Messages

301

Programming / Re: Preferred/ideal development environment/language for roguelikes?

« on: April 18, 2012, 08:38:04 PM »

No such thing exists. There are good and bad features in all languages. You want everything, but it's rarely possible.

Gee, Thanks.

In my Opinion the best Language for your first(I think?) roguelike would just be the one you are most comfortable with.

I don't have comfort issues. I just want the better tool for the job.

You might not be 100% satisfied with Java, but here's something you should know: It is an absolute pain to write a program in a language you aren't 100% familiar with. This applies even more for a roguelike. If Java is the language you know then go for it!

I disagree. I had to develop a web-published game a while ago- I wanted to do it as a Java applet or webStart so that I could take advantage of a wide variety of libraries I was familiar with and runtime performance increases. My boss refused and insisted on flash, which I had no prior experience with. I think it took about week to learn and another week to master- all in all the development time was very reasonable despite having to learn something completely new. If I'm going to embark on a serious project or series of projects that are similar, I want a toolbox that is specialized for that particular task. I already know how to program games.

I haven't met a scenario yet where I had to abandon JavaScript because it got too complex.

Very interesting! I would never be waste my time doing anything remotely Dwarf Fortress-like, but I would like to be able to manage a lot of off-screen updates. How would pure JavaScript perform in that regard?

Otherwise HaXe looks sexy. I considered using it for flash development once a long time ago, it has matured a LOT since then.

IIRC even Apple supports java apps.

Moot- I could care less about apple.


Thanks guys- more advice is always welcome.

302

Programming / Re: Working on new Roguelike (Sci-fi setting?)

« on: April 18, 2012, 05:18:32 PM »

Thanks again for the great ideas all.  Really helping me flesh out what I want to do.  I made alot of progress today and have Movement, LoS and basic pathfinding stuff setup.  Tomorrow I figure out the style of levels I want to generate, what do you guys think on those.  Open spaces?  Lotsa Corridors?  Generate it room based so its more like a building instead of random cave (leaning this way for scifi).  Maybe I should save level generating for last so I can build the levels around how the gameplay mechanics end up working out?  (Wide open vs tight corridors etc).

Thanks again for the brainstorming!

I think you should have buildings, not caves. Unless it's a fallen-sci-fi-type-thing-where-they-got-kicked-out-of-the-city-and-have-to-survive-with-minimal-technology kind of thing

Or crashed on an alien planet.

303

Programming / Preferred/ideal development environment/language for roguelikes?

« on: April 18, 2012, 10:22:31 AM »

I've been meaning to get started on my roguelike project, but the more I prepare to write the first line of code, the more I feel like my traditional Java-based setup may not be ideal for me. I love Java, but I'm going to want an approach more optimized for prototyping that is also less verbose.

I want to have a wide array of deployment options (I would sacrifice this though)-
Compile to mobile device, compile to binary, publish on web, or distribute via source without ridiculous compilation instructions (Obviously, development will have to be considerate of the restrictions for each platform). If I can publish to web the other options aren't really important.

I willingly sacrifice control for productivity-- I don't really want to deal with pointers or memory management. I like code-completion and every form of spoiling that Java developers are used to.

I want concise or ultra-readable syntax. I want to shorten the number of times I have to push buttons.

I need the option to run CPU-intensive simulations.

I want to be able to prototype quickly- either making changes while the program is running or quick compiling. An interpreted language with the capabilities of compiling to native would be cool.

I want my environment to be easy to set up or come mostly already set up. Nothing obscure.

I'd like my deliverable to be lightweight. If it must include a VM it should be tiny.

I don't care about preexisting libraries- basic data structure management is all I really want.


I know QBasic, C/C++, Java, AS3, LISP, and some dumb scripting languages. I firmly believe that a programming language and development environment should do as much work for you as is possible. I want to think less and implement more. I've already done the whole 'compile linux from source' thing and am not in this for my personal education of programming concepts.

I've looked at Scala and I like it, but it's not great for prototyping because of its bad compile times. "Drink the koolaid" cults are also not very desirable.
I've looked at Python, but its obsession with dynamic features makes it a poor choice for CPU-intensive simulation.  This could be done with C/C++ extensions, but I easily get sucked into optimizations and want to avoid thinking too much.
There are some Python derivatives like RPython that looks interesting, but I haven't heard anything about them from another human-being.
Javascript/HTML5 has very good performance now a days, but I don't think it's a great choice if things get complicated.
I'd just use Flash, but the VM, on account of a sub 7-mb distribution size, is total shit.
I'd just use Java, but I don't want to require the JVM.

Anyhow-- what are y'all's experiences?

thanks.

304

Temple of the Roguelike / Re: Welcome to Temple of the Roguelike Forums! (a.k.a. Introduce Yourself Thread)

« on: April 18, 2012, 05:10:04 AM »

You essentially need to have your limbs/body parts replaced with higher quality materials.

So, I cut my legs off and replace them with metal legs so I can run faster?  Sounds interesting. 

Well, Kicking is the default unarmed attack. It can get VERY powerful, especially if you're wearing boots of kicking. The main reason why you trade-up body parts is for the HP. If it's an arm it also helps increase your ability to use gear. But again, the game gets harder when you get stronger. So you have to make those decisions strategically.

305

Temple of the Roguelike / Re: Welcome to Temple of the Roguelike Forums! (a.k.a. Introduce Yourself Thread)

« on: April 18, 2012, 04:13:04 AM »

I never introduced myself-

I've been a hobbyist programmer for 10 years. My focus is typically on implementing and developing concepts more so than publishable content- primarily out of personal interest. I've typically worked on 3-dimensional concepts in collision detection, but I want to put more thought into more complete game development projects. I'm working on a roguelike for a thesis project, but won't expect any time to work on it for a while.

My first and favorite RL is IVAN.

Welcome requerent.  Cool, IVAN? I've heard about it, but never really played it.  That's the one that has oh, say a bajillion stats for everything, right?

The game's difficulty ramps up depending upon the gear in your inventory. High level gear requires certain stats to be higher, which pretty much exclusively depends upon how often you use similar gear. If you just use what is reasonable, you can progress without thinking too much about the stats, but they are important.

What's more important are strategic decisions in regards to deities. It's a very god-driven game. You essentially need to have your limbs/body parts replaced with higher quality materials.

306

Temple of the Roguelike / Re: Welcome to Temple of the Roguelike Forums! (a.k.a. Introduce Yourself Thread)

« on: April 18, 2012, 03:30:26 AM »

I never introduced myself-

I've been a hobbyist programmer for 10 years. My focus is typically on implementing and developing concepts more so than publishable content- primarily out of personal interest. I've typically worked on 3-dimensional concepts in collision detection, but I want to put more thought into more complete game development projects. I'm working on a roguelike for a thesis project, but won't expect any time to work on it for a while.

My first and favorite RL is IVAN.

307

Programming / Re: Working on new Roguelike (Sci-fi setting?)

« on: April 18, 2012, 01:20:38 AM »

I would avoid character creation and leveling all together. Sci-fi lends itself to logistics and resource management more so than races and such. It may be interesting if you could find terminals to replace lost limbs with artificial ones or acquire permanent enhances to strength via genetic modifications.

The interesting thing about technology is that it makes men more equal. The rifle allows for a commoner to easily kill a warrior-- it's these types of principles that I think are lacking in Sci-Fi. Classist and Racist systems detract from the fact that Sci-Fi is interesting more so for how tools are utilized than what a person is capable of.

There are obvious skills, but I think a Brogue-ish approach of developing your character as you progress does a LOT to make a game more accessible.

Suppose players are autonomous teleological robots but have somehow lost their primary directive. Your system automatically defaults to questing for some lost piece of data or equipment. Perhaps you directives become decrypted from your memory banks as progress is made-- perhaps you can swap out discovered modules to give you skills that you wouldn't have normally.

It'd be interesting to make it entirely item/inventory dependent. Inventory restrictions determine how many classist-type of abilities the player has available to switch on and off. Effectively giving them a dynamic character. The tough questions revolve around what items to keep and which ones to give up. This way everything discovered is likely to be useful.

I would have skills involved with this, but I would make those skills only accessible via items/equipment. Maybe randomize a player's starting equipment so that they have to think about start-game strategy in a different way each time, but they then have the freedom to mold their character however they can. It might also be interesting to have 'blank' chips that they can use to drain properties from other artificial lifeforms to gain new cocktails of skills or to mix existing chips in a way that yields unexpected but consistent results (an alchemy cookbook).

Anyhow, just some thoughts.

308

Other Announcements / Re: a RL that requires skill?

« on: April 18, 2012, 12:51:05 AM »

It is in the OP; it is the whole point of the thread. Much of what he has posted is in support of that assertion. It’s in the damn thread title.

It seems like a double standard to criticize Ancient for a lack of diplomacy when the OP is basically asking a community of Roguelike developers and players if there are any Roguelikes that don’t suck.

Tell me you've never made a rant post.  Ancient's post was pretty ranty too.  Besides, OP has a legitimate point.  If you're playing a new (to you) game and you don't know all the spoilers, luck plays a main part in the game.  Ancient himself said that randomness was a pivotal point in the rogue-like genre.  

And maybe I wasn't clear, I'm pretty sure he's only said "RL's require dumb luck" insultingly only once.  The rest of the thread has been pretty level.  Just because he has a different opinion doesn't mean he's actively insulting you.

Come on, PB has been whining this whole time. He says dumb things and then reiterates them in a slightly more esoteric manner and continues to do so until people concede to what he's saying. It's both bad and unproductive reasoning (I know because I like using it  ).

I think the discussion is useful, because there exists opportunities for developers to consider some things they may not have before, but he's not even close to being on target.

Roguelikes are hard.
Roguelikes don't provide much in terms of positive feedback.

That pretty much sums up what PB is saying. His use of words 'luck' and 'skill' are just ways for his ego to conceal these amateur complaints that exist for any new player in a new genre of gaming.

His entire argument is admissable ONLY because there are some developers who would like to hook a larger population of gamers. And its only valid as a study of how a new player reacts. We should be thankful he's expressing his opinion in the way that he is- it provides important psychological data on how roguelikes can be designed to be more addictive. Anything he is actually saying, however, is pretty pointless.

309

Traditional Roguelikes (Turn Based) / Re: No PCG (!?) Deterministic 24HRL: MUTANT ARISTOCRATS!

« on: April 16, 2012, 07:54:06 PM »

I actually think I would take back the bit about it being both. It's about as much a puzzle game as DOOM is. Just because the levels are predetermined AND the game is deterministic doesn't make it a puzzler. It can be 'solved' via strategy and repetition like DOOM can, but that isn't really enough to call it a puzzle.

310

Traditional Roguelikes (Turn Based) / Re: No PCG (!?) Deterministic 24HRL: MUTANT ARISTOCRATS!

« on: April 16, 2012, 09:16:57 AM »

I voted yes and yes.

It's a roguelike because the outcome of arbitrary actions is too intangible to determine. Even if you repeat the same map, Fog of War makes it impossible to recognize the outcome of an action. In this way, there is a general set of rules that you discover and apply to any given situation; however, it isn't possible to determine whether or not the applied skillset is actually the solution to either a sub-problem or the complete problem.

It's a puzzle because you can solve it methodically. Is solitaire a puzzle game? I would say it is even though it's not always solvable and the board state is randomly generated. There's no reason why the inverse can't be a roguelike. A genre designation describes the experience that a player has when they play the game.

If a player plays it a second time they should immediately realize that it is deterministic, at which point they can begin thinking about the game as a puzzle. However, the solution to the puzzle is NOT going to be a memorable set of key-sequences, but a methodology. The refinement of a player's problem-solving method and the near-infinite possible solutions as well as the inability to determine the outcome of minor actions makes it a game that requires the same skillset as one would implement for any other roguelike.

The player can't have expectations that some set of actions will lead to a solution just because a particular set of actions is a solution to a subset of the problem. A solution to a subset can be a failure of the entire game, as it would be in Sudoku. However, the rules of this particular game suggest pretty clearly that progress in the game is more immediately discernable relative to the information available. In other words, we know when we're progressing despite the fact that we also know the unforeseen outcome of our actions is deterministic.



This makes it very much both a roguelike AND a puzzle game.

311

Other Announcements / Re: a RL that requires skill?

« on: April 16, 2012, 04:53:14 AM »

@Ancient,

The complaint that the people have here has more to do with learning curve than it does with 'skill.'

In a fighting game for example- before you can even think about strategy you need to understand the mechanics and principles governing gameplay. Once you've learned enough of the game, you can start employing strategy that favors your specific skillset. At that point, the game becomes an application of skill rather than (or in addition to) an accumulation of it. Once you reach the application phase, the game becomes cognitively interesting.

Punkbohemian is complaining about balance when he's still fumbling with the metaphorical controls.

I personally feel that this is a legitimate complaint, but he's completely wrong in calling it 'skill.' His complaint is on the game's accessibility.

312

Other Announcements / Re: A side-scrolling graphical Roguelike...

« on: April 15, 2012, 12:03:43 AM »

Though as a sidescroller it naturally lacks some of the typical tactical depth of a roguelike,

Woah, disagree. The type of 'board' has very little if anything to do with the tactical depth of a game. If anything, a sidescroller cuts out an 'unnecessary' dimension and requires the developer to focus on more important aspects of 'tactical depth.'

There are only two aspects of the board that is important in regards to gameplay, whether it is 2d or 1d- the distance from enemies and the features of the terrain. There are ancillary qualities like LOS, but that can be represented in 1d or 1.5d. There is nothing about 2 dimensions that make a game inherently more tactical.

If we consider a roguelike, what does the 2nd dimension of the board actually do for gameplay? It's more enjoyable to explore from a player's point of view, but it inherently does close to nothing for 'tactical depth.' If anything, it creates the artificial illusion of depth, but having that extra dimension of movement doesn't really do anything.

I create distinction between 1d and 2d which isn't immediately clear- I put a platformer in the 1.5d category-- Jumping, relative to its meaning in gameplay, serves as a 'portal' to another area of the level. If we separate a level in terms of what areas must be jumped to, then we have a perfectly one dimensional game. Roguelikes can similarly be reduced to a literal one dimension (moving to the right is a 1-bit right shift, moving down is an 8-bit right shift- you can even play chess in 1 dimension).

In this way, sidescrollers and grid-maps have no literal difference in terms of INHERENT complexity. If you want a tactically interesting game, design it to be interesting in the first dimension and transliterate it to the second. Games that start in the second may 'feel' more 'tactical,' but whether they are or not has nothing to do with the dimensionality of movement.

Fighting games, for example, are typically one-dimensional games. You can either move toward your opponent or away (jumping is a shift in gravity or position-- a stance, you don't actually move relative to the board), there isn't really anything else to consider- only the distance between you and your opponent. Fighting games are, also, incredibly tactical.

313

Other Announcements / Re: Liberated Pixel Cup Game-Making Contest in June/July

« on: April 14, 2012, 11:15:21 PM »

I have to say I was more interested in this before I saw there was a prescribed art style.

Constraints are both useful and necessary in the creation of art. I imagine that they may run future competitions with different style guides. IMO I think it's brilliant. A variety of artwork submitted by a variety of artists with the expectation that they'll all potentially be useful together. It's pretty ambitious and a style guide is totally necessary.

My only complaint is how do we ensure the appropriate quantity of animations for character assets? The diversity of assets will determine how much freedom coders have to present gameplay principles. It may not cause problems though- it just depends.

314

Off-topic (Locked) / Re: Jo's Off Topic Rant

« on: April 14, 2012, 08:46:54 PM »

Holy shit- I cried.


If you have problems primarily with infections, then there may be a nutritional course of action that can be helpful. One problem with being born premature is an opportunity to fill your gut with the good bacteria necessary to supplement your immune system. I think... at any given point in time the average human has about 10-100 (it may be more like 10,000, I can't remember) good bacteria for every human cell on their body (we are mostly foreign critters O_O). Antibiotics are NOT trivial because they kill both. Necessary if an infection occurs, but afterwards it's necessary to get those good bacteria back because they help ensure that digestive resources get used effectively (necessary for proper immune function).


Whatever you do, don't go vegan or vegetarian. They aren't good ideas for people with health problems. As far as nutrition is concerned, you want easy energy that can't be absorbed by bad bacteria. That means eliminating sugar as much as possible. Not just 'sugar,' but naturally occuring mono and disaccharides (IE lactose from milk). If you can't eliminate, try to keep it to under 1 gram per serving. What you want are these things call oligosaccharides- they are short chain starches or complex sugars that will ferment in your gut, stabilizing your GI tract's Ph and feeding good bacteria as it gets broken down into simple sugar. Our primary source for the oligosaccharides necessary for building gut flora is in mother's milk- after that we don't find large quantities of it in average diets. Asparagus, garlic and Soybeans are good sources of the three types.

http://en.wikipedia.org/wiki/Inulin#Natural_sources_of_inulin
http://en.wikipedia.org/wiki/Oligosaccharide#Sources

You also need good bacteria. The easiest way to do this is by making your own Kefir. You just need a 16oz jar, kefir grains (you can get 'em freeze-dried online) and milk. The Kefir grains are yeast/bacteria cultures that eat the lactose. Just let them ferment at room temperature for 24 hours in a glass jar with milk. strain out the grains and drink the liquid. Reconstitute the same grains with milk and repeat. It's very easy and in 2 months you'll probably no longer need to use toilet paper (in a good way). I like recommending kefir because it's incredibly inexpensive and it's an easy solution to keep up with. If you did only one thing, this would probably help quite a bit. Kefir will still have lactose in it (contradicting the point above), but it's going down with lots of good bacteria.

This obnoxiously awful website pretty much started the 'Kefir revolution'- it's a good resource but there may be more succinct sources elsewhere http://users.sa.chariot.net.au/~dna/kefirpage.html.

Otherwise, try to stick to pickled and fermented foods. Truly picked foods, in the ingredients list, will have only salt, water, and the food being pickled (plus spices)- it won't have vinegar or any kind of sugar. On the nutritional facts, pickled food should also have a very low or nonexistent sugar content (less than one gram). Sauerkraut and pickles are the easiest things to find-- but if you feel like this sort of thing is helpful to you, making pickles is incredibly inexpensive. Salt, water, jars (you can reuse jars bought from the store- they all have pickling lids that will repop from the fermentation gases), and stuff to ferment- plus the internet for some guides.

Stick to sourdough for bread- don't need to shy too far away from beer, undistilled fermented beverages help fight infection (stick to original bottle fermentation)- if your liver can handle them in small quantities. Also highly recommend supplementing with Zinc- it will help your liver and other aspects of the immune system.

Mmm... Oh. Recommend high quality (care not cut) beef. A chuck roast, when prepared properly, will run you about 4 dollars a pound and taste like rib-eye. Bivalves are also good.

315

Other Announcements / Re: a RL that requires skill?

« on: April 14, 2012, 08:36:24 AM »

I've been keeping up with this thread but haven't really had time to chime in.

I opened a thread in development over here-> http://roguetemple.com/forums/index.php?topic=2220.0 that tries to quantify a similar complaint into a problem that can be explicitly addressed. I haven't had a chance to reply to that thread either, but I think there is some decent thoughts tossed around on the matter.


I know of a few instances in academia where developers create an AI that handles randomly generated levels. The AI first generates the level according to a certain criteria of expected difficulty and then quickly playtests the level several times at varying degrees of 'skill.' In this way, the AI can determine exactly where the level is most difficult and make fine adjustments. Needless to say, indie developers don't really have time to take this sort of approach unless the rules of their game are fundamentally simple (say, to the point that they could be reduced to a CSP- idea? Yes!). We want greedy or simple algorithms that can give us reasonable expectations of difficulty. Which is why so many Roguelikes use depth-based difficulty.

If the difficulty is determined by depth, then our ability to descend depends upon what equipment/bonuses/enablers that are randomly generated for us. The more we play the game, the more we realize what items are important for when we approach certain depths. Your utilization of the equipment and the terrain may increase your chances, but you might simply never received enough goodies to succeed. Even if no aspect of combat is random, you're success still depends upon the tools that the RNG makes available. The game becomes less a matter of skill and more a matter of knowledge- kind of like Chess.

If a PGC considers the character's current strength when generating challenges and equipment, then difficulty can be regulated in a more reasonable way. Doing that, however, doesn't necessarily make the game more interesting or skillful-- but it will both prevent inescapable failures and expose how terrible or boring a 'system' actually is.


What set of criteria satisfies the skill-minded player?