Temple of The Roguelike Forums
Development => Design => Topic started by: AlexPT on November 17, 2013, 07:10:27 PM
-
Hello all. ;D (sorry if I posted this in the wrong place, but I think this is the right place for it? :-\)
Ever since I was a little fella, I've been wanting to make a game... but I think that having my ideas in one would suffice. (I could do concept art, but I'm still just a beginner artist and I'm only practicing, still not evolved enough, I could also try to sprite for someone who is willing to do these ideas, but I'm not sure if they would even come out good)
I'll be throwing all of these for you guys who might be interested in making an Old West Roguelike. (and with a tiny bit of fantasy!)
An open world of:
--A vast, randomly generated world, consisting of:
A Desert, where there would be:
Ruins as Dungeons (deep in the dungeons there could be cities like El Dorado and other lost cities);
Towns with shops, a diversity of NPCs and random quest givers (some towns with mines as dungeons);
Nomads, some hostile others paficic/neutral;
Desert animals, and maybe an option to catch them and make them your pet/mount!;
Maybe monsters? Like, giant sandworms.
A Jungle, where there could be:
Tribes, some hostile where you could maybe dialog with them, some paficic/neutral;
Ruins and temples as dungeons (and maybe deep in them lost cities too? :D)
Jungle animals too, of course, and maybe, too, an option to catch them and make them your pet/mount!
Maybe monsters?
Maybe even more if it's possible? Like, some isles, some full of dangers and others paradisiac to build your house :D
--Weapons: all of these weapons from the 1630s to the 1880s old wild west, and some not, well, it's up to whoever picks up these ideas, I did say, though, that they'd have some fantasy
Rifles;
Pistols;
Swords;
Spears (from the indian tribes or not?);
Shields (from the indian tribes or not?);
Maybe "we" can even put a bit more of fantasy and include more, for example, warhammers, etc... ?
--Armors: all of these armors/cloaks, too, from the 1630s to the 1880s old wild west and some not, well, it's up to whoever picks up these ideas, I did say, though, that they'd have some fantasy
Cowboy hats (normal hats too);
Ponchos as cloaks (and maybe indian cloaks);
clothes as armor (maybe vests too?);
armors (what you normally see in a fantasy game and such)
Leggings (leg armor too);
Boots (cowboy boots, armor boots, etc...);
Indian headwear (you know what it's like)
--Mounts & Pets:
Horses;
Camels;
Raptors as mounts? (come on, that'd be awesome! :D y'know, part of the fantasy stuff I talked about)
Dogs;
Parrots;
etc...
--AI Companions
--Character Generation/Creation:
Races:
Humans (sub races: Caucasian, Hispanic, Indians?, etc...);
Dwarves?;
and more fantasy races? :D
Classes?:
Gunslinger;
and more (I just don't know the right names for them. :/)
--Gameplay
Real time roguelike (or turn based, but real time would be different (and imo, better), and perhaps, the old cowboy duels would be possible in it?);
Plot? (I don't really like to have a plot in a roguelike, but that's me, and I'm just giving ideas)
Random quests and bounties;
The ultimate goal would be up to you, for example, conquer this old wild fantasy west, become the best cowboy, help people; survive, etc...;
Crafting;
Bulding?;
Cowboy Duels (if it's possible);
Any mount Races?;
Multiple choices (chose to be good, bad or neutral, well, I think that's covered in the "The ultimate goal would be up to you", oh well xD)
And more?
That's it from me, folks. If anyone is willing to pick these up, please do say so here and let me in on the dev updates! (I also want to test it!!)
Also, I don't know if anyone has these ideas tho, can't read minds, unfortunately, and if there is already something like this in developtment, which I doubt?, link me to it!!! :D
-
Why not fire up a beginners python tutorial when your fresh off a high from thinking of all these cool ideas, start making your game on paper (write down the features you want, plan the gameplay etc).. Then after a month or so when you feel comfortable with python (it's not too difficult to learn), grab a copy of libtcod and get making the game yourself (in ascii at first).
It's a much easier to ask for help making some tiles for your game than it is to try to 'inspire' someone into making your perfect game ;)
Besides, what could be more fun than watching your game gradually come to life as you become a better and better programmer? Isn't that part of the point of making a roguelike?
-
Yeah, there's no shortage of idea guys in the roguelike world. Programming isn't as difficult as it seems, and a simple roguelike is an ideal beginner's project.
-
Raptors as mounts? (come on, that'd be awesome! :D y'know, part of the fantasy stuff I talked about)
A wild west roguelike with raptors as mounts? WHY ISN'T THAT THE TITLE?
I actually like that, and the idea of the sand worms a lot. Maybe shamanistic magic could be real also.
I think the real question, though, is whether this game will include authentic wild west dialogue, like: "I'm gonna fill you full of lead, you yellow bellied coward!!!", and judicious use of the word "mosey", as in: "I'm gonna mosey on down to the saloon to get me some whiskey!"
-
You sir have come to the right place for throwing out ideas.
I too am more idea man than programmer or spriter. I LOVE designing systems and little games, but rarely get far in the coding.
No one is going to actually make your game for you though, or include anything except maybe your artwork. Or maybe an RPG system you designed. I've seen both of those happen. The thing is hobby developers have their own ideas, probably 100's of them, and they make RL's to express them. Making someone else's game is all work and no fun.
But hey, I could be wrong, I've just never seen it done.
I use a program called Game Maker to make my ideas come to life. It's actually a pretty easy system to use, I've done several 7DRL's now with it. The learning curve is MUCH lower than C++ or Python.
Also there are several sample roguelikes out there you can convert over to your purposes, and there's the T-Engine as well.
One caveat to throwing out ideas. They aren't protected at all. Some people think they are and then get butthurt if they see someone else do something similar. Please don't be that guy. Expression is protected, the actual words, designs, code. Not the idea. Myspace couldn't sue Facebook, for example. It makes sense too, the expression of an idea is the real work, the idea part is all fun, of little economic value.
Probably you didn't need to know all that, but sometimes people are confused.
-
Hi, welcome, and thanks for an interesting post. Your ideas resonate more than a little with what I envision for my own project Land of Strangers (http://forums.roguetemple.com/index.php?topic=3221.0). It's still in its very early phases of development, and going slowly. So I'm more than glad to see some Weird West ideas being tossed around. Raptors as mounts! :o that's definitely on my wishlist now, unless you mind ;)
If you didn't know it, check out Abura Tan (http://roguebasin.roguelikedevelopment.org/index.php?title=Abura_Tan), which was developed up until about ten years ago. Also, there were a few failed attempts at cowboy RLs in this year's 7DRL competion. Let's see … Here (http://hickathriftgames.co.uk/gun.html) and here (http://7drl.org/2013/03/17/lawman-failed/).
As always,
Minotauros
-
@Trowel & Vanguard
Well, I've thought of learning code myself, but it all seemed very confusing, but maybe I'll actually try and learn. :)
@Gr3yling
Ahah, I actually never thought that my ideas would get replies, but I'm very happy that many people like them.
Shamans? Oh boy, why didn't I think that? O:
Wild west dialogue is pretty badass and that would be very awesome. :D
(and this is why, two heads thinking are better than one, I also forgot to include bows in the weapons, lol)
@Jo
Yeah, what you say is indeed true. (if I had money I'd actually hire professionals to do my dream game XD :b jk)
Well, I might try to learn programming, also, I've used Game Maker before and made some games moking a friend of mine, lol, but it was something really simple and stupid, and well, I might actually learn programming for real this time. :) But I don't know if I'll use GM, tho. xd
Yeah, they aren't protected, I know that, and I will not get butthurt. :D
@AgingMinotaur
Ahah, I definitely don't mind, am actually happy that I'll get to see raptors as mount in a wild west game! :D
Thanks for the links! I'm definitely keeping an eye on your project. :P Stalker style. 8) not xD
-
That's great buddy! I always try to point that out when a newer person raps about ideas.
If you look at my past threads I yap yap about ideas and mechanics all the time. It's fun, sometimes valuable to other people, often not, no biggie.
If I had the money my Super Hero Epic RL would be a reality. Guaranteed. I know I lack the drive to do it, I just don't like programming that much, but the idea is great.
As for Gamemaker, you CAN make a good RL with it. I've seen it done more than once. Any programming language will do though. Most of the big ones have libraries that make the basics pretty straight forward.
In the end if you want to make a game you gotta know how to code at least a little. I stick to mostly board and card game making because of the coding issue, but have branched into RL making in the last few years because you can do one without a ton of skill overhead, meaning you don't need a team of artists, sound, effects, etc...you can do ascii bumping around and still have a pretty awesome little game, I love that.
-
Also if you were interested in Gamemaker at all I would send you the source to any of my RL games, which is only 2 that I still have. Klingon and Sun Crusher.
-
Ideas ftw! ;D
Pheraps one day you'll have the drive to do your dream RL game. :)
Game Maker always seemed very limited to me... :X but then again, when I used it, it was a lot of years ago. XD And I'm also a beginner, so everything will prolly seem limited?
I don't think that I'll use it.
Well, I don't really know if I'll learn programming, I was all fired up yesterday, but I didn't get very far on the tutorial. XD I'll definitely give it another try, though.
I'm also working on some pixel sprites for these ideas I posted. (are they sprites even if they don't get animated? w/e)
-
What exactly are you trying to do when you say you looked at a tutorial to "learn programming"? What language?
If you're really just interested in making games, and not really that interested in programming, I'd recommend not trying to learn a major language like C++. It'll take a long time for you to even get to the point you're making a game, because you'll have to learn all this low-level stuff first.
Someone mentioned python and libtcod. I've never used python, personally, but I've heard it's a pretty easy language to program in. Libtcod is a library made for roguelike developers that handles a lot of the not-so-fun stuff already (like display and line-of-sight) so you don't have to worry about it. Again, never used it myself, so others here might have more to say about it.
There's also rot.js (http://ondras.github.io/rot.js/hp/), which I believe is a Javascript version of libtcod, for developing on the web.
You could also try writing a mod for the T-Engine (http://te4.org/). It uses a language called Lua, which is also very easy to use.
My game Possession (http://forums.roguetemple.com/index.php?topic=3159) was written using a Lua-based game framework called LÖVE (http://www.love2d.org), and there's a port of rot.js (https://github.com/paulofmandown/rotLove) for it, though it came out after I made Possession. There are also several other Lua-based game engines out there, as well). I'd be willing to let you look at the code for Possession, but I can't promise how readable or good it is...the majority of the code was written in only 7 days.
Game Maker is, again, something I haven't used, but it's not as limited as most people think. Some pretty complex games have been made with it. Off the top of my head I know that the original version of Spelunky, and Hotline Miami were both made with Game Maker.
So, I guess, there are a lot of options out there...even more than the ones I've posted. You just have to find what works for you and lets you make the games you want to make!
-
If there are dwarves, raptors, or sandworms, then it is not a Wild West game, IMO...
Indian magic, shamanism, and mythology would be OK, though. Or maybe the Mayan starship from Palenque.
-
@Rickton:
Well, I tried to learn Python, so I went to this website: http://www.learnpython.org/ and I got stuck on the String Formatting part, since I got stuck, I never got around to trying out this tutorial, http://roguebasin.roguelikedevelopment.org/index.php?title=Complete_Roguelike_Tutorial,_using_python%2Blibtcod .
Maybe I went the wrong way about it? :|
I don't know, or maybe I'm just not meant for it?
Anyways, thanks for the links, I'll definitely check them out thorougly. :)
@Z:
Well, dwarves, pheraps, but the thought of a cowboy riding a raptor seems very old wild fantasy west (does this genre even exist, tho? XD) to me, but that's me. :) And you indeed said IMO.
-
the thought of a cowboy riding a raptor seems very old wild fantasy west (does this genre even exist, tho? XD) to me, but that's me. :)
Cowboys and raptors somehow makes me think of Bravestarr (http://www.youtube.com/watch?v=_oBkWd_WHYs).
-
What exactly are you trying to do when you say you looked at a tutorial to "learn programming"? What language?
If you're really just interested in making games, and not really that interested in programming, I'd recommend not trying to learn a major language like C++. It'll take a long time for you to even get to the point you're making a game, because you'll have to learn all this low-level stuff first.
Someone mentioned python and libtcod. I've never used python, personally, but I've heard it's a pretty easy language to program in. Libtcod is a library made for roguelike developers that handles a lot of the not-so-fun stuff already (like display and line-of-sight) so you don't have to worry about it. Again, never used it myself, so others here might have more to say about it.
There's also rot.js (http://ondras.github.io/rot.js/hp/), which I believe is a Javascript version of libtcod, for developing on the web.
So, here's a question. Learning Python sounds tempting, but I noticed that Crawl, Nethack, ADOM, and DDA were all written in C or C++ (at least according to roguebasin). Is it possible to write a game with the types of features that ADOM has using Python? What are the pros and cons of C/C++ vs python in general?
Also, how long would it take someone completely ignorant of computer languages to learn python well enough that they could start making a game? I realize that's kind of a vague question, since games come in all sizes and levels of ambition. So, let's say we are talking about a game that consists entirely of randomly generated dungeon environments (no overworld map, towns, or non-hostile NPC's) and includes a few classes which are capable of using melee combat, ranged combat, and magic. Let's say there are also 50 total equipable and consumable items and 30 enemy types.
Could anyone give me a rough estimate about how long that sort of project would take? Assume I don't want to use a pre-existing engine because I'm stubborn and irrational.
-
Well, dwarves, pheraps, but the thought of a cowboy riding a raptor seems very old wild fantasy west (does this genre even exist, tho? XD)...
You just invented it. Not many people can say they gave birth to a genre, you know.
Honestly, somehow it does all fit to me, in a weird sort of way. Perhaps there was some sort of dimensional confluence, or perhaps the past and future were superimposed due to the discovery of a powerful Mayan (read: alien) artifact. I'd be interested to hear what you end up using for backstory.
-
I have read somewhere that a talented programmer is more than 10x faster than a typical experienced professional programmer. So it is hard to estimate, because it depends on how talented you are.
C++ is much, much faster than Python. It could matter for complex algorithms, like AI or other advanced pathfinding or procedural generation. In my experience, C++ things are easier to run (I sometimes run into problems with Python executables). And some libraries are only available in C++, like NotEye (although probably Python bindings could be created, like it has been done for libtcod). On the other hand, Python is easier to learn and beautifully designed. C++ is around for a much longer time, so it is possible that some experienced programmers use it because that's what they know best.
-
So, here's a question. Learning Python sounds tempting, but I noticed that Crawl, Nethack, ADOM, and DDA were all written in C or C++ (at least according to roguebasin). Is it possible to write a game with the types of features that ADOM has using Python? What are the pros and cons of C/C++ vs python in general?
I don't know if Python was even around when the games you mention started development. There are many younger RLs which utilize Python and similar languages. In any case, C and C++ are much faster than Python, but probably not so much as to make a huge difference when writing a Roguelike. You may encounter some lag, for instance with complex AI and map generation. It would definitely be possible to write an ADOM clone in Python. I make this claim having no knowledge of C, but having used Python a bit.
Also, how long would it take someone completely ignorant of computer languages to learn python well enough that they could start making a game?
Well, you'll be writing the game from the outset, wouldn't you, learning as you go along? ;) When I started my first RL, I was at about the point you describe. I had already looked a bit at Python, writing some very simple programs to manipulate huge chunks of text, but was more or less a blank slate. I think if you're able to do the design part (ie. envision a rudimentary rule system), you can get your hands dirty pretty much straight away. Maybe do some exercises to get to know the basic data types (strings, integers, lists, tuples, dictionaries) and start reading about classes (object oriented programming). You'll also need some external module just to handle drawing a window, getting keyboard input and so forth (not something you want to implement from scratch, no matter how idiosyncratic you are). libtcod might actually be worth looking into, although I use pygame myself, which is really simple if you need it to be, but still supports some advanced stuff you might want later. In addition to a basic Python tutorial to get a feel for the language, the official documentation really contains all you need to know. Same goes for pygame: Get started by googling specifics like "pygame draw sprite" or "pygame keyboard input" – just cut and paste the code you need – and then rely on the official documentation.
I think my very first attempt with python (before I threw pygame into the mix) was at a character creation interface, that I threw away. Then I looked up pygame, and had a "@ walking around an empty map" within a few days. I started from scratch several times, scrapping months worth of development at a time, but after a year or two I had a working game of about the scale you mention: a simple dungeon crawler with enough stuff to make it feel worthwhile. I decided to abandon the game three years ago, just about when it could have started to become interesting, but at least I learned enough from the process to feel comfortable with python and pygame.
As always,
Minotauros
-
Java is a good beginner language. It's pretty much easymode C++. The syntax and concepts are similar enough that if you want to learn C++ in the future, it'll be a quick adjustment.
-
Java is a good beginner language. It's pretty much easymode C++. The syntax and concepts are similar enough that if you want to learn C++ in the future, it'll be a quick adjustment.
No, it would be slow and painful. If you want to learn C++, start from C++.
-
Yeah, when it comes to what's easiest to learn in the Roguelike context, whatever you want really.
I started with Pascal. I betcha I could write a fucking AWSOME RL in turbo pascal.
-
No, it would be slow and painful. If you want to learn C++, start from C++.
It worked for me~
-
It worked for me~
Of course it works. Slowly and painfully :). Java experience does not make it any better.
-
Well, dwarves, pheraps, but the thought of a cowboy riding a raptor seems very old wild fantasy west (does this genre even exist, tho? XD)...
You just invented it. Not many people can say they gave birth to a genre, you know.
Honestly, somehow it does all fit to me, in a weird sort of way. Perhaps there was some sort of dimensional confluence, or perhaps the past and future were superimposed due to the discovery of a powerful Mayan (read: alien) artifact. I'd be interested to hear what you end up using for backstory.
Well, I have many many ideas, but the sad part is the lack of ability to implement them on a game, in this case, a roguelike, but I can learn though.
When I threw all these ideas together, I hadn't thought of a backstory, pheraps in the future, if I can make this game, I'll think of a backstory. (tho, I'll admit, I'm good at throwing ideas, but stories, I'm not so sure, it doesn't hurt to try though)
-
Thanks for all your advice, girls/guys.
I've started playing with python a little using the tutorial in AlexPT's link. So far it does seem a lot easier to understand than C++ (I made a very brief attempt to learn C++ but then stopped because I didn't know if it was the best language to learn for my purposes).
In python it seems so far that you can learn by playing with the commands, by switching out components and seeing how the results differ. With C++ it seems like trying to do that usually just resulted in a error.
I, personally, think I'm going to stick with this language.
-
Python is a great language, and it has good library support for game authoring in pygame and libcotd.
When performance becomes an issue, python supports you dropping in optimized small chunks of C/C++ to do the really CPU intensive bits, and most libraries for python are actually implemented in C/C++ under the hood, so the performance is very good compared to a "pure" interpreted language.
If I were starting a roguelike from scratch, I'd most likely build it in Python and only do performance-critical bits in C/C++, and this is from a professional programmer specializing in C/C++ :P
-
Hey AlexPT, I think you should check this site out: http://www.codecademy.com/tracks/python. The pace is much slower than the other site that you linked (which is good for me). Each concept is broken down into a number of small lessons, and that helps too. You might like it if you are still getting discouraged with string formatting.
-
I love this idea, I can really picture an ASCII Red Dead Revolver with permadeath and a dash of the Jurassic.
I'd be tempted to pick it up if I wasn't already a couple of months down my own RL rabbithole
Break out that Python Tutorial and go for it. By the end of that tute you'll have a reasonably comprehensive set of mechanics and it explains what it's doing every step of the way. Well done Jotaf, where-ever you are (presumably ascended).
Coding it is where the idea really begins to solidify - no idea will survive exactly the same once you start implementing it.
-
@Gr3yling - thanks for the link! I'm following it now, and hopefully I don't get stuck. X)
@CrowdedTrousers - yeah, I'm going for it. :)
-
So, I'm having some problems starting my roguelike. I am trying to follow the "Complete Roguelike Tutorial, using python+libtcod" on roguebasin, link here:
http://roguebasin.roguelikedevelopment.org/index.php?title=Complete_Roguelike_Tutorial,_using_python%2Blibtcod
First, I downloaded python 2.7 and libtcod 1.5.1. Then, I made a new folder for my project containing several files from libtcod just like it said. I then went in the python folder, opened python, and tried to type in the first command that the tutorial instructed me to: rename libtcodpy as libtcod. And then python told me that there was no module named libtcodpy.
Any suggestions? I'm guessing maybe I need to move more files around, but I don't want to experiment and end up messing something up.
-
That's weird. So, if I understand correctly: Your own script is sitting in the same directory as libtcodpy.py, and contains the line: "import libtcodpy as libtcod". Then, when you try to execute your script you get an error like:
Traceback (most recent call last):
File "foo.py", line 1, in <module>
import libtcodpy as libtcod
ImportError: No module named libtcodpy
If that's the case, I've no clue what may be causing it. Just to be sure, I tried to follow the first few steps of the tutorial, and here is what I get (on Linux, using ls and cat to display the contents of the folder and the file respectively):
simen@coupdepoudre:~/data/rl/libtcod-1.5.2/test$ ls
bla.py libtcodpy.py libtcod.so
simen@coupdepoudre:~/data/rl/libtcod-1.5.2/test$ cat bla.py
import libtcodpy as libtcod
print "Hello world."
simen@coupdepoudre:~/data/rl/libtcod-1.5.2/test$ python bla.py
Hello world.
simen@coupdepoudre:~/data/rl/libtcod-1.5.2/test$
As always,
Minotauros
-
Greyling didn't run the script in the folder containing libtcodpy.py, but rather started the python executable in the python executable directory (where libtcodpy.py can't be found).
-
Greyling didn't run the script in the folder containing libtcodpy.py, but rather started the python executable in the python executable directory (where libtcodpy.py can't be found).
So, I'm still confused. What should I have done?
I am still working on learning the python language in the tutorial that I mentioned earlier, but it's not going to cover things like this, is it? Is there a resource that has beginner level information about this kind of stuff? Like, what scripts are, and how to use them, that sort of thing?
-
So the tutorial is set up for creating a projects folder in which you create a new file called myroguelike.py (or similar). Then in this folder you also add the following files from the libtcod directory, (direct snippet from the RL tute)
Now to create your project's folder. Create an empty file with a name of your choice, like firstrl.py. The easiest way to use libtcod is to copy the following files to your project's folder:
libtcodpy.py
libtcod-mingw.dll on Windows, libtcod.so on Linux
SDL.dll on Windows, SDLlib.so on Linux
A font from the fonts folder. We chose arial10x10.png.
I think the libtcodpy files are not visible from the CLI compiler you're typing that first command in. I think they could be if you copied those files into the Python27 directory, but they're not that useful there either.
So I suggest: create your project folder (under the python27 directory, mine is Python27\Projects\firstrl.py) copy those files and then start your .py file with the import statement. That works for me anyway!
As a side note the python CLI run-time is great for quickly testing (non-libtcodpy) commands to see what effect/value they produce.
Good luck
-
At the very end of the python tutorial I'm using there's a section on file input/output. I guess I just need to study it so that I can understand the types of things I'm asking questions about. Thanks for all your help, though.
-
You might have a look at this tutorial (http://docs.python.org/2/tutorial/), which seems to cover the basics pretty thoroughly.
As CrowderTrousers said, the CLI (that you get from just typing "python" (or what the executable may be called) in a terminal) is good for quick experiments. But you should save your script as a text file, which you can then execute with "python \path\to\file.py". If a copy of the libtcodpy module is located in the same directory as your script, it should be imported normally. Hope that helps.
As always,
Minotauros
-
You might have a look at this tutorial (http://docs.python.org/2/tutorial/), which seems to cover the basics pretty thoroughly.
As CrowderTrousers said, the CLI (that you get from just typing "python" (or what the executable may be called) in a terminal) is good for quick experiments. But you should save your script as a text file, which you can then execute with "python \path\to\file.py". If a copy of the libtcodpy module is located in the same directory as your script, it should be imported normally. Hope that helps.
As always,
Minotauros
Thanks again, AgingMinotaur. I'll check out the link, and I'll try to apply what you are saying.
-
I made some sprites. (it's the first time I tried spriting, these are the first sprites I've ever done)
And I'd like to know if they look nice for a roguelike. Also, I revamped the Cowboy a little, and have two versions of sand, but I just don't know what the hell to do with that sword... no matter what I do to it, it just looks bad... :s any ideas??
http://imgur.com/a/3lIVY
Edit: I may also revamp the Colt Revolver, it's handle looks too small. xd
-
They look alright :-) Me, I'd try to add a little padding to all the sprites. That way it wont look too messy when the sprites are placed next to each other. Also, if possible, keep a consistent angle for all items in the game. Depending on what backgrounds the sprites are going to be placed upon you might consider an outline.
-
Some quick tips:
Make sure you pay attention to color contrast. Most of the the cowboy dude's sprite is a dark, unsaturated brown where it's kind of hard to make out individual lines. Then you have his extremely bright, saturated skin that draws too much of the viewer's attention.
You can make objects appear larger or smaller through your color choices. That's why the cowboy dude's arms look thicker than his sleeves do.
Read about anti aliasing in pixel art so you have better control over shape and color.
Try to get as much mileage out of your colors as possible. If you can make something look good with three colors, don't add a fourth. Make each shade you use distinct from the rest.
The best advice for learning any visual art is to look at the world around around you. Look at the way light behaves and how light and shadow affect color. Use real objects as references when you draw rather than than trying to intuit or remember what things look like.
E:
Here, maybe this will help:
(http://i.imgur.com/1irQmJw.png)
I did a quick re-do of that sword you were working on. The main thing you should notice about it is how I used darker pixels to smooth out the edges of the blade and make it slowly taper off near the tip. This works because the darker pixels partially blend in with the black background, giving the appearance of a sort of half pixel. This particular kind of anti aliasing won't work as well against a brighter background because the darker pixels will stand out rather than blend in.
It's not perfect but maybe it'll give you a better idea of what to do.
-
@malignatus: Thanks for the tips, however, I'm not sure what padding a sprite is. Care to explain? ^^''
@Vanguard: Thank you for the tips too! Yeah, I noticed that in the cowboy dude, and was actually pondering a darker skin tone, just never got around to it. x.x'
I'd read about AA before, guess I'll go and see images and read about it again. :)
Yes, I actually do both of those last advices. Or at least, I try. xD
That sword looks good!! I think that type of AA you did will work for atleast the UI of the game, I'm planning for it to be a dark-ish brown, I'll try to apply that type of AA with the sword I did, let's see how it turns out. :)
Btw, if you're curious, here's the sword I'm trying to "replicate": http://www.sailorinsaddle.com/media/images/product/display_897_AMERICAN_HORSE_ARTILLERY_SERGEANTS_SWORD_CA_1812_633731221135625000.jpg (http://www.sailorinsaddle.com/media/images/product/display_897_AMERICAN_HORSE_ARTILLERY_SERGEANTS_SWORD_CA_1812_633731221135625000.jpg)
(or maybe you already knew this was the sword xD)
Also, I've revamped the cowboy with a darker skin tone:
(http://i.imgur.com/NRDdURY.png)
-
@malignatus: Thanks for the tips, however, I'm not sure what padding a sprite is. Care to explain? ^^''
To add some margins around the sprites.
(http://i.imgur.com/dO3sEIC.png)
-
@malignatus: Thanks for the tips, however, I'm not sure what padding a sprite is. Care to explain? ^^''
To add some margins around the sprites.
(http://i.imgur.com/dO3sEIC.png)
Oh, I see what it is now, thanks! :)
It does like less clustered, looks kinda nice.
-
Gosh, I love this horse.
-
Thanks, I'm glad you like it! ;D
Any suggestions and more tips are welcome. ^^
-
The sprites are looking good, AlexPT. Keep it up.
So, I'm still dutifully trying to learn python. One of the assignments in the tutorial I'm using is to program a battleship clone. The playing field in this version of the game is a series of lists.
So, this made me wonder, are the playing fields in most roguelikes also represented with lists? Just curious.
-
Roguelikes use a very wide variety of data structures to store map grids. A list of lists is a reasonable option as are:
- a flattened list with a function that links 2d positions to list indices
- a map / hash / dictionary indexed by coordinate pairs
All of these have different advantages and limitations, several variations exist.
-
The sprites are looking good, AlexPT. Keep it up.
Thank you, I'll try to keep it up. ;D
-
I finished the battleship clone. I don't want to brag, but it's probably the best game ever. It has everything:
Love!
Betrayal!
Battle!
Ships!
Battleships!
1 Race(s)!
1 Classes(s)!
A total of 1 race-class combination(s)!!!
Really the only thing left is to decide what to call it. I'm torn between "Rogue Battleship" and "BDOM" (that stands for Battleship Domains of Mystery, for those who were wondering).
The next version will, of course, add raptors as a mount type.
Feedback?
-
What about Rogue Battleships of the Domains of Mystery? ;D
This thread reminds me of the old Six-Gun Shootout. That was more of a tactical squad game (like Laser Squad, but without lasers) than a roguelike though. Maybe it would still be an idea to have a look at that for an inspiration.
If you are still interested on reading more about Python and Roguelikes at somepoint, there's an interesting series of blog posts at: http://therealkatie.net/blog/tags/pygame/. It's not libtcod based, but instead of uses Pygame. Basic concepts still should apply though.
-
So, I finally finished the online python tutorial I was working on. I'm sure that it would seem like very basic stuff to most of you guys, but I'm actually really proud of myself for getting through it. It was quite challenging to me. I'm hoping that now I can make some sense of the game making tutorial on roguebasin and start to create a very basic game.
Wish me luck.
-
So, I finally finished the online python tutorial I was working on. I'm sure that it would seem like very basic stuff to most of you guys, but I'm actually really proud of myself for getting through it. It was quite challenging to me. I'm hoping that now I can make some sense of the game making tutorial on roguebasin and start to create a very basic game.
Wish me luck.
Good luck Gr3yling. Great to hear that you finished the tutorial and are going to continue with the programming.
-
So, guess who has an @ sign moving around a blank background now?
(Hint: it is not akeley)
I just copied the code from the roguebasin tutorial, but I still feel like it was quite an accomplishment, considering I had no idea what scripts were until a day ago. I'm not even going to say how much I misunderstood what the tutorial was telling me to do to begin with, but lets just say I was pretty far off.
-
That's great. The start is always the hardest. Do you have plans what you'll try to tackle next? Like walls or items?
-
Right now I'm just trying to go through the roguebasin tutorial and comprehend the code. I more or less understand what is going on up to the part on dungeon generation, so I'm maybe 10 to 20 percent of the way though the tutorial.
I think I'm a long way from being able to implement any features without the tutorial holding my hand, though.
-
I don't know, or maybe I'm just not meant for it?
There's no such thing as not being meant to learn programming, it's just a matter of finding the method that works best for you.
I personally have always felt the need of someone teaching me, instead of just following a tutorial. It helps set the pace and keep myself motivated.
Since having an in-person training is not always an option, maybe online courses could be a good alternative for you. If you like this approach, I recommend looking for Python courses here: https://classpert.com/python-programming (https://classpert.com/python-programming)
Good luck on your studies, and don't give up!