once we have the basic framework of the game going (a menu and such), we can hide it behind some eye-catching 'beginning of the game' thing (or a loading bar if we suffer a general faliure of imagination).

*yaaaaawn* Now I can sleep. ^_^
>> No. 44469
File 136371793188.png - (6.83KB , 1282x719 , harold-vs-harold.png )
> still just getting the 403

Don't worry. I get a 403 too. I've been having trouble uploading text files to Ponychan. I've put it here for the moment.

Or maybe I should make a pull request to the Wiki? I don't feel I'm saying anything particularly worthwhile here.

>35 states
So that's what

>FPS now 100

I tried again. I think it just takes more time to speed up correctly now so it does see to run at a good speed eventually. I check the first version to be sure and its still running at 300 FPS.

Now I suspect it is just the same behaviour as the "pauses at startup" problem. Its just that those pauses were much less apparent before. As you've mentioned already, the start up time also increase by comparison to the previous version.

I just tried to not touch anything for a minute at the start of the game. but its still choppy when I start moving (and it gets better as the game progresses).

Maybe its slows down the first time it reaches each state? That would be weird.
>> No. 44484
[email protected]:~/Downloads/APFG-master$ ant build.xml
Unable to locate tools.jar. Expected to find it in /usr/lib/jvm/java-6-openjdk-amd64/lib/tools.jar
Buildfile: /home/john/Downloads/APFG-master/build.xml
Target "build.xml" does not exist in the project "APFG".

Total time: 0 seconds
[email protected]:~/Downloads/APFG-master$
>> No. 44507

That's all pretty much right.

>String can be altered at runtime.
That is one reason I want state stuff stored externally. Balance tweaking would be much faster that way. ^^ There are one or two other reasons I want it that way though. Copyright is the big one. This game can never ship with any licenced intellectual property. I won't speak for the users at home, but I want to play this game with characters from a certain animated series in it. To make that happen, I will need to replace art and sound assets, and possible change the behaviour of the related character. Now, for me this is no problem. I can just paste the changed file(s) and recompile. For someone with an .exe file who doesn't know or care what 'recompile' means it's just a niusance. Hopefully this won't be necessary, but you never know. ^^

Another reason I did it this way is to make the character development process easier in the long run. It probably actually makes it a bit harder right now, having to learn a new set of rules and so on, but i'd like to make a set of tools at some point to create and modify files like this. Something that lets developers browse and search the possibly thousands of states and maybe visualize state trees and stuff like that.

>Slows down first time it reaches a state?
That would be really, really weird. When I run it, I have the pause at the beginning and then everything's fine. I need to test it on more machines, see if a pattern emerges. :/

Not really sure how to deal with this. Are you sure you have the JDK, and not just the JRE?
>> No. 44508
I did sudo apt-get remove openjdk-6-*
And installed openjdk-7-jdk using Synaptic Package manager.
Can you automate the process using the .Also does have Ant.

[email protected]:~/Downloads/APFG-master$ ant build.xml
Buildfile: /home/john/Downloads/APFG-master/build.xml

Target "build.xml" does not exist in the project "APFG".

Total time: 0 seconds

There seems to be improvement. Here is an ANT Tutorial
>> No. 44521
What are the requirements to run this game?
>> No. 44530
File 136453802797.png - (218.94KB , 1024x754 , fate_seeker_and_destiny_gazer_by_staticwave12-d5z8tdg.png )
Presenting APFG newest characters Fate Seeker and Destiny Gazer
Fate Seeker and Destiny Gazer by Ak74smith
Drawing done by Staticwave12

FateSeeker is generally shy and keeps things to himself, but he's easy to get along with and is open with friends close to him. At times, he gets smug and cocky once his ego inflates. He has the ability to distinguish anypony's soul of either being good or evil. Detecting a good soul will cause his angel wing to light up, and evil souls would cause the demon wing to be enshrouded in a dark-purple aura. His horn can also cast spells, pick up items with telekinesis, and can even create anti-matter platforms to step on for levitation/flying purposes.

DestinyGazer is the Rule 63'd version of FateSeeker

Last edited at Thu, Mar 28th, 2013 23:38

>> No. 44535
Erm... Right now? Hard to say, exactly. I haven't done much real testing in that regard. It's not really doing much work, so the requirements should be fairly minimal. I'm hoping to keep the minimum requirements as low as possible, but to be honest there's not much I could do to raise them. The most expensive tasks the engine will be doing will all be graphical and there's only so much stuff you can put into a sprite game. If we ever add shader support (and I would like to) that might kick it up a few notches, but that will be optional stuff and won't affect the minimum requirements.

Basically, if you have a graphics card you're probably fine. If you ask it nicely the engine will actually run without one, but I don't recommend it. :P

Cool. So that's one weather pegasus and one magicky unicorn. If we can get an earth pony who has cause to lug an artillery piece around we'll be halfway there. ^^
>> No. 44536
[email protected]:~/Downloads/APFG-master$ ant build.xml
Buildfile: /home/john/Downloads/APFG-master/build.xml

Target "build.xml" does not exist in the project "APFG".

Total time: 0 seconds
[email protected]:~/Downloads/APFG-master$

I think the solution exists in this article
>> No. 44537
I was wondering if we should join this?
>> No. 44538
File 136478007593.png - (265.49KB , 845x713 , crossfire_by_x_tatik-d4agpld.png )
Introducing CrossFire
She is shy'ish towards new ponies.
Enjoys entertainment / entertaining
Loves her close friends.
CrossFire by CrossFirePony under creative commons 3.0 Attribution
Drawing by Xmetalmilkshake

Last edited at Sun, Mar 31st, 2013 18:37

>> No. 44539
[email protected]:~/Downloads/APFG-master$ ant build.xml
Buildfile: /home/john/Downloads/APFG-master/build.xml

Target "build.xml" does not exist in the project "Compile and run".

Total time: 0 seconds
>> No. 44548
I have no problem with it.


As riveting as everyone on /collab/ no doubt finds these error-dumps, they probably don't belong in the thread. I've made a bug-report on the git, could you please go here:
and give me some more details about the bug?

Last edited at Tue, Apr 2nd, 2013 06:44

>> No. 44555

You're running `ant build.xml`, shouldn't it be `ant compile` or `ant run` ?
>> No. 44576
File 136516808895.png - (51.43KB , 314x320 , mlfw2788_small.png )
By Celestia's beard and booties, is that really it? Did I really just waste a day trying to find some way the ant script would behave differently in linux?

Anon is correct, John. 'ant build.xml' gives me the same error as you. Run 'ant compile' then 'ant run'. Or you can just run 'ant' if you're lazy, and don't mind waiting for three seconds while it compiles every time.

Thanks anon! Now I can get back to work. :)
>> No. 44578
I tried that have you read this tutorial

Last edited at Fri, Apr 5th, 2013 09:16

>> No. 44612
File 136566921132.gif - (2.65MB , 500x500 , shaderDemo7.gif )
Well, now that the build errors seem to be sorted out, I thought i'd share the first test of the lighting system i'm working on. :3

This images on the left and right are exactly the same, but the left-hand ones have some normal mapping applied to give them a sense of shape. There's still a way to go with it, but i'm pretty pleased with this so far.

What will we do with it? I don't really know. XD It's safe to say we could make some pretty snazzy environmental effects with this, & at the very least we can make particle lighting for projectiles and so on.
>> No. 44628
File 136587631173.txt - (363B , example_json.txt )
That looks pretty cool. I didn't know Harold was a 3D character!

I hope you'll consider an option to easily turn effects off though if framerates become a problem.

This version works fine for me. But the game crashes now. If two different players are selected (I love how they are all called some variant of Harold) then the OS outright kills the process after (112) is loaded for the second player. Presumably this is a memory allocation problem. How much memory does it need? The entire asset directory is only 25MB.

If I pick the same character for both then the keyboard doesn't respond once inside the fight.

This reply is way late but
>Runtime movesets

Ah yes, I keep forgetting that recompiling isn't so easy for everyone. I think its a good design decision then.

In that case, though, how about using some existing standard (instead of worrying about parsing a new mini-language unless that's part of the exercise). Maybe JSON ( It is fairly simple and has parsers and writer available in many languages?

I've put an example (uploading text should work now). Now that I've made this, it seems strange that conditions are on the target states instead of the source state. (What happens if two conditions are satisfied simultaneously?) Looping also seems like its just shorthand for a condition.
>> No. 44633
>An option to turn effects off.
Absolutely. They'll be off by default, and once the options menu exists, that'll be one of the first things in there.

>Game crashes now.
Gotta admit, this kinda has me stumped. :/ During the load the game should log any errors and then try to keep going anyway ( which incidentally caused that 'ignore keyboard' error).
I tried taking memory out of the test machine down to 512mb, & it compiled happily (if slowly). It did fail when I went down to 256mb, but that was less than the minimum requirements for the OS, so I was sorta impressed it compiled at all. :)
An earlier build went overboard on the graphics card memory, but it gave a tidy little error message about it... *shrugs* I'll keep digging, I guess.

>Keyboard doesn't respond.
Fixed! :) Once again, me and linux are having difficulties with each other.
Since the errors ocurred at the same place, i'm basically hoping this makes the crash magically go away. :D

>Using an existing standard.
Yes. This must happen. IIRC, I started making my own mini-language because it was late and I couldn't be bothered to learn anything right then. XD A few weeks down the road, the plan is to make a seperate character-making program that will write all this stuff out for me, and I was going to make the switch to .XML or something then, but .json looks pretty swish, so I will definitely investigate that.

>Conditions on target state.
I can see what you mean about it being strange. Sort of looking at what's happened instead of what will happen. With the top-down approach, you can look at a state and see everything that it can do, and changing the end behaviour for a state would be a lot easier.
On the other hand, in some states you'd get twenty to thirty conditions in a list, while most others would have none, and if you wanted to change the entry behaviour for a state you might have to track down a bunch of different references...

Looking at it now, i'd say you're right and the top-down approach is a little better. :) Since it doesn't carry any huge benefit though, i'm probably going to leave it as-is. The character-maker will hide it all anyway, and hopefully have the best of both worlds.

>Looping - shorthand for a condition.
Yes and no. Looping is shorthand for ALL conditions. If any of them are true, the state resets. You lose a bit of control with it, but it's easy. :3 It'll either be used on a case-by-case basis, or gotten rid of entirely in the final product.
>> No. 44662
File 136621931651.png - (154.70KB , 1280x764 , shader.png )
>taking memory out of the test machine

That's some seriously thourough testing! And to clarify, it does compile fine. The problem happens when running.

But I noticed that my description isn't accurate either. The OS simply kills the process at that point (during the "Loading file assets/chars/..." phase). I think that's usually what it does when it runs out of memory (pick a memory hungry process and force kill it) but I don't know the details of the murder. The game seems to have asked for at least 2G of ram before getting killed.

I even tried adding a -Xmx512M flag but it did nothing (a lot of memory is still used up and the process still gets killed). But then again, wishful debugging has never worked before. This is where I wish ant showed the commands it ran by default (now I know -verbose gives this effect.

>Keyboard response

That's indeed fixed (I can play just fine when picking the same character for both players)! The start of game delay from way back is also gone now (and so is the sky jump).

And looking at the commit logs, I did notice the file load order changed (now in increasing order of value).

I'm getting the "flat" shading" (right copy from >>44612) for both Anonyrixies (and Anony Pie as well) in the shader test. I don't know if this is expected.

>Using an existing standard.
That makes sense. Its better to get the thing working first I think, which is what you did. Being able to use an external program to view/edit the moveset is certainly part of the reason I asked. XML would work nicely too (although its slightly harder to read as plain text).

You could also have conditions completely separate from states although I don't think I'd advise doing that.

from state = "Idle"
to state = "WalkF"
condition = "button held"
button = 6

It seems harder to track things down this way (although minor alterations (for testing) are a bit easier to make).

And one there is an external viewer/editor, this may not matter so much anyway. (I certainly wouldn't bother changing this unless needed.)
>> No. 44690
>Taking memory out of the test machine
It's a virtual machine. I just click buttons. :)

Oops, I said compile when I meant to say load. :( Silly me.

I looked a little deeper into this memory thing, and it seems you're right, it is hogging an extraordinary amount of memory. The main offender seems to be the spritesheets. It never occured to me that they'd take up more space in memory than they would on the hard disk. A bit of guesswork puts the combined weight of two characters worth of sprites at about 800mb, minimum.
So there's that. /)_-
The rest I assume is coming from java's relaxed approach to garbage collection, and my relaxed approach to memory management. :P
I've been working today on making the sprites more efficient, and should be able to cut their memory usage by 50% at least, but even that may not be enough in the long-run. :/ If that doesn't work i'm just gonna have to redo the whole display system.

>"flat" shading.
There should still be shading on the left-hand images and Pin- er, the pink pony. :) The lighting map from >>44612 was handmade, and much more obvious than the current one. The shading on Trixie was so dark and dramatic because i'd left out 30% of it, and done another 30% wrong, so it was basically ignoring a lot of incoming light. XD
This one was made by a program, it's a lot more subtle and only affects the edges of shapes, but it is there. I should change the square back to the way it was, so it's totally unmistakeable. ^^

>Seperating conditions.
Hmm. Since conditions are sort of dealt with seperately during load-time anyway, this kinda makes a lot of sense. I'm going to have to totally recode that whole section anyway, so making changes is not really an issue.

Well, the character creator is only about 1/4 done, and it's going to stay there until i've played with memory-management a little. Plenty of time to think it through. :)
>> No. 44750
I was wondering if you could open up a tumblr with a donate button.
>> No. 44782
Um... I haven't the faintest idea how to do that. :) Would it be tied to someone's paypal account or something? It's a bit moot for now, since no one is going to donate to the "Grey Squares Touch Other Grey Squares with White Squares" game. XD

Isn't it also kind of dangerous? Sure this project is legal, but it walks a fine line. If money starts changing hands, i'm going to have to be a lot less cavalier about what I use as placeholder art. :P

*Should* we take donations? What would people be donating to? I can't explain *why* exactly, other than the legal stuff, but the idea makes me kind of uncomfortable.

What would we do with it? Both the tumblr and whatever money it collected?
>> No. 44786
The tumblr would be used to keep people updated on our progress and the donations will be used to improve the game.
We could use the money for creating all the game art, the story and hire more programmers.
We should also have a twitter feed on the tumblr.

Last edited at Sun, Apr 28th, 2013 16:23

>> No. 44809
Welp. I'm sick and I have exams coming out of my ears. :P Don't expect anything from me for a couple of days at least.

I'm kind of surprised you were asking me to do it at all. :/ You're at least as qualified as me, and you actually know what you want. Don't let me stop you, dude. :)

I'm not much of an artist, much less a graphic designer, so anything I make will look gross, and asking me to manage the accounts seems silly considering I have no idea what will be done with it.

Speaking of which, if you're going to ask for money, I think you should be really clear about what's happening with it. If someone's giving you their money, they at least deserve to know how it's going to be used, IMO.

For instance, what if you get enough money to pay one artist, and then the donations run out? Will other artists be expected to contribute for free? Will it be divided between them? How are going to choose artists, musicians and so on? (since if you have money to play with, the final product will be much more dependant on your choices, rather than just whoever shows up.)
>> No. 44843
I am asking you because A you are programming the game engine so you will be the first to know all the latest updates and 2 I don't do paypal. Here is an example of a programmers tumblr
I use the deviantart commissions search feature to find artists.
>> No. 45054
This person has joined our project and he is a programmer can you assign him a job
>> No. 45060
Is this still being worked on? There hasn't been a real update in a while, at least on here...
>> No. 45061
Yes it still is being worked on.
>> No. 45063
File 137035189527.png - (307.95KB , 800x847 , watermelon_pony_by_spyrothefox-d67rfl1.png )
Introducing a new character WaterMelon Pony
Water Melon Pony Created By Glaceon
Concept Art Commissioned By SpyroTheFox
>> No. 45064
File 137035784922.png - (138.04KB , 1266x829 , charScreen.png )
Yeah, we're still plodding along. There just hasn't been much to talk about lately. I was pretty sick for a few weeks, & since I got better i've mainly been making the engine do what it already does, but better. Anything I could have said would basically have amounted to 'Check it out everyone! The engine looks exactly the same as it did last week, but it's totally not, trust me!' :D

I suppose the least boring things that have happened would be; the character creator is done, mostly. (pic relevant) It's ugly and there's a lot i'd like to add, but it saves a lot of time.
The algorithm that determines collisions is MUCH more efficient now. During tests it had a decent framerate with up to 500 objects colliding simultaneously. I really, really hope we never need that many. :P
Erm... This isn't totally implemented yet, but projectile behaviour is different now, I guess. Before they would each have been programmed seperately, but now they can be made with the character creator, & can do anything a character can do. So 'projectiles' can potentially have extremely complex behaviours. (For example: minions. Changelings! Shadowbolts! Expendable diamond dogs! Wheeeeeeeee! :) )

Oh, cool!
Ermmmmmm... I guess i'll clean up the issues section of the git and put up all the jobs to be done there. I'll try to get that done tomorrow sometime.
>> No. 45066
So... is the game in it's final form going to be
Main Menu
>Single Player
>Character Select
>Character Select
>Level Select

With a second program being the Character Creator? I'm all for following a 2-D fighter where I can create my own characters, but it's hard to get behind something without understanding what it will be.

Will you be able to create more then just ponies, for instance could you create a 3 legged tripod walking machine?
>> No. 45067
>it's hard to get behind something without understanding what it will be.

My intent is to create a fun, balanced fighting game that is accessible to new players but has enough depth to still be interesting to those with experience. That also happens to be pony-flavoured.
That's what I want out of this, and that's what I hope to make.

I think I may have overstated the importance and power of the character creator a bit. It's more of a developer tool than a feature of the game. I mean, in theory it would let you make a character out of anything you want, be it an abstract grey rectangle, an aeroplane made of bees, or an origami pig. (CARDBOAR!!!11!) The problem is, you'd need to supply the sprites. The creator just lines them up and lets you tell the actual game what to do with them. So, if you wanted a three legged tripod walking machine character that would be fine, but you'd have to draw and animate it first. :/
Even if you have all the animations, making a complete character may take days, or even weeks. Testing to make sure it's actually balanced will take even longer.

On that subject, once players start changing the stats of characters, balance goes right out the window. I don't mind releasing the creator seperately, i'm sure if people are actually interested in this later then some awesome stuff will be done with it, but I intend to set up the multiplayer stuff so that both players have to have the same version to play together. Or maybe there'll be a 'no-holds-barred' version that will just accept anything. I dunno.

I made it mostly so I don't have to write stuff like this:
... {"name":"WalkF","preserveInput":true,"canTurn":true,"openState":false,"allowGravity":false,"block":null,"transition":"WalkF", ... {"name":"Frame: 30","url":"sprites/Walk F/Harold-10096.png","offsetX":206,"offsetY":233,"guardHigh":false ...
by hand anymore. It gets old pretty fast, especially when you have >1000 lines of it. :)

>Character Select

I'm not sure what you mean by 'interact' here.
>> No. 45084

Sorry I was pretty tired at the time and didn't really explain, what I mean is interact with the NPC's or progress the story line, something along those lines.

What I think happened is I got too excited and didn't think logically about it, I've always wanted to create a character for a fighting game. Don't know shit about vectoring, but I'll think about giving it a shot if it means I can make the character I've had in my head for the past decade.
>> No. 45101
Cool! This looks pretty nice. Glad to see this is still moving along!

I can only walk and hit with "A" in the latest build though (no jumping, no knockback from a hit and the field looks smaller). There's also no main menu anymore.

I think its also choking on the

{"name":"Action: 1","type":"DAMAGE","parameters":null}

in Air A.json for anyone that's not Harold or Final-Harold. Probably on the line

String pObject = statusObject.get("parameters").toString();

of the function populateTargetActions in src/svb/

I'm also getting Unknown Source everywhere instead of a line number on an error (such as

[java] java.lang.NullPointerException
[java] at svb.StateFactory.populateTargetActions(Unknown Source)
[java] at svb.StateFactory.buildState(Unknown Source)
[java] at svb.StateFactory.populateJsonMoveList(Unknown Source)

) which is odd and a bit annoying for figuring out problems.

Oh and a suggestion. Even though the moves aren't meant to be edited by hand, maybe you could still have them pretty printed. (For me, that's just so I can quickly find the lines with keywords like DAMAGE and have an idea of about where they are in the tree.)
>> No. 45109
>...what I mean is interact with the NPC's or progress the story line...

Yeah, i'd like to have a story mode or something. I have no concrete plans about it and i'm not going to really think about it until the main engine and a few characters are working. It'll probably need to be someone else's job anyway. A writer I ain't. :P

>No moves, no menu.
The moveset is being made from scratch, so it has basically nothing in that build. As I re-add mechanics i'll be adding moves again to test them. The next few builds will probably add a dozen or so states each.
The menu is still there, I just got sick of clicking through it every time I compiled. :P I'll put everything back the way it was as soon as the menu actually does anything. :D

>Everyone but Harold and Final Harold cause hideous explosions.
*facepalms* Silly, silly Jubilee.
Whenever i've changed the movelist i've only been copying it to the default characters to save time. Evidently I forgot to copy it to the others. 6_6

>Unknown Source
Not sure where this came from. :/ It's not happening in the IDE. My guess would be something to do with the ant .xml file, but I don't remember changing it.

>pretty printed moves
Sure, I can do that. In the next build each command will be on a seperate line, although some of the really huge ones will still be textwalls. I'll try to make the move files a little cleaner every build.
>> No. 45118

>copying movelist
Ah that would explain it. It happens.

>Unknown Source
I would guess ant too (or some kind of compilation flag that got changed) but I know even less about ant than you.

>No moves, no menu.

Oh, I thought the movelist from before could be directly translated. But I guess the engine changed (or is changing) at the same time.

Although for me, that was mostly to "see" what the states and conditions were (without learning the file format).

>pretty printing

Actually, I thought a pretty printing function is included in the JSON library. No need to bother if you have to manually add something to do that. (I mean, I can certainly parse and pretty print (arbitrary) JSON on my side with no problem and presumably, not a whole lot of people read it yet. Some automatic pretty printing can be added later on if needed.)

Come to think of it, it feels a bit strange that so much is being populated and built as json parsing. Is the parsed format not suitable for direct use?

Is one of them (either internal storage format or external format stored as text files) better than the other? (In which case maybe the worse one should be changed to the other, better one.) Is Java unable to "generically" create instances from field name and values? (I don't know much about Java but I don't see why you shouldn't be able to create them from arbitrary string inputs.)

Or maybe you are doing this for validation? (But it looks like a lot of work just for that, although it certainly worked to catch invalid input early on.)

When I suggested using an existing format, I was thinking that the change would let you remove whatever parsing code you had (and replace it by something short that would not change as the specification for the external .json files changed (the engine would change, but the parser wouldn't)).

I mean everything seems to work now so I wouldn't normally suggest something that could break everything. But since you are altering the engine, this is as good an opportunity as any.
>> No. 45119
By the way how is linux version coming along? Aslo have you assigned the new kid some work?
>> No. 45167
Just something I felt like sharing
>> No. 45171
>Pretty printing function.
Hmm. The way the files are being written at the moment is object-by-object, maybe that's why I didn't notice it. It should be simple enough to change that.

>Is the parsed format not suitable for direct use?
In some cases, no. It's totally possible i'm doing too much with it, since I had everything else working that way already, but some things are definitely inconvenient in pure json. A bunch of things are just using the json values though.

>...I was thinking that the change would let you remove whatever parsing code you had and replace it with something that would not change...
Well, it *shouldn't* change. Much. :) A few tweaks here and there maybe. It's basically just going:
character.value = jsonObject.value.parseInt
a few bajillion times, so even if everything goes fruit-shaped and I have to change it around it won't take long.

I am reading this at 4am, so i'm sorry if i'm misinterpreting or missing the obvious. :(

>By the way how is linux version coming along?
...They're exactly the same. I test every build in linux now. Is it not working?

>Aslo have you assigned the new kid some work?
Uh no, sorry. I'm in the middle of finals and i've been making a concerted effort not to do any work on this until they're over on friday. :/ And honestly i'll probably spend saturday eating ice cream and weeping. :D So, i'll get back to it on sunday.

I also have no real way of contacting the guy. I can put stuff here or on the git, or I can send a PM to the indiedb account he has apparently never accessed. ("time online: 1 second") I have no way of knowing when or if he will ever recieve any of those.
>> No. 45199
I sent you a new message about it in the issues section
>> No. 45292
How is the progress going?
>> No. 45314
The update from a few days ago fixed a bunch of bugs, re-added an improved blocking system and added the projectyile system.

The update from a couple of minutes ago added functioning hitstun, a combo tracker & wallbouncing. I'm going to spend the next few days tweaking and fixing those. (the combo tracker has a bug, wallbouncing feels really yucky & I want to change how hitstun works a bit)

Once that's done, i'm going to rip the sprites from another fighting game and try to replicate some of a characters moveset. (I won't be able to upload that version, unfortunately.) I'm expecting that to bring up about a dozen things i've forgotten to do, or have done poorly.

So, yeah. It's gettin' along. :)
>> No. 45318
We might find some legal to use sprite sheets from opengameart like this
Here is a list of sites with open source content
>> No. 45377
File 137343755577.gif - (1.75MB , 1850x906 , rvgc-final.gif )
I tried the latest version and its working without issue. Harold is still missing some moves as expected (no slide and uppercut, no air attacks).

> Pretty printing function.
I just noticed that git diff is not as meaningful on these JSON files (it just tells you the entire file/line changed) so maybe this is actually worth bothering with. (Depends on how you use git, of course.)

Apparently, if you use org.json.JSONObject instead of org.json.simple.JSONObject, you can just call toString with an indent parameter, like

I hope this doesn't needlessly add bloat the project though!

> Using json values directly
Oh, well. That's too bad. That is exactly what I meant.
> jsonObject.value.parseInt
I'm a bit surprised that it doesn't give you a Java int directly. The might even be the real issue here (the parser isn't giving you something immediately usable).

> if everything goes fruit-shaped
Now there's an idea for a fighter!


While you are here, you should know that the cutie mark in these are embedded raster images, not vectors.
>> No. 45384
I was wondering if you could create a PPA for Awesome Pony Fighting Game APFG kinda like this
>> No. 45416
The point of using existing sprites is less to have more sprites and more to force me to follow a pre-existing design. I already know what the charcacter is supposed to do and how they're supposed to feel, so I don't have to spend time designing their moves.

Rarity OP. Buff Giant enemy Crab pls. :P

>Pretty printing.
I'm doing something wrong somewhere. I can make pretty printing work during runtime, but when it writes to a file it gets kinda scrambled. I'll figure it out at some point, and the code to prettify them won't be part of the engine, so no bloat. :)

json.simple seems to want everything to be in Longs. :/ Kind of annoying, but not really a huge problem.

That's a way of distributing to ubuntu specifically, right? That's something that will probably happen eventually, but it's more of a long term goal. I'd like work on the engine to calm down a bit before focusing on specific distributions.
>> No. 45506
How is the progress this month?
>> No. 45524
File 137572278887.png - (399.86KB , 1215x1709 , devkitsthroughtheages.png )
My laptop exploded. -_- The replacement arrived two days ago, and i've only just gotten everything transferred over and re-installed.

So, i've mostly been writing stuff down and working on the dev kit. (the only thing that wouldn't make my laptop scream in pain and then paint everything a random colour)
I've added a fair bit and generally made it nicer. (pic relevant)
There's still a lot of features i'd like to put in and i'm still tweaking it pretty much every time I use it, but for now... Well, it no longer disgusts me. :P

As for the game itself, 'bouncing is working pretty nicely, i've put in a few combos which all seem to work as intended and after noticing that aerial combat felt really slippery I started writing in some hitpause too. I need to change the way a few mechanics work a little before mid-air fighting can really work so i'll be doing that for the next few days.
Oh, and I also maybe wrapped my tendrils around another programmer, possibly an artist as well.

On that subject, what is the plan re: the art side of this?
>> No. 45529
I will find some artist to draw this month. When will you update the repo on github though?
>> No. 45660
How is the progress going? The repo hasn't been updated for 2 months
>> No. 45674
Yeah, we've been using a lot of sprites that are part of one IP or another. :P To upload the current build, we'd have to delete all of the characters, so it would be unplayable, even if it ran.

Since we still need those characters for a couple of things, we won't be uploading a new build for a little bit. We're working on making them obsolete, and we aren't adding any new stuff that's copyrighted, but for now we're testing a lot of stuff that'd just take too much time to implement again.

Azura (the new guy) is working on a guage/meter system that will also take the place of the health bars, round victory counters and similar. Since that's most of the HUD, he's going to be making that too, along with everything else that will be drawn straight on to the 'camera.'

Once the meter system is in place, there are only two mechanics left to make before the generic behaviour of every character can be implemented. ( I hope xD) With that in mind, i've started making Harold 2.0, who will have examples of all the mechanics we've made so far, rather than just whatever I happened to be working on at the time like with the old one. I'm also drawing some really rough sketches for his sprites, so he'll be slightly more interesting to look at than a deformable square, and have animations that are tailored to what he's actually doing. I'm also finishing off the sub-actor mechanics; the stuff we need for pet characters, swap-out characters and better control over projectiles than we have now.

Last edited at Sat, Aug 31st, 2013 10:15

>> No. 45857
How is the progress going? I am still in talks with an artist to draw for this project. Hopefully he will agree.
>> No. 45873
Youtube embed play button
Eheh, sorry for being away so long. I assumed this would be completely forgotten about after a few months with no activity. :P

It's all progressing fairly well, all things considered. There are a lot of little things that need cleaning up, and there are obviously still a lot of major features that need to be added before it's a fully-fledged engine, but the 'fighting' part of 'fighting game' is basically done now, and it doesn't look totally disgusting. :)

Fair warning, the 'art' in the video is all mine, and each animation has a frame rate about one twentieth what it could actually be. With a real artist at the wheel, and enough frames to fill out the animations, this could obviously look much better. :P

Last edited at Thu, Oct 31st, 2013 04:49

>> No. 45889
File 138367612482.jpg - (2.80KB , 100x100 , 51IAx+tovtL__SL500_SS100_.jpg )
neden geç açıyorsunuz?
>> No. 45899
Youtube embed play button
  pikie pie vs apple jake
>> No. 45939
to jest super gra
>> No. 45952
File 138618253606.jpg - (10.94KB , 225x225 , fighting.jpg )

too close to a another pony fighting Game
>> No. 45959
File 138641976969.jpg - (8.45KB , 264x191 , spoiler.jpg )
a game cool
>> No. 46036
>nuked by C&D
but i'd love to see clone with at least half of mane6's quality
>> No. 46455
File 139409601878.png - (1.16MB , 4049x2912 , celery_pony_reference_sheet__commission__by_facelessjr-d76y0d2.png )
Things are looking up
>> No. 46456
Done by the talented faceless JR
>> No. 46751
File 139897319597.gif - (440.20KB , 640x600 , zombie-vertical-preview.gif )
Did this stop? I hope not.

This looks an awful lot like what you were doing with lighting sprites.

Is this a common technique? I've never seen it before here.
[Return] [Entire Thread] [Last 50 posts] [First 100 posts]

Delete post []
Report post