Thanks for the offer, man, but at the moment we have closed the doors to new contributors because introducing new people might slow things back down. Later on, should we need more help, it would be awesome.
-
Hey Guest. Check out your NeoGAF Wrapped 2025 results here!
Indy Game Development: any GAF'er ever make their own game, or even make money on it?
- Thread starter Mengy
- Start date
There are lots of games i would get if i could pay less than they want.... a good example is Atom Zombie Smasher, since i played it recently because of the steam coal thing
They ask 10 dollars for the game ... i would NEVER pay 10 dollars for this game because it is totaly not the kind game i like ... but included in the random humble bundle i got it, it was totaly worth the money i spend on it.
So, for games that aren't in bundles, piracy includes people who would buy and people who would only buy at an insanely ridiculously low price
And that is only one example
ThatObviousUser
ὠαἴÏÏιÏÏÎ¿Ï Ïαá¿Ï εἶ
Bottom-tier writing, makes it really hard to read. :-/
Hattifattener
Banned
90's Dos games are goldmine of indie game ideas. Trials HD was Elasto mania rip-off example.
Someone should use RedLynx receptie whit other Dos games.
Liero is realtime worms(1).
Wings is realtime worms but worms are spaceships.
Mine Bombers is Bomberman boulderdash hyprid.
The Ultimate Tapan Kaikki is realtime xcom.
Someone should use RedLynx receptie whit other Dos games.
Liero is realtime worms(1).
Wings is realtime worms but worms are spaceships.
Mine Bombers is Bomberman boulderdash hyprid.
The Ultimate Tapan Kaikki is realtime xcom.
HP_Wuvcraft
Banned
.... I love this beautiful thing.
Jeez, Gamemaker is making me crazy. Can't have perfect collision. I want PROFESSIONAL feeling collisions god damnit.
It's theorically possible to make though. The "Hello Engine 5" guy totally did it --- and in a freaking similar way to me, I can't hardly tell why my stuff isn't solid like his.
((
It's theorically possible to make though. The "Hello Engine 5" guy totally did it --- and in a freaking similar way to me, I can't hardly tell why my stuff isn't solid like his.
((ThatObviousUser
ὠαἴÏÏιÏÏÎ¿Ï Ïαá¿Ï εἶ
Yeah, GM and even C2 don't do collisions 100% perfect, which can be annoying.
Jeez, Gamemaker is making me crazy. Can't have perfect collision. I want PROFESSIONAL feeling collisions god damnit.
It's theorically possible to make though. The "Hello Engine 5" guy totally did it --- and in a freaking similar way to me, I can't hardly tell why my stuff isn't solid like his.
Are you implementing your own collision logic or using the one Gamemaker provides?
I always go in GML and never uses icons. I made some "collision rectangles" system at 30 fps. Results were not all that awesome even when factoring my movement speed and multiple other things. What I am trying right now is to make the game 60 fps. So far it's working ALOT better. My bottom detection now doesn't even need a "move contact solid" to land correctly. I am very positive right now.
EDIT: ... hmmm, not perfect with diagonals. I still have this wierd bouncing going on. I'm only able to remove that by using "move contact solid"
Diablohead
Member
Now that christmas is over and done with i've returned to my iOS game, after about 4 weeks of not working on it i've come back to what looks like a huge mess because I was cramming too much in there like level ups, extra weapons, complex hud and map system, way too much if I want to release the game in a few months at the latest!
So i've taken a new approach, the HUD is gone apart from a floating health value above your head (visible when standing or being hit) and added in a very fine but curved border to give the feeling you are playing on an old sdtv, along with this starting the app copies what happens when you turn an old 8-16bit console on with the image shaking briefly, a very quick "all rights reserved" message like on the megadrive/genesis and of course very faint but cool scanline effect.
I wan to speed up my work so I can release this game so i'll show off some newer screens at the end of the week.
So i've taken a new approach, the HUD is gone apart from a floating health value above your head (visible when standing or being hit) and added in a very fine but curved border to give the feeling you are playing on an old sdtv, along with this starting the app copies what happens when you turn an old 8-16bit console on with the image shaking briefly, a very quick "all rights reserved" message like on the megadrive/genesis and of course very faint but cool scanline effect.
I wan to speed up my work so I can release this game so i'll show off some newer screens at the end of the week.
Understandable. Good luck on the game, looking forward to it!
I've kind of only recent started making games - at that, I've only recently started programming. It's hard work, and I'm amazed at anyone who can actually get a game running.
So far I've really just been doing Actionscript 3 and a bit of Java, which are both really easy languages to get into. So far I've made two little browser games, one is just a shitty sidescrolling beat 'em up / shooter hybrid. The other's a physics / gravity / puzzle game.
I have a question though. To everyone currently making cool games - how long did it take you to be able to make your first good, "finished" game? How did you move away from prototype ideas and placeholder mechanics / art / menus?
Because I'm still at that stage. I probably will be for another two years, assuming I keep making things. I guess I really just want to know if it's a common thing to not be good when you first start.
Also, are any of these concepts connected to not using actionscript 3 >_>
(something something craftsman blaming his tools)
So far I've really just been doing Actionscript 3 and a bit of Java, which are both really easy languages to get into. So far I've made two little browser games, one is just a shitty sidescrolling beat 'em up / shooter hybrid. The other's a physics / gravity / puzzle game.
I have a question though. To everyone currently making cool games - how long did it take you to be able to make your first good, "finished" game? How did you move away from prototype ideas and placeholder mechanics / art / menus?
Because I'm still at that stage. I probably will be for another two years, assuming I keep making things. I guess I really just want to know if it's a common thing to not be good when you first start.
Also, are any of these concepts connected to not using actionscript 3 >_>
(something something craftsman blaming his tools)
The first few things you make will suck (in terms of realizing a vision). Get through them, and you'll better start to understand the overall process and what you personally need to focus on to see a project to successful completion.
This is the sort of thing that can drive one nuts about GameMaker style tools, it can work great to get everything on the screen and get mechanics almost working, and then you may be left with crazy floaty physics and/or weird glitchy collisions and bouncing unless you spend a bunch of time on it...and in a game jam event, that's time you almost certainly don't have.
You should write a tutorial telling everyone how to get awesome collision working.
The Technomancer
card-carrying scientician
Yeah, when my brother and I built a platforming engine in GameMaker like five years ago we ran into an issue where the character would adhere to walls, so we just turned it into a feature Megaman X style.
Well, this iteration turned out being mostly plumbing. Got pausing, menu placeholders, hud placeholders and a lot of other basic code for inventory you can't see. Variable sprite frame sizes and durations, attacking, sideways digging. I had to push off enemies (the real meat) until next iteration because I was missing necessary pieces.
I feel a little burned from this release because there's not enough complete and interesting (like every feature is half-way implemented) but it's still my biggest change so far. Hopefully I can find the strength.
http://dl.dropbox.com/u/13787691/caverns/caverns.html
I feel a little burned from this release because there's not enough complete and interesting (like every feature is half-way implemented) but it's still my biggest change so far. Hopefully I can find the strength.
http://dl.dropbox.com/u/13787691/caverns/caverns.html
Well...yeah.
"Hey guys if that TV was only $100 instead of $900 I would have bought it."
This is the sort of thing that can drive one nuts about GameMaker style tools, it can work great to get everything on the screen and get mechanics almost working, and then you may be left with crazy floaty physics and/or weird glitchy collisions and bouncing unless you spend a bunch of time on it...and in a game jam event, that's time you almost certainly don't have.
You should write a tutorial telling everyone how to get awesome collision working.
I don't feel attached to code, if you're serious I can tell you.
Yes I was serious. It would be awesome for future users to have a reference. I know it might be a lot of work to make a full tutorial on how to set the proper collision up, but even without a tutorial, if there are any special tips or tricks those would be good to know.
AcceptableGhost
Member
For the basics of a basic 1:1 bounding rectangle collision system that I made way back it was basically this (for people who want to program in their own collision system or whatever):
(brief summary obviously)
class "Body" with:
private vector2 position
private vector2 proposed
public method move(x, y) - add to proposed (used in normal gameplay for moving dudes around)
public method internal_move(x, y) - add to position (only collision engine uses this or if you need to move without collision checking)
public get methods for position/proposed/rectangle
public method CollisionStep() - add itself to collision engine for collision checking
class "CollisionEngine" with
public method Update() that
1. apply gravity to bodies using move
2. step through Y, THEN X of list of body's rectangles which... (repeat 3-6 twice, once for Y, THEN X)
3. apply proposed x or y
4. do quick three quadrant sorting (six would be easy too, i was just lazy) to make sure you only do checks against bodies that are close
5. check the x or y axis against bodies in the same quadrant
6. if intersecting then push opposite (using the internal_move) till not intersecting (doing it this way you can also easily report to your body that it was pushed LEFT/RIGHT/UP/DOWN or whatever with an AfterCollision delegate or something).
7. clear proposed vector on all bodies and clear the list of bodies
Fairly quick and fast (and your entities know which way they are being pushed), and you can add in checks against your baked in level collision or whatever. Can also add in whatever hacks you need like frame advantage checks or collision type checking or whatnot. I made it after dealing with Farseer (so from the front end it acts like that in use in my game) and I decided I did not want to deal with collision engines based in real physics for most of my games.
(brief summary obviously)
class "Body" with:
private vector2 position
private vector2 proposed
public method move(x, y) - add to proposed (used in normal gameplay for moving dudes around)
public method internal_move(x, y) - add to position (only collision engine uses this or if you need to move without collision checking)
public get methods for position/proposed/rectangle
public method CollisionStep() - add itself to collision engine for collision checking
class "CollisionEngine" with
public method Update() that
1. apply gravity to bodies using move
2. step through Y, THEN X of list of body's rectangles which... (repeat 3-6 twice, once for Y, THEN X)
3. apply proposed x or y
4. do quick three quadrant sorting (six would be easy too, i was just lazy) to make sure you only do checks against bodies that are close
5. check the x or y axis against bodies in the same quadrant
6. if intersecting then push opposite (using the internal_move) till not intersecting (doing it this way you can also easily report to your body that it was pushed LEFT/RIGHT/UP/DOWN or whatever with an AfterCollision delegate or something).
7. clear proposed vector on all bodies and clear the list of bodies
Fairly quick and fast (and your entities know which way they are being pushed), and you can add in checks against your baked in level collision or whatever. Can also add in whatever hacks you need like frame advantage checks or collision type checking or whatnot. I made it after dealing with Farseer (so from the front end it acts like that in use in my game) and I decided I did not want to deal with collision engines based in real physics for most of my games.
There definetely are some tricks to know about for collisions to work properly in GameMaker. I won't code them here but I can certainly tell you how to proceed, what to use, etc.
- First off, put that thing in 60fps. This saved me ALOT of wierd engine behaviors for some reason.
- Your collision should be a block, whatever size (preferably not less than 8x8) and put to "solid".
- Your character sprite should have its origin to "center" and the bounding box set to "full sprite" and "rectangle". Not having a difference between your collision mask's size and the bounding box of your sprite will simplify many things and prevent glitches.
- Put all your detection and collision related stuff in a "Step Event" in your character. Having seperate stuff everywhere or seperated for some reason in "Begin Steps", "Step" and "End Step" is asking for trouble.
- For the character's collision now: Have 4 "collision rectangles" around the bounding box of your sprite (1 on each sides, covering the whole side). They should be 1 pixel wide and adjacent to all the different sides of the box. They need to stretch outward according your "speed" variable. (the faster your character will move, the bigger the rectangles will get and you will detect farther away from your character. This is a must when moving.) They need to detect your collision block (sorry if its captain obvious..)
- Now your character should collide properly (almost) but will sorta bounce on the ground or stop a couple of pixel before actually touching it for some reason (yay for Gamemaker!). The solution is to have a second collision rectangle under the character that is 1 pixel of height but this time it's placed exactly where the bottom of the bounding box is. With this you can go with a condition like: "if that bottom collision rectangle does not collide when hero is on ground" and then do "move contact solid" with a max distance of 1 or 2 pixels (just so the character looks exactly like its on the ground).
There's one last thing. Your character can still glitch a bit through corners. You can place some rectangle again, flush with the bounding box corners and something 2x2 pixels. When they touch, do a "move outside solid" in the direction you need, the character will never stay stuck anywhere.
- First off, put that thing in 60fps. This saved me ALOT of wierd engine behaviors for some reason.
- Your collision should be a block, whatever size (preferably not less than 8x8) and put to "solid".
- Your character sprite should have its origin to "center" and the bounding box set to "full sprite" and "rectangle". Not having a difference between your collision mask's size and the bounding box of your sprite will simplify many things and prevent glitches.
- Put all your detection and collision related stuff in a "Step Event" in your character. Having seperate stuff everywhere or seperated for some reason in "Begin Steps", "Step" and "End Step" is asking for trouble.
- For the character's collision now: Have 4 "collision rectangles" around the bounding box of your sprite (1 on each sides, covering the whole side). They should be 1 pixel wide and adjacent to all the different sides of the box. They need to stretch outward according your "speed" variable. (the faster your character will move, the bigger the rectangles will get and you will detect farther away from your character. This is a must when moving.) They need to detect your collision block (sorry if its captain obvious..)
- Now your character should collide properly (almost) but will sorta bounce on the ground or stop a couple of pixel before actually touching it for some reason (yay for Gamemaker!). The solution is to have a second collision rectangle under the character that is 1 pixel of height but this time it's placed exactly where the bottom of the bounding box is. With this you can go with a condition like: "if that bottom collision rectangle does not collide when hero is on ground" and then do "move contact solid" with a max distance of 1 or 2 pixels (just so the character looks exactly like its on the ground).
There's one last thing. Your character can still glitch a bit through corners. You can place some rectangle again, flush with the bounding box corners and something 2x2 pixels. When they touch, do a "move outside solid" in the direction you need, the character will never stay stuck anywhere.
The Technomancer
card-carrying scientician
Yaaayaaay, I got sensors working in Box2D so I can restrict jumping to when you're standing on a flat surface. All I need to do now is adjust the player's in air-momentum so that you can't do a perfect 180 in mid-jump and I'll be ready to start implementing my data framework around this system.
RelentlessRolento
Member
I had collision problems last week as well in GameMaker but started trying to use "check if [object] is at x,y position"... does the same thing as collisions without the restraints.
I had some crazy idea ... but before going any further i need to know .... how much of the pallete of a sprite can you change with programing ?
I know that you CAN change sprite palletes with programing ...
But is it possible to do more specific changes like going from
to
with only programing or you can only do this with new sprites and what you can only do with programing is make a "red tint" or something like that ?
I know that you CAN change sprite palletes with programing ...
But is it possible to do more specific changes like going from
I had some crazy idea ... but before going any further i need to know .... how much of the pallete of a sprite can you change with programing ?
I know that you CAN change sprite palletes with programing ...
But is it possible to do more specific changes like going fromtowith only programing or you can only do this with new sprites and what you can only do with programing is make a "red tint" or something like that ?
That should be doable. I don't see what could prevent you from changing the colors --- an image itself is code after all. However, it does not mean that all editors or scripting languages would let you do that though.
Been making some good progress lately. Learning a lot really fast. Still super early, no artwork still playing with primitive characters, no sound, GUI is basic etc. The gameplay feature set , in one mode, is about 50%. But it's actually starting to get almost fun. Think StarCraft2 Desert Strike + Frozen Synapse + Battlefield. Can't wait to get it to a preview state.
I had some crazy idea ... but before going any further i need to know .... how much of the pallete of a sprite can you change with programing ?
I know that you CAN change sprite palletes with programing ...
But is it possible to do more specific changes like going fromtowith only programing or you can only do this with new sprites and what you can only do with programing is make a "red tint" or something like that ?
This is exactly what pixel shaders do. How clever you want to be with it is up to you, for example you can use the alpha of each pixel to set whether it's locked or not and tint all with alpha > 0 or something.
I'm guessing at a simpler level, paletted/indexed images can do that too, right? I don't know if there are certain ways to do it more efficiently however, where each pixel of some image contains (for example) a small integer that indexes into your palette of colors, and drawing the image with a different palette just means changing the palette before it's drawn.
I seem to recall that Notch did something like that on a simple level with his Minicraft game from the last Ludum Dare, but his code was not the easiest to read.
It's late so I don't know if this post makes sense. XD
I seem to recall that Notch did something like that on a simple level with his Minicraft game from the last Ludum Dare, but his code was not the easiest to read.
It's late so I don't know if this post makes sense. XD
The Technomancer
card-carrying scientician
Makes sense to me, that's how the program Zelda Classic works
If you are looking to compress your assets beyond single image formats that's another strategy.
ThatCrazyGuy
Member
Sorry to butt in guys. I am been doing some research about possibly making an iOS game the last couple months.
Then this article popped up on gamasutra about iOS games and who is making profit and how:
http://gamasutra.com/view/news/3965...igures_that_might_help_an_indie_developer.php
Does some of this ring true to you guys?
Then this article popped up on gamasutra about iOS games and who is making profit and how:
http://gamasutra.com/view/news/3965...igures_that_might_help_an_indie_developer.php
Does some of this ring true to you guys?
One final note on the palette swap thing, if memory isn't a big concern then the easiest thing is probably to write some code (or use an existing utility) to color swap the sprites before the game starts, so you just have multiple images of different colors and don't have to worry about drawing them in funky ways at all.
Do you guys know where I could learn the workings behind and how to make a 2D scaling game like Outrun? You know those old racing game 2D engine I am talking about?
Rad Racer 2
I am curious how it's done and then figure out if it can be done at all in GameMaker or some other simple engines.
Rad Racer 2
I am curious how it's done and then figure out if it can be done at all in GameMaker or some other simple engines.
Teknopathetic
Member
I actually looked into that a week or two ago and found this page:
http://www.gorenfeld.net/lou/pseudo/
http://www.gorenfeld.net/lou/pseudo/
AcceptableGhost
Member
If I remember correctly it just scales the sprites up or down in size to make that effect. I doubt the NES could do that in real time though, so I bet it had all the sprites sizes pre-made on the cart.
Edit:
I guess scaling is a part of it then.
IchigoSharingan
Member
30% cut.
Bleed us dry, bleed us dry.
I help with design/marketing for www.godofaxion.com indie devs. A lot of the topics covered in that article are covered but not much because none of us have eyes bigger than our stomachs.
But what REALLY stands out is this:
Myth #4: Being visible on the App Store just takes a good post on reddit or a good viral video
Once you have a good game, the key to success is visibility on the App Store. Another tale I’ve been told many times (and that I actually wanted to believe) is that you can leverage big communities with a nice forum post or a cool and cheap video. I believe now that this is a waste of time. You can’t influence a community unless you've already been in this community for a long time. And viral videos suffer even more from the "jackpot syndrome" than the apps themselves, in the sense that you can’t at all predict if they will get 12 million views or 300 (although 300 is more likely).
Just accept it: being visible will be a long and tough battle that you’ll have to fight from the day you start to code, to a year after the launch.
This could not be any truer. A lot of indie devs fight over this. Some still believe in 'overnight' marketing. I'm sorry but communities are not stupid. You have to let them know who you are over time, if they even give you the opportunity. And if they don't, respect their choice and try elsewhere.
hmm, looks like quite a crazy thing to reproduce in GameMaker lol
You mean without installing Itunes on your PC?
The page of each game is available online but I don't it it's possible to "browse" the whole store.
Anyone know a good site for free SFX? Specifically ones aimed at games maybe?
Having a decent budget is very good, but not a total necessity. If you are with two guys and you both have all the areas covered to get a game out, you don't need a big budget (or any really). You can compare it with a scale if you will. If your budget is big, the talent and skill won't have to be as big, and vice versa if your skill is beast then your budget doesn't have to be very big
This probably isn't what you are looking for, but in case you or anyone else isn't aware of SFXR: http://www.drpetter.se/project_sfxr.html
There's also a web version that looks more powerful: http://www.bfxr.net/
Lastly there's a Google code page that MAY have a more up to date version of SFXR, but I'm not sure what has changed: http://code.google.com/p/sfxr/
Any of these versions are especially useful for game jam projects since you can mess around with the sound effects until you get something that matches what you want.
They are tools to generate sound effects as opposed to websites that have a bunch of premade royalty-free sound effects.RelentlessRolento
Member
I use free sound project (google it... On iPod right now)
This is perfect! I forgot to mention it but I am always searching for more sound creation programs. I use free VSTs as well to make sounds and it works wonderfully. Some easy to use ones I've worked with are magical8bitplug3 and Basic_64 (commodore sounds). You can make whole soundtracks with them if you desireThis probably isn't what you are looking for, but in case you or anyone else isn't aware of SFXR: http://www.drpetter.se/project_sfxr.html
There's also a web version that looks more powerful: http://www.bfxr.net/
Lastly there's a Google code page that MAY have a more up to date version of SFXR, but I'm not sure what has changed: http://code.google.com/p/sfxr/
Any of these versions are especially useful for game jam projects since you can mess around with the sound effects until you get something that matches what you want.
Will check this one out tonight
This is my goto site as well but I found it harder to search for VG sounds, I've found some pay-for sites like soundrangers.com that have wonderful categories perfect for videogames but at 2 bucks a pop I'd only use it for serious projects
RelentlessRolento
Member
If you need game stuff this is awesome... I've used it before and can back it up. Plus the band who made it is already awesome.
http://www.ymck.net/magical8bitplug/index.html
I mentioned it right above your quote
But I can vouch for it being awesome! Haven't heard much from the band except for a song or two they made with the plugin but it sounds real good
Similar threads
- Locked