next »

[sigfarter]

Hi guys,

Alongside my other project Cyborg War, I have been working on a roguelike that makes use of some of the same technologies I have been developing for the former.

All of my games use similar code bases I have written entirely by myself. These consist of:

Brain Damage API (BDAPI) - Rudimentary codebase written in ISO C99, supporting all of the inbuilt data types, string and data handling functionality and cross-platform coding synchronization.

Chaosforge - Random and procedural generation library, supporting a wide variety of common and uncommon implementations of randomness, such as random string parsing, visual and auditory noise generation with support for a number of different noise generation techniques, including white, pink, brown and perlin noise. High level functionality includes map generation, real-time and reactive music and sound generation and computer-generated storytelling. Not related to another Chaosforge.

Cranium - Cross-platform game library supporting TTY, console, Windows 95+, Macintosh, Linux, BSD and Solaris. Provides error, logging, window, keyboard, mouse, joystick, video, audio and 2d and 3d graphics handling amongst other things. Similar to libraries such as SDL, SFML and HGE, except entirely scratch-built and not based on any of the aformentioned. For roguelike and text-based games, Cranium offers support for traditional consoles (boo, no more fullscreen consoles in Windows Vista or 7), emulated consoles and tile-based graphics.

The vast majority of my time spent over the past year or so since I began working on the Brain Damage API and its associated technologies has been spent Cranium. I've been pushing 100+ hours of development a week consistently for over a year to get this far; my commitment is primarily based on my desire to be an independent developer in every sense of the word; I work with no one but myself.

That being said...

Bloodshed is a roguelike game I am developing. I've been planning on making a rogulike game ever since I first played NetHack nearly 8 years ago. Bloodshed is basically my own take on the roguelike genre after playing dozens of roguelikes over the years and differs in many aspects from them. My intent isn't mearly to make a roguelike game because I am a fan of the genre, but to make a compelling, story-based adventure game that carries that player through a long-term, thought-provoking journey that remains varied and surprising even after years of play.

Differences from other roguelikes are:
- Not based on any established fantasy worlds. An entirely new universe without elves, orcs or dragons.
- A completely new system of game mechanics that on the surface appears similar to the D20 ruleset merely for the purposes of familiarity but functions entirely different.
- Detailed cause and effect interpretation and notification. For instance, combat is not based on 2 or 3 rolls of a die, but a complex system of multiple conditional tests and finely-tweaked randomness. Bloodshed won't tell you that you simply "missed the monster" on an attack, but elaborate on the very reason that you missed in a narrative fashion that might be as minutely sensitive as the reflection of a bright light that reflected off of your sword or a twitch in your wrist.
- Permadeath exists only in circumstances that the player decides to set at the beginning of a new game. Bloodshed is set in a universe where no afterlife exists and its sentient inhabitants are forced to reincarnate after death for all eternity. Upon death, the player character is born into a new body and continues the game from there, unless they have no soul to begin with and the game simply ends. Residually recalled memories allow player characters to remember information based on their past lives and this ability is elaborately expanded upon in the game world. Your soul might be newly born or literally hundreds of thousands of lives old, and this has an effect on yor player character. Your character is effectively a hybrid of two components: the permenant soul and its personality and characterists and the temporary body, which is more or less a traditional RPG character minus the morality and some of the psychological stuff.
- True sandbox gameplay. There is no scoring (ignore that in the screenshots) or goal to accomplish set by the designer (me).
- The world is not generated randomly everytime. Well, it partially is; it is generated based on a constant seed, so you get the same world everytime you play. However, your choices, actions and external factors have a profound effect on the Bloodshed world and it shaped accordingly. We aren't just talking about the little things, but the fate of entire civilizations and solar systems. Death is not necessarilly to be feared in Bloodshed like it is in other roguelikes; since Bloodshed encourages long-term play, every game will be completely different, even though they start entirely the same. A lot of these gameplay mechanics share similarities with the tenants of chaos theory and the butterfly effect. The only difference that effects the initial condition at the start of a new game is who you choose to be.
- Insanity. The toll of endless rebirth has its effect on the spirit, and the older and more mature souls that inhabit the Bloodshed universe exhibit symptoms of profound insanities that can cause predictable or unpredictable things to happen to you in your journey...
- Lack of a morality system. I have always been firmly against systems of moralies in, hell, everything, including games, fiction and real life. You are free to interpret the unfolding events of the game as you see fit.

So basically Bloodshed's primary inspirations at both the design and player levels are fed by my musings on existentialism, nihilism and misanthropy.

Here's some shots of what I've got working right now. I'm working on the graphical tiles right now, but this is what is there:

Console mode:




Graphical tiles mode:




I'll keep posting more when I get stuff finished.

Until then, C+C would be encouraged.

I wish you great power and influence.

[Netsu]

I don't play rougelikes, but your description is interesting, seems like this could be quite a deep game. Looking forward to a playable build.

[wilbefast]

a compelling, story-based adventure game that carries that player through a long-term, thought-provoking journey that remains varied and surprising even after years of play.
(...) True sandbox gameplay. There is no scoring (ignore that in the screenshots) or goal to accomplish set by the designer (me) (...) your choices, actions and external factors have a profound effect on the Bloodshed world and it shaped accordingly. We aren't just talking about the little things, but the fate of entire civilizations and solar systems.

Intriguing. Tell me more

Lack of a morality system. I have always been firmly against systems of moralies in, hell, everything, including games, fiction and real life. You are free to interpret the unfolding events of the game as you see fit.

I second this emotion

[hareball]

Sounds interesting, I'm looking forward to seeing more of this. Is this going to be primarily keyboard input?

[Dustin Smith]

This sounds utterly awesome. You have a clear sense of vision, interesting ideas, and if I believe your 100+ hour/week claims (I do) you are certainly dedicated to your craft. Best of luck with this man, I have my eye on you.

[AlexanderOcias]

I'd kill for a good roguelike with graphical tiles, definitely will be keeping an eye on this one, it sounds great!

From what there is to comment on in the screenshots so far:
1. Some bigger margins and more leading would go a long way to getting people to read all that text. With that font, you only want about 6 words per line - I can't get past the second line.

2. I understand HP/MP, but why are all the stats in different colours? Is there a reason to draw the player's attention to there, or to create a link to things that appear in descriptions else where? or something else?

As far as the description of the gameplay goes, the only thing I'd question is the complex narration of combat - doesn't combat in roguelikes eventually become a very automatic process for players? It would be a shame if you put in all that work and then these descriptions were just ignored by everyone. :S

[RotateMe]

<detailed messages>

Keep in mind that (at least I think) there will be many many fights and eventually the player just wants to know "hit or miss?". If the result is the same, after some time, different messages become irrelevant and even annoying (after you read all of them anyway). Instead of more messages add more things that can actually happen when missing (like losing your weapon, getting blind a few rounds (because of said flesh of light), stumbling one field in some direction) and add short messages for those. Of course, this is just an opinion, but I really never saw how long, frequent messages add to the game.

<explains semi-perma death system>

I like that idea a lot! This way you can exactly define what will be kept and what not. A permanent storage for items is a feature I also like a lot. Think Shiren the Wanderer (I can't think of any traditional roguelike that has that).

<sandbox gameplay and world generation / manipulation>

That's a high goal, but sadly one which I have read a LOT of times on the rgrd newsgroup (by the way, if you haven't posted there, do so, those guys are SUPER helpful!) and ALL of the times the game didn't get beyond a few releases. In practice it's very hard to keep some action going on if there's very few to nothing predefined in the world. For example Dwarf Fortress' adventure mode is mostly about doing the same things over and over with some background that gets no real attention by most players.

Just a thought on that: Make an overworld map in the size of ADOMs, add some towns and dungeons in a set pattern, do some standard quests (hunt down amount x of y, find item i, etc.), focus on fun game and dungeon features THEN add the whole character interaction thing.

Another problem with a random world is, is that it gets hard for people to talk about it, and roguelike players DO like to talk (or rather write) about their game experience.

<Screenshots>

In my opinion the interface looks great. The text, while not well readable, makes a great title screen. Some person above me talked about graphical tiles, but I didn't really got that from your decsription, so what's it going to be?

Some last words in this way too long post: I've been developing a roguelike before, I've spend some years on it, and one thing I've learned is that it's important to keep the fun features in focus, the ones you're having fun developing and playing. Playtesting is a great part of the development process (even if you have beta-testers later on) so make your game enjoyable for yourself to play. Keep posting progress (I would move this post into the Creative/Devlog section btw) so you have the feeling that you owe something to the few people who wrote something about your game, sometimes this is the only thing that keeps you going. Finally, think a LOT about your code design, as after some time, for roguelikes, it always becomes an issue if you didn't (but reading about your codebase I have no worries about that).

Edit: got the graphical tiles thing now, 'twas a misunderstanding I like the graphical tiles a lot! Have you considered using libtcod btw? Has some neat features too!

[sigfarter]

Intriguing. Tell me more

The story telling aspect is primarily handled by the inclusion of multiple civilizations, characters and other organizations. All of these entities have unique agendas in the game world and their clashing interests (sometimes with the player) evolve differently in each game. Obviously, the player can have a direct influence on this evolution of narrative.

You may be familiar with chaos theory's butterfly effect. Slight variation in the initial conditions can have a massive impact on the outcome in the long term, so the player's decisions might hvae as profound an effect as the destruction of an entire civilization which is handled in a manner that I as the game designer have not decided on.

A lot of games that boast true decision making typically have a number of generic paths the player can choose from, such as the "evil" or "good" endings in a story. As well as not necessarilly having an ending, there are no set paths to be taken in the story as the evolution of narrative is processed by the game itself in ways I can't even begin to anticipate or comprehend.

[sigfarter]

1. Some bigger margins and more leading would go a long way to getting people to read all that text. With that font, you only want about 6 words per line - I can't get past the second line.

An astute and fair observation. You were the first to state this after I made the game announcement, but I see a number of other people that feel the same way. I'm going to make it one of my priorities before making any releases to make the interface much more modifiable to the user's wishes.

2. I understand HP/MP, but why are all the stats in different colours? Is there a reason to draw the player's attention to there, or to create a link to things that appear in descriptions else where? or something else?

The intent is that the player will be able to find a particular piece of information on the screen quickly. I can add an option to remove the coloring if you wish.

As far as the description of the gameplay goes, the only thing I'd question is the complex narration of combat - doesn't combat in roguelikes eventually become a very automatic process for players? It would be a shame if you put in all that work and then these descriptions were just ignored by everyone. :S

Quite right. Most roguelikes handle combat by simply making the player run into the monster. My intention is to make a much more sophisticated system where a player becomes "locked" into person to person combat with one particular entity at a time. On top of making encounters with more than a single enemy at once much more challenging and dangerous, it allows me to temporarily remap the keyboard into a "combat mode" where you have many, many more options to choose from.

The combat system is designed based around several real-world martial arts theories and techniques. Take for example Wing Chun, which places a great emphasis on the "center line" that maintains balance in a combatant. The belief is that sufficiently bending or distorting someone's center line leaves them vulnerable to being knocked off balance or over completely.

I am not directly basing the combat system on real systems of martial arts, but it is techniques like these that inspire my creation of a versatile roguelike combat system.

[sigfarter]

That's a high goal, but sadly one which I have read a LOT of times on the rgrd newsgroup (by the way, if you haven't posted there, do so, those guys are SUPER helpful!) and ALL of the times the game didn't get beyond a few releases. In practice it's very hard to keep some action going on if there's very few to nothing predefined in the world. For example Dwarf Fortress' adventure mode is mostly about doing the same things over and over with some background that gets no real attention by most players.

I suppose I'll just have to think and work harder than the cynics.

Edit: got the graphical tiles thing now, 'twas a misunderstanding I like the graphical tiles a lot! Have you considered using libtcod btw? Has some neat features too!

For personal reasons, I have a principle never to use anyone else's code if I can write it myself in a realistic time frame. At the moment, the only external libraries I make use of are OpenGL and OpenAL. I don't want to write a completely new alternative to DirectX or something like that; things like that take hundreds or thousands of active developers to create and maintain and I don't have the time for that.

[Dustin Smith]

For the most part I'm a genre newbie (I've only dropped little over a hundred hours in RLs), but if you ask Greg Costikyan nice enough he'd probably give you some insightful advice -- at least as far as game mechanics are concerned. He's got literally dozens of years of experience designing games and he's an avid Nethack player.

[email protected]

[sigfarter]

What roguelikes has he worked on?

[ஒழுக்கின்மை (Paul Eres)]

he's mainly designed pen and paper rpgs, but i would expect some of the skills required cross over

http://en.wikipedia.org/wiki/Greg_Costikyan

[sigfarter]

Looks like he charges for consultation. Might not be too good of an idea, unless he's interested...

Sounds like a dream job; get paid to give people advice on how to make games.

[Dustin Smith]

With this byzantine labryinth of game mechanics you have going on here he might sympathize. *cue shrug*

[sigfarter]

He seems like a pretty interesting guy and probably deserves more recognition than what he has, but I don't have a reason to ask for his advice yet.

I am open to receiving feedback in this thread, however. I also have a policy of implementing features in my games requested by a single person if there are no major conflicts of interest; the least I can do is provide optional functionality for the requested feature.

I figure if one person wants something and takes his time to address it to me personally, then there are probably a thousand other people who would appreciate its inclusion.

[Netsu]

I am open to receiving feedback in this thread, however. I also have a policy of implementing features in my games requested by a single person if there are no major conflicts of interest; the least I can do is provide optional functionality for the requested feature.

I figure if one person wants something and takes his time to address it to me personally, then there are probably a thousand other people who would appreciate its inclusion.

Then maybe I'll let you know, that I am awfully repelled by most hit point systems in both pen and paper and computer RPG systems. The amount of HP usually goes up with each level, sometimes characters can differ in HP drastically, one can have 20 and another 350.
How does this make any sense? I understand that with each level you learn new abillities and maybe train your body, this is how gaining experience works in my opinion. But getting more and more immune to death?
The systems I always liked the most was those that had a fixed amount of HP (varying between characters of different constitution, but no so much that one character has 10 times more HP than another) or had no HP system at all, and instead based on some sort of a wound system.
Hope you are of the same opinion.

[sigfarter]

stuff

I agree with this. To be honest, I haven't even settled on using hit points (or mana points) yet. If my combat system turns out to work as good as I think it will be, I can probably do away with hit points entirely and replace it with an injury based system.

I like how Dwarf Fortress handles injuries to specific body parts, but things like this are quite pointless unless you can actually target specific body parts (at the very minimum).]

My ideal combat and hit point systems would be where combat is determined by your skill as a fighter and your bodily attributes. That way you don't have to hammer a guy in the face ten thousands times until he dies.

The same criticism can be aimed at games like Stalker where you can shoot a guy in the head with a sniper rifle several times and he doesn't die. We need to do away with things like this because they are thin-skinned cover ups for laziness.

[sigfarter]

On another note, I do think that players should still have ways to increase the amount of damage they can take, whether by magic, exercise, berserk trance, etc. I do not think that you should be harder to kill simply because you've been playing the game longer than the other guy.

Are there even any RPG games out there that do away with player levels? I know some don't tolerate hit points.

EDIT: zelda lol

[Netsu]

On another note, I do think that players should still have ways to increase the amount of damage they can take, whether by magic, exercise, berserk trance, etc. I do not think that you should be harder to kill simply because you've been playing the game longer than the other guy.

In the systems I like the most, you must resist pain from your wounds to be able to keep fightning (otherwise you are given drastic penalties to all your rolls) and you can wear armor to reduce the seriousness of wounds you recieve (for example a heavy-wound-inflicting slice stopped by an armor of thick skin is reduced to a medium wound). In addition to this, the amount of wounds you can take before you die/pass out can depend on your stature (for example 2 heavy wounds for a skinny/little guy, 3 for a typical male and 4 for some kind of a giant strongman).

So yeah, you can survive longer by wearing armor, by being a tough badass, by being a berserker or whatever (you'll be obviosuly completely resistant to pain in such state) by bing plain fucking huge or by some kind of magic.