• entries
8
20
• views
8301

Updated w/ plenty of visual joy.

## Project #2: Elemental. Demo + screens of CO.

So I have two projects in progress currently - my primary long-term project is still Cloud Ocean (for which I have created a day/night cycle and a menu); however, my new short-term project (scheduled for release in August) is Elemental.

Elemental is unique due to its gameplay - it's sort of a puzzle-platform game utilizing mostly mouse control.

Each level, you start with a set of Elementals - Earth, Air, Fire, Light, Dark, and Water. Your goal is to get as many as possible to the level's end. You do this by combining Elementals to create devices to help you on your way... for instance: Combining Fire + Water Elementals creates a Steam Jet, a directed flow of hot air which boosts your Elementals up cliffs etc. Or you can combine Earth and Light to create the Clay Globe, a giant ball perfect for smashing walls - if properly directed. You don't get spent Elementals back - it's up to you to find the most efficient path through each level. The wide variety of potential gameplay elements, in my opinion, means a very engaging game.

Oh, and there's a demo. It contains the four tutorial levels - all I have done for now. Download here.

In CO news, I've added a day/night cycle, adding sunrises and sunsets, as well as a sun-distortion effect around the horizon:

The height of the horizon and the radioactive water will be fixed, I promise. Also, I intend to edit the colors quite a bit, "until I'm satisfied." Right now it's way too yellow and not enough blue...

Also, I have a near-final menu design:

## Issues + Green

The more you use an engine, the more you notice its deficiencies - for instance, the 3d engine I'm using for Cloud Ocean doesn't render billboards (for particle effects) that well - 100 billboards cost about 30 frames per second to render. I'm not sure what to do to get around this right now; but I'm working on it.

Additionally, this engine seems unable to load animations from any file formats but SMD and md2 - the former is the one format it seems Blender can't export to, and the latter is prohibitively inefficient. What's more, the engine seems unable to handle some aspect of my animations - I suspect bone weights - and slows down immensely - 30 frames per second loss for a 4000-polygon model. (My 20000 polygon city model, by comparison, only costs ~5 frames per second)

Both these issues will need to be resolved before I hit release; I remain hopeful in both cases.

In more positive news, work on CO's interface and engine proceeds apace. The trade and repair windows are in working order; remaining to be done are the upgrade window, the diplomacy window, and the location transferral system. My overall goal is to complete the engine, in its entirety, by the end of this year. Then there'll be another year or so to add story and content, and...

I wrote up a simple design sheet for a sort of "cooperative minorly-multiplayer survival adventure RPG" (what a mouthful) that I wish some company would decide to make. The document follows (Beware of casual writing!):

"Green": Design Document
* Procedurally-generated world
* Minorly multiplayer - only 2 to 16 (undecided) players per game
* Casual - players don't need to spend much time at a time on the game. Players can be on alone or together - but with seven players a server, there will normally be at least one other player on.
* Task-based - aside from the starting 2-3 hour spurt, players will spend their time in the game dealing with and recuperating from challenging "world events." Things like floods (save your belongings by moving them to higher ground; don't be washed away), illness (one or more players unable to function and must be tended), or the introduction of new elements to the game world (rampaging rhinoceri!)
* Taskbar notifications? Perhaps the game should indicate to the player when a disaster is occurring...
* Very long advance trees allow for hours of "vanilla" gameplay time - gathering materials, including entirely new ones as they appear; creating tools (with a minigame?); exploring the world (including new areas as they form), building houses, repairing things, etc.
* Social interactions (Illness and times of stress should form interpersonal bonds quite well - or cause negative situations); and the dynamic of surviving as a group in an increasingly hostile world.
* Dynamic, changing world. Time passes whether you're in the game or not; and there's always a very real sense of danger and impending loss. This is supplemented by the fact that you can actually lose in the game, causing you to have to restart from scratch...
* Personal characters that change with time. Customize hair, eyes, and body, then watch as you change over time - eat too much, and you'll be able to survive a famine etc. - but you'll get fat and slow. Too little, and you'll go gaunt. Exercise, and you'll be able to run longer distances... etc.

Story: You and a number of other people suddenly find yourselves transported from the modern world to a fresh, pristine paradise. At first, the land around you is completely flat and empty, but as you watch (in the introduction), plants spring up around you and you are suddenly in a lush terrain. That's it. Into the game you go...
Your stomach reminds you that you need food. Looking around, you find that none of the plants nearby are familiar to you - but none nearby are poisonous, so you may eat them. Having filled your stomach, you search for water. Then it's time to set up camp - the game walks you through creating basic structures.
Finally, you are given a list of tasks to perform - things like searching for more food for later days, exploring the area, building new and better huts, etc.

A blinking taskbar icon reminds you of your character's status when you're not on - your character will stay around the camp, eating and drinking as need be - unless he's ill, in which case he can't do much of anything and the other players must take care of him. (bring food and water to his hut, protect him)

If a player is inactive for too long (say, a week), the other players can hold a vote to replace him - at which time a "jump in" opportunity becomes available for multiplayer players... if all the players are inactive for too long, the game ends due to group starvation...

There is a certain element of trust involved in gameplay. Players have a chance to kill other players - it is even possible for groups of players to split up into factions and conduct miniature wars with spears, bows, and clubs. Of course, such possibilities are quite unlikely - but if there is a major conflict...

Game themes: Competition and cooperation. Game length: Lengthy. Maybe a week, maybe a month.

Specials: Hidden themes in the world. Ancient ruins with no explanation; strange artifacts... there should be a rich backstory present that can't be uncovered in a single playthrough.

...The ruins and artifacts are the leavings of an alien civilization - strange palaces and devices may be uncovered which have either positive or negative effects on the players... if you choose to delve deeper, you can uncover ancient ruins in an alien language... which can be translated to reveal the backstory of the game.

## A pictolog again...

...lots of pictures of Cloud Ocean:

1. The textured top of CO's first pillar. The city in the distance is a temporary model; updated version follows.

2. Corealis' complete model. (texturing to be redone) Also the basic idea of the interface. The thing most lacking in this image is shadowing; honestly the engine I'm using doesn't have any reliable way to implement it.

3. Night. Also, the second pillar of the CO world.

4. An indoor scene. Showing the model from a few posts ago, now textured properly. (although not thematically adjusted for the CO world) The room itself is generated procedurally, but a little sparse...

5. And here's an updated room view. The engine I'm using doesn't mind high-polygon displays that much; hence I have ~30k polygons in each of two cabinet displays. The difference showing vs. not showing: 84 fps showing, 89 not showing. Since my goal is to stay above 30 fps, I have ~50 fps left to spare.

## Cloud Ocean

So, I've decided to rework the setting of LotA2 to something completely unique, never-before-attempted in the world of games. Well... except for final fantasy, of course - but even there it tends to be just a minor focus. Here the whole setting is built around it.

I have an opportunity to create, with Game Maker, a game not only artistically beautiful but technically competent and functionally unique. I have added a glow layer, unlimited antialiasing, and a world of gigantic scale.

LotA II is now... Cloud Ocean. A game of airships, piracy, and adventure.

Above: Glow on, 0x AA, and a color filter I call semisepia (a sepia-toned overlay alpha mapped from intensity)

The brawl engine will likely remain part of the game.

## Introspective Short 1

Christian games are the puny weaklings of the gaming world. They try, but they always fall flat.

Is there still a game genre waiting to be discovered?

Is this genre well-suited to Christian games?

Can I do better?

Can a Christian game truly succeed in the commercial realm?

I've created what I believe to be a finished, working candidate for the YYG contest. Fun times.

Condition: Stressed.

## Log entry, April the 16: The Pass and the Furious

Item of business 1. An image of The Pass, my 2d arcadish RTS.

All the graphics are Blender renders, and are designed to fit into memory (with GM, 20mb of png images are uncompressed to ~400mb of memory), hence the low resolution on the units. On the upside, that amounts to ~20,000 frames of animation, which is no small number.

As I mentioned earlier, I've reached "feature-lock" stage and am now ironing out bugs...

Item of business 2. The Brawl engine now takes into account various factors for each combatant to decide:
1. If they even care about the conflict
2. What side they'll enter on
3. And what they'll do when they do enter it.
The character's disposition, your fame/infamy, differences in allegiance, and the character's drunkenness all play a role in deciding this. Some people may even be so disagreeable that they'll attack or run on sight of you.

I also added primitive line-of-sight elements - drawing your gun/sword will only affect those who can see you do it... but shooting it/killing someone/destroying things will draw attention from the entire building.

Next up: Integration of the mouse-driven combat interface with the AI engine.

## Log Entry, April the 15th: The Brawl Engine

LotAII will feature land fights (hence my character models); I plan for these to be quite dynamic and engaging - like a hybrid FPS/action game.

What I have:
* 1 character model, half done
* 1 mouse gesture recognition system for combat, mostly complete
This is quite an interesting affair. Your mouse can be used to stab (left button), block (right button), switch weapons (scroll wheel), perform advanced maneuvers (for instance, just moving the mouse left in a sweeping motion will cause your character to slash to the left; there are 5 unique moves to be performed in total, many with directional variations), and fire as in an FPS (with a gun selected, the left mouse button fires; probably a mouse maneuver (move cursor down, then up) will reload; right will block with rifles or shotguns; and moving the mouse straight down will bash with the selected gun)

It works. 'Nuff said.

* One AI maneuvering engine, titled Brawl. So far, it handles retreating and cover; eventually, it will handle split-second decisions as well. It has the potential to work on procedurally-generated maps (although at present the map used was custom-designed) and should act realistically in a fight.

Above: The base "starting" configuration. All enemies spend about two seconds "looking around" for a hiding spot if you have a gun and they don't. Possibly eventually if they /see/ you with a gun they'll run; if they see someone else running they may come to investigate.
Enemies are red dots; you are the green dot; little white dots are "cover spots." Certain cover spots have "leaning" tendencies; characters near here can lean out to shoot at enemies. Other cover spots are associated with actions; so, for instance, an enemy can tip over a table to create additional cover. Different spots have priorities associated with them as well, making AI response more realistic.

Above: The situation a few seconds later. Note how most of the enemies took up hiding spots facing away from you and as far away as possible. (Although proximity to the enemy also plays a large factor)

Above: I moved up to scare away all the enemies in the left side of the room. After a bit of rabbit chasing (they may just walk around the table to get away from you) they were all corralled in the other room. Note that certain enemies are still running; if a hiding spot is too crowded, they won't tend to favor it. In this case their response is more one of "terminal situational indecision."

Above: And again, I scared them back across the room. Some enemies will cluster around the sides of a table facing you; it's far enough away/close enough to them that they feel safer there (but probably aren't). My stated goal for these fights is that they should realistically involve up to 10 characters in a single region or area. I think that this system does adequately complete my goal; hence, I am quite pleased with it.

Basic rules of engagement:
One or two swordfighters may engage someone else in a swordfight. People with guns should stand a ways away and fire at their enemies (moving back if needed; they can be cornered if you can avoid the stream of bullets)
If you have a gun and someone else does as well, they may either begin firing at you or (if more prudent) run for a coverspot with a leaning opportunity, then lean out and shoot at you.
If you have a gun and they don't, they run for cover, as seen above.
If they don't have any cover (say, a square room), they may simply charge you, or cower in fear or back away. The goal is to make your enemies seem human.

Combat details:
You have an extremely limited number of hit points (5 at the moment; attacks may take up to 4 at once) and 8 balance points (needed to perform various attacks). Balance points recharge with time, but slower when blocking. Almost all attacks and blocking take balance points to perform; the blockbreaker will damage an enemy's balance and bring him out of block. Blocking and using initiative is crucial; oftentimes a major enemy attack will leave him open for a few seconds.

## Log entry, April the 12th, 2008: Who are ye anyway

So, I finally decided to pony up the $13 to get some public critique/ego-bolstering from my fellow game developers... ...Hence, this Log. Items of interest: 1. Who am I? 2. What am I doing? 1. I am (at least, as far as I know) a jack-of-most-trades programmer/modeler/designer/tester/scavenger. I do most of the work on my projects: ideas, design, programming, 2d and 3d graphics creation, and testing. I use online resources for sound and music. I am a Christian, a bit of a grammar Nazi, and an avid game player. I am currently in college, where I study computer science, make games, and play Nethack. I have in the past placed 5th in the 4E5 competition here. More on this later. I have produced a number (more than a dozen) of games using Game Maker - which, I have found, is generally regarded as a child's development tool. I hope to prove otherwise. 2. At the moment, I am working on a pair of projects semiconcurrently: an entry for a game contest and a sequel to Lions of the Atlantic, my 4E5 contest entry. The former is called The Pass. It is a very fast-paced arcade-style RTS. * Combat normally occurs within a minute (versus AI) and the game is normally over within 30 minutes. * The game centers around your hero, who can lead units into battle, attack, and cast spells. The viewpoint is locked to your hero's position; only your hero can gather resources or command troops. In other words, if you happen to be out of town when the enemy marches in, you're in trouble. * The goal of the game is to destroy your enemy's Base Flag - sort of like Defense of the Ancients or the various sidescrolling RTS games out there (Cortex Command). You create units with gathered resources; make them follow your hero; attack with them; and eventually overpower the enemy. * The game features a persistent hero "RPG-lite" mechanic. Your hero gains experience from battles; this can be applied towards upgrades in stats and spells. There are 3 nations to choose from (Egypt, Greece, and China); 3 classes (Swordsman, Archer, Knight); and 3 gods per nation. Spells work similarly to Age of Mythology - you have a fixed mana pool for each game, and can cast known spells from that pool. * The focus of the game is on fast-paced multiplayer gameplay, although it also features reasonably competent singleplayer AI. The Pass is currently in feature-complete, content-complete stage (AKA late beta) and is set to release before April 29. If I beat my competition (which should be easy, really) I get$1000.

My other project is codenamed LotA II. It is a "naval trading/combat simulator" set in the Caribbean (think Pirates! etc.)
It will feature the same type of gameplay as the original, but with some added sections (third-person adventure/fighting). It is currently in the early preproduction/concept stages - but I do have some stuff to show off anyway.

My basic human model, with "Rich merchant" clothing, untextured. It has animated eyes, mouth, and a full skeletal rig. Time to create: 2 days. Polygons: 4000.

A very early test-rendering of a ship model and my ocean setup. Uses Game Maker; features reflection and animated water. The ship model is 1300 polygons.

I also have a map of the Caribbean which I am currently converting into some 200-odd land patch models for third-person exploration.