Jump to content

  • Log In with Google      Sign In   
  • Create Account


Member Since 11 May 2010
Offline Last Active Today, 08:37 AM

Topics I've Started

So I got another random idea

07 July 2014 - 12:28 PM

Okay this time the idea is small, and I kind of implemented a fast prototype for it already. I kind of want to get some kind of feedback.


So the game is a maze game, but not any old simple maze game. 

You traverse the map using "commands"


Currently, I have implemented only "move" and "turn". 

When you send your program to traverse the maze, you must return to the start point for your program to provide feedback to you.

Any program that didn't return to start point will just "terminate" and no feed back will be provided.


The maze will look something like this

Attached File  maze.png   14.48KB   0 downloads


The program is a "String"

It must start with "b" and ends with "e"

So the simplest program you can write to traverse the maze will be "be"


Attached File  basic.png   15.75KB   0 downloads


So when you first enter the maze, it will tell you where you can find empty space. 

The next step will be to move and explore more.


So here I ran the command "bm4tbm4e"

Which is saying "Begin", "Move 4", "Turn Back", "Move 4" , "End"


Attached File  advanced.png   33.5KB   0 downloads


If you didn't get back to the start point you will get this

Attached File  fail.png   5.67KB   0 downloads


So this is really the basic idea. The idea of the game is to say, try "data" that are hidden in the maze. You need to uncover the maze to find all the data.


Additional planned challenges. (if i continue to work on the project)
1) Different movement cost. Right now all tiles cost 4, but later there will be tiles that cost 1, 2, 8, 16, which will kind of confuse your map.

2) Marker command, that allows you to place a "marker" on the floor and allow you to detect "loops"


I am planning(maybe) to write this as a "web" game, which is why I use python. That will allow me to convert to say Flask/Django easily. I probably can get it done by tomorrow with a good interface. But I kind of want to get a feedback of this stupid random idea that came into me today.


oh and anyone who want to play this game can get it at https://github.com/ZwodahS/mazehack


It is CLI at the moment, to run it you just need to find the test.py, and run "python test.py <seed> <instructions>" and add a "DEBUG" to the end of it if you want to see the map

The seed is there to make sure you get the same maze tongue.png


So to start, you should just type : python test.py <seed> "be" 


I figured that most of the people here are developers and should be at least intrigued by this >.< but if this is a stupid idea, feel free to flame me tongue.png

Let me leave you with one final screenshot

Attached File  funtime.png   107.98KB   0 downloads

Need help - can't find the fun factor

27 June 2014 - 12:10 AM

I wrote about an idea of a Fantasy Guild Simulation game here about 2 months ago and I have been working on it on and off and kind of hit a brickwall. I can't find the fun factor in it at the moment.


So here is the idea so far.



The main premise for the game is that you control a guild of fantasy-based heroes, mages, knight etc. They work for you to help finish quest that towns have. The game is fully text-based at the moment, similar to that of "Long live the queen". 

The game operates in on a day-basis. Each day begins by you looking at what have happen the previous day, collect the rewards from town and returning the heroes back to town. After that you decide what you want each hero to do for that day. Those that are still on mission will not be assignable. After that you "end the day/turn" and the world will simulate your decision and cycle continues.


Towns provides various type of quests. Some requires you to slain various type of monsters in various places, or gather certain amount of items from various places.

You can send heroes to locations to "explore" them, allowing you to kill monsters or gather stuffs from there. Combat will occur when you encounter enemy and will be resolved automatically.


Combat resolution is done using a ActionPoint/Speed system, like most FF title, except that instead of you choosing the moves that the unit will perform, they will choose randomly, or based on their traits. For example, if a character has a "Thrifty", he will more than likely choose a skill that cost him less energy, or if a character is "Flashy", he will more likely to choose skill that is more specular.

The other thing you can do is to train the heroes. This do not gain you gold/resources but improve the character faster. However, they generally don't like to train since they also don't get gold if they don't go out on mission.


Heroes with specific skill can also create items; Weapon making or potion brewing for example. The quality of the item that is made is also depend on the skill of the unit. 




So far I have implemented basic exploration of places, a simple combat resolution, and some basic recruitment stuffs. I implemented them mostly using curses instead of a graphics library. 


There are a few problems that I realized after playing a bit of the game. Firstly, it seems fun to watch your character grow at first, but there are no way to see how much they have grow.


Take Pokemon/FF game for example. They have a simple combat system and as your pokemon/char grow, they learn new skills and you directly see the impact of this growth. In this simulation game, since the combat is automatic, the feedback is not that obvious. If you can't see the impact of the growth of the characters, then the growth fun factor isn't that fun anymore.


On top of that, it seems like this game has only 1 core feedback loop and it is not as fun as I thought it would be. All you seems to do is sending units on missions, train/upgrade units, then send them on more missions. 


So the question for fellow game developers are 

1) Do you think the game I describe have any audiences ?

2) How would you improve to solve the problem I mentioned ?


I really like this idea but I can't see myself enjoying it, at least not with the problems I mentioned. I might be too close to it to see the "fun", so I hope you guys can give me some feedback on what I have now. 


(I did thought about using a non-automatic combat resolution, like FF-tactics or as simple as traditional turn based combat. However, I realized the core part of the game is character building and not combat/tactics. Having the player directly control the combat will shift the time that the player spend on management stuffs from 100% to about 15 ~ 20%, meaning the combat will take up most of the player's time. Each day will be longer, since after you decide what you want to do and where you want to explore, you will have to spend even more time resolving them.)

Switching between languages/API

23 April 2014 - 01:00 AM

Hi, not really a beginner question but probably fits here than anywhere else.


I was coding SFML/C++ for the entire year of 2013, and switched to Libgdx/Java this Jan to port a game I made in C++ to android. After playing with Libgdx for about 2~3 months, + a 1 week break without touching any coding, I faced a problem. I wanted to prepare myself for Ludum dare this week and intended to use SFML/C++ instead of Libgdx, since I like SFML more. I realized I couldn't remember most of my C++/SFML stuffs. (I will probably have to read through my old projects). It is not that I forget the syntax but to do extremely trivial stuffs, like drawing sprite to screen, using my own personal code, or even just loading of assets, I have to reference my old projects to get started.


So my question is, what are some ways that you guys do to prevent or reduce this "confusion state" that happens when you switch from one language to another after using one for significant time.  I don't think I am the only one that experience this. Does cheatsheet helps ? I tried creating one in the past but it became extremely tedious and pointless when you are using the language but become extremely useful when you forget them.



Fantasy Guild Management Sim

22 April 2014 - 01:05 AM

The basic idea that I have is some sort of a simulation where the player control a guild/clan in a fantasy world. I originally thought of this when I played FFTA, and I really like the part where you send out units on mission and they grow as they do more missions. The idea reappear when I started reading the manga "Fairy Tail". I did my research and also posted a thread here to ask for similar games. I couldn't find anything close to what I want. The idea was put aside for a while as it seems too big for me to tackle as a newbie back then. Now that I am a bit more "experienced", I kind of want to visit this idea again.


The basic experience I want the player to have is to manage the members of the guild. The core experience lies at managing who to send out on mission, managing the "financial" aspect of the guild, like expansion, forging of new weapon etc. This game is not meant to be a challenge for players, but an experience like that of dwarf fortress.


Basic Limitation of the game

I first set out the ground limitation/features for the game. Here are some that I have and explanation for them.


Mission success/failure is not random, but heavily based on who you send out and how many you send. (Duh ?)

In additional to this, I also want the player to have minimum decision when the party is executing the mission. FFTA make it such that you totally have no control over the success of the mission the moment you decide who to send. I kind of want the player to know why the party fail to complete the mission and how to improve on the success rate in the future.


Combat should also be fast and simple

Since I want the whole experience to be about managing the guild, and not combat, the time the player spend on combat should be relatively shorter than the time he spend on deciding what to do, planning the training schedule etc. 


Evolving world

The world must evolve based on the player decision. For example, evil guilds may pop up and you may want to stop them from growing. If you don't they might start to kill off other good guilds or even assassinate your party while they are on missions. You may decide to help a town build up their farm by sending a few members to help out, and in return they provide you with some foods.


Story generating

What I want is a game where people can share their story on what happens in their world. Like how one of their members became so strong that he destroy the entire evil guild single handedly, or how the guild leader was assassinated by someone within the guild etc. 



These are the general ideas I have on the game, and here are some ideas on the mechanics that I want to try. They probably don't work as well as I think they will but perhaps I can get new ideas from you guys.



This is the main reason why I am revisiting this idea. I was on a very long bus trip and wonder what is a good alternative to a "send and pray" method when you send people out for mission. I kind of came up with a simple die roll idea and evolve it to this state.


So mission are basically series of "stats check". Let me start with an example.

Say for example, your mission is to explore a dungeon that the people believes that someone is performing dark magic in. So say you send out 3 party members. When the party reaches the destination, the mission begins. In this case, the mission will be a series of "events". Each event will look something like this.


[Event : Trap] : Perception check > 5. 
This means that it check if any characters has a perception value of higher than 5, if it does, the trap is not trigger.

However if no one has a perception more than 5, then the trap triggers.

[Event : Trap triggers] : Roll Dexterity > 5 , else take 5 damage.

This means that each unit in the party needs to roll their dexterity, and if it is less than 5 that unit will take the damage. 

Rolling a stats means converting their stats to a 1D(S) die, where S is the value of the stats of concern (i.e. if the unit has dexterity of 8, it will roll a 1D8 die).


Some events do not have a negative impact, but required them to clear it before the player can proceed. For example :

[Event : Boulder in the way] : Roll total strength > 20

This requires the player to roll a total strength of > 20 using all the roll from the party. In this case, sending out additional members would help in this event. However, sending too many members will waste valuable manpower.


There are many other type of events, checks that can be thought of. I am just stating some of the basic ones that came to mind. Some events can also be combat instead of checks. In order to complete the mission, the player must complete all the events in the mission. Some missions have partial rewards if the player couldn't complete the mission.


To allow the player to have some agency in the stats check, I thought of implementing a energy system(not the kind in mobile game). Each unit will provide a certain number of energy, which can be used to do many thing. One of which is to allow the player to reroll a specific die, like in the case of the event "boulder in the way", the player needs to reroll the dice until they succeed. If the energy runs out, the player has no choice but to abandon the mission. In the first Trap event, you could also use energy to reroll the dexterity die to try to prevent the damage.


I am not really sure how fun this system for mission is but it seems to be better than any "send and pray" mechanism.



Now this is something that has been bothering me for a while. I can't think of any short and fast combat that has agency. I want it to be related to the mechanics I used in mission, to make it more consistent for the player to follow.



There are 2 problems with the world that I am concern. The first is whether to craft the world or generate it. The second is whether to use a grid style world or a graph style. Since they have no effects on the other systems, this problem can be discuss independently.




The reason that I am posting here is to get some inspiration for the combat systems and to see if this game idea in general appeals to any one. If there are anything that is unclear, let me know.

Question regarding Box2D

07 April 2014 - 09:54 PM

I had implemented a few platforming games (just to learn the basics), and had always use simple AABB collision test. 

If I just want to make a simple platformer game like say Mario or Megaman, the kind without complex Physics, is there any reason to use Box2D or is there any reason NOT to use Box2D ?


I have always been told by friends that I should use a physics engine instead of writing my own collision code, but I always have the impression that Box2D will be overkill for a simple platformer game. Advice?