I managed to carve a good chunk of time working on city generation over the 4-day Thanksgiving, but I wish I were done. My core accomplishments was in creating single-wide alleys, more crafted patches for both 2-lane and alley roads, and most importantly: Allowing patches to be rotated when placed.
This allows me to create more variety, not just because things look different when rotated, but also because I can spend my time on crafting unique areas without having the create four direction rotations of them. It’s getting there…
The biggest problems beyond variety is playability. The most interesting maps should have some interesting tactical spaces, and more importantly enough loops that give the players a better sense of exploration as well as not punishing them as much for getting some speed going (which dead ends can completely wreck). Typical Roguelikes (Dungeonmans included) introduce at least a few loops so that exploring the map doesn’t force endless backtracking.
A whole bunch more work is needed, including:
Make additional large patches with interesting tactical features such as open areas, wide runways and so on. This will help keep the game from just being a bunch of corridors.
Create an evenlarger 4-lane road type and seed the map with a couple of large roads. This should present some neat places to build up some speed.
Checking patches before they are laid down to make sure that they are not blocking off an adjacent road (this leads to a series of roads that lead to nowhere).
When the map has been expanded as far as it can be (usually a set value, such as 1000 attempts at placing a patch), make sure that all the “unexplored” road tiles are capped with dead ends or are connected up to their neighbors. (Again, this avoids road connections that terminate abruptly.
Add some “overlay” rules and tagging that allows for “optional” entrances into a block. This way if I try to lay down a block next to another block, I can more easily “bust a hole” between the blocks as needed. This will be huge for generating loops.
Create a few rules to evaluate a “good” map, including if there is enough space to explore and that there are areas suitably deep in the city where boss areas can be placed… and if those criteria are not met, throw the whole map out and start over.
I basically need to take this as far as needed until it’s fun, and then step away from it and worry about making it perfect later. Something like this could take all of my time for many months if I let it. Over time I’ll try to add new models, streetlights, textures, buildings and so on, but I need to work towards something I can play again.
Once the map generates well (hopefully in another week), I’ll need to properly populate it with encounters, and create a boss area fairly far from the entrance that players need to play towards. Once I’ve got that I’ll be back to having a game loop and can push to sharing a build out. That’s something I really want to get done before the holidays.
Another big gap between the last update and this one. Part of it was a good thing (two-week vacation to Japan) and part of it was a bad thing (two weeks of airport-spawned sickness after going to Japan). Then for the last month I continued to purge that damn technical debt.
It turns out there were a lot of things that the game generated (including things I wrote this summer when creating the overworld) that was mired in sub-awesome way of storing things. I can’t say things are perfect now, but my Perforce tree is starting to fill up with all the datafiles I can now create, search and copy to my heart’s content.
The datafile system (again, inspired by the one used in Dungeonmans) allows me to use tables to reference other tables, as well as arbitraily roll dice for whatever I want. I still have entities to squeeze into the system but I think everything else is in a pretty good state. Now I just have to train Notepad++ to parse them just a hair better and I’ll be in a great place for editing.
One thing I won’t be converting into text files is the Patch System I wrote for my initial attempt to create cities. I wrote that system to integrate well into the entity editor, and I can drop in a bunch of tiles into an object and then drag the object right into an info panel to bake it out. I’m not great at tools but I’m pretty happy with that one. I’ve only just started to fill out variety with it but I think it can produce some really crafted setpieces in the midst of the pile of procgen that a Roguelike generally has.
Coming back to working on the generation of a quality city, I decided to double down on the value of Patches and try a generation system that capitalizes on it. We colloquially call it the “Crown Royal bag” method, once again inspired by Dungeonmans (Jim mentioned in his talk at last year’s international Roguelike conference). It was actually first written about by Mike Anderson on RogueBasin. The idea is to keep a list of all your legal walls that can hold doors in a list and draw randomly. Pick one of those walls and consider busting a door into the side. Then pull a precrafted “room” from your virtual Crown Royal bag at random and see if it fits onto that particular doorway. If it doesn’t fit, throw it out and start over. For hundreds of draws you might only place 40-50.
There are number of issues using this for a city. First of all, as you can see here, my tiles are created with entrances and exits built in… I had to create an entrance and exit tagging system to figure out what tiles were valid connections to each exit. That wasn’t too hard luckily, although I have to create connections now for single-lane roads and just plain asphalt and dirt, which are more interesting alleys and short cuts than the core big, two-lane roads.
Second, the technique basically creates a tree rather than a nice, playable dungeon with loops. I’ve got a ton of ugly dead ends and roads that lead up to other roads but don’t connect. It’s important for the proper feel that intersections look right and create nice setpieces with crosswalks and driveways and stop signs and such. To accommodate this I need a more sophisticated system, in particular a way to overlay roads onto other roads and have them convert to intersections. That actually means that I’d be heading into creating a road link system instead of relying on handcrafted patches, but I think I can integrate some smart tagging on my tiles to handle this.
The battle continues… Trust it’ll look better soon.
I’ve had some pretty great results so far with generating my terrain and dynamically placing road decals as needed. The Easy Decal 3D solution ended up being pretty great. The folk at &u Assets were super-helpful, and with clever use of their baking and decal combine features my roads went from 20 fps to 200 fps. So, basically solved, and it’s worked out pretty well for putting interesting terrain features when staring down at splat-mapped Unity terrain starts to get tiresome.
I feel pretty good about the look and feel of my overworld, as well as the scale (which was a concern since it’s extremely exaggerated, even though I wanted a gritty feel). I still have a ways to go before I can finish up my core loop and release a build, but things have leveled up a fair amount.
This past week I’ve been polishing a new feature for procedural generation, patches. This allows me to hand-create a collection of assets (both single-tile and multi-tile) that gets placed down in a singular area by the terrain/city generator. It can be a solid square or only fill some of the squares within. I can import any patch by dragging in a group of objects from a sample scene. The tile system already converts the patches into the tileset currently in use by that map, and mixes in variants as they are available to the tileset.
The patch system has a lot of additional features yet to do. For example, allowing the definition of patches in a nine-slice style where I can scale them to an arbitrary size with consistent edges. This would allow me to create arbitrary-sized plazas, but more importantly, I can create a road patch that has signs, lights and various elements on it, that scales according to the length (and width if desired) of the road.
I also want to put in randomized prop points: object stubs that will randomly place a specific prop in that spot, anything from a trashcan to a stain to an ambient effect. It also allows me to place variants of an existing entity, such as a destroyed version. This will be important so that my generated city doesn’t look all pristine the way it does above.
I didn’t expect to get dragged into city decor so early in AutoFire’s development. It’s certainly a topic of interest to me, but the push in that direction really came from trying to find props for the game. My 7DRL city was fine, it was basically a dungeon. A dungeon can be a twisty maze of passages and hallways don’t necessarily need a specific direction of travel defined. However, it felt like 90% of modern-day props required placement with some thought… You can’t just sprinkle in mailboxes, street lights and stop signs via Random.Range(0, size)… There needs to be established clusters and strips of materials. I can construct hand-built areas anyway, but to extend everything into procedural town I’m pushing the limit of my limited home-grown tools.
Rolling from all this, I’ve also discovered as my game’s scale gets better established that I’m reaching a point where my single-square building assets are looking weirder and weirder… I want the cars to be big and bold, and characters in the world to be larger than life also, but right now they are about two stories tall. I was okay with something more abstract and representational when I was in 2D, but working in 3D makes that scale a bit more important. Aaaaaand the only way to get my building “walls” closer to the “correct” size is to come up with a system for merging building squares into larger structures. That will take a whole new post-process that… I’m going to wait on.
I’m quite aware that I’m running the risk of entering a bottomless pit of effort… Making a city look like a city is hard, and many games make that their only thing, whereas I have
so much more to do. So, I’m going to finish up this patch functionality and pull back to environments that are a bit easier to generate, such as frontiers, shanty towns, junkyards, and so on. This way I can push back into mission generation and create some assassination missions and encounters. This will finally lead me back to the promised land: the refinement of driving mechanics, combat and its associated UI.
So unfortunately I’m going to have to chalk up one in the “fail” column for my February #1GAM, which was to be titled “Enter the Catrix”… I had until this evening to get it submitted, but after over a week of computer problems (complete drive failure) along with some Unity quirks that waylaid me, I’ve decided that despite a last drive for completion I am not comfortable releasing this thing. It just wouldn’t be something I want people laying hands on at this point.
It’s a shame really… It took some effort to turn a couple of cats into the animated characters, and then casting them in this point-and-click combat system. The idea was for the player to choose between slower free movement on the floor and speedy movement from “perches” as he maneuvered the ninja to the right angle for slaying the robot cats.
The original concept for this thing was simple… Make a fast, mobile combat experience with some flavor similar to what we created in X-Men Legends… dashes, knockbacks, walls getting destroyed… Of course I knew what I could get done in a month would be rough, and I’d be at the mercy of the animations I created, but I still thought I could capture some of the mobility I was after.
One of the core ideas was the idea that the map was filled with “perches”… These were ninja-style balance points that the cat could hop on crane-style and move quickly from point to point, hopping down for a quick fight. This provided the possibility that the player could dash to these points for a quick maneuver around the back of an opponent, but these points were not always available, requiring the player to hop down and run around instead.
My choice of perspective was one of my first problems… I needed objects with height to be perches, and yet I wanted some really fast horizontal movement when on the ground. For an easy point-n-click interface it was okay, but making things look tall and having them easy to hop onto was a bit of a pain that could probably have been solved with time, but was non-ideal since it forced me to move the camera more to a 45-degree angle rather than using a simple top-down perspective that worked with my 2D-animated characters. I got the characters to orient properly to the camera, but I didn’t manage to get the collision and controls to quite feel like they were going exactly where the player wanted them to.
In a more practical sense, I also wanted to use a modern or future city, or perhaps an industrial complex, merely because of my use of “robot cats” as enemies. (I didn’t want to create a game about my cats actually killing each other, as you might imagine). However, I didn’t manage to get a lot out of the Unity store, so I was working with a lamp post, mailbox and a fire hydrant as perches and not much else… Not a great method of immersing the player. Briefly I just littered the world with Iron Monkey-style poles and assumed it would just be a natural part of the world, but tall objects failed pretty hard.
I would have loved to consider this point-n-click method for iPads and touch devices, although I did use a right-click mechanic for throwing shurikens. However, I needed to make those more important for specific situations, so that melee, maneuvering and projectiles all had a part in the combat tactics.
Anyway, I don’t think this particular experiment is just a dead end, but I’ll need some time to re-explore it. With only a month to create the whole thing, I didn’t create the robust animation state machine that I really wanted, nor the proper input and physics… the push to finish resulted in more prototype-y things which can fall apart easily. I hope to work on it later and get some more crisp systems that I can leverage into games at a future date.
Recently I was looking at my old WoAP post and thinking about how the fundamentals of creating a good, solid-feeling weapon are evolving, and which basics are the same as they were 15 years ago. Developers are definitely better at these things, but experiences are still all over the board.
I gathered more footage from games released in the last year in preparation of a WoAP section on this site (which is still locked for now), but I fell behind in capturing the newest of the new. While that certainly won’t deter me, I did realize that I did have one source of state-of-the-art titles with readily-prepared footage… E3 2010! Sure the games are in-progress and the available footage varies in quality, but I thought it would be fun to see what the next year could be like in terms of nice, beefy implements of destruction.
As usual, I examined a number of elements to consider whether they contribute to a sense of power. Check out the original article if you wonder, for example, why a real-world weapon still needs to follow these rules:
Gun Visuals: The visual effect of the weapon’s muzzle flash and bullet trail, and the animation of the weapon itself firing. It’s the first thing people think of when they consider how the gun looks and feels.
Sound Effects: How’s the audio, from the firing to the impact sounds? This can be one of the hardest to get right, because the sound has to bear repetition thousands of times, and it has to compliment the rest of the arsenal.
Effect on Environment: Here we look at the visual impact of the bullet on a wall or floor, as well as environmental shifts such as lighting. We need these because the results of stray bullets give the player a sense of affecting their environment and with a potent weapon.
Effect on Opponent: This refers to the visual impact on an enemy, and how the enemy reacts to being shot, which is critical to give the player feedback that he is successfully damaging the foe.
Combat Effectiveness: I’m completely subjective here, which can be unfair when taken out of context, but it’s worth talking about. If I see a weapon limply pile bullets into an enemy with little sense of accomplishment, that weapon doesn’t get high marks. Maybe it’s designed to be a pathetic weapon, I don’t know.
Each weapon is grouped with its general type: They aren’t all identical models, weights or even technology, but each group of weapons fills a general role for players. I found as much unadulterated footage as I could… nothing from trailers or anything less than the game being played. This is obviously not final weaponry, and I had to make adjustments for the situations where the audio or visual quality wasn’t the best. Suffice it to say, they previewed their games to us and basically I’m previewing the weapons right back. Wah.
I didn’t get footage from everyone I wanted (technical issues denied me Brink and Killzone 3), but there was plenty of material (in unbiased alphabetical order of course):
The Agency, Sony Online Entertainment
Blacklight: Tango Down, Zombie
Breach, Atomic Games
Bulletstorm, People Can Fly
Call of Duty: Black Ops, Activision
Crysis 2, Crytek
Halo: Reach, Bungie
Medal of Honor, Electronic Arts
Rage, id Software
(Too many one-word titles that start with “B” this year…)
And away we go. Watch the footage and check out the results:
Almost 10 years ago Halo and Counterstrike created the first pistols that we actually wanted to use, and more recently Call of Duty cemented the concept of pistols as a quick-draw backup weapon. This trend continues, although video is hard to find because these weapons aren’t “sexy” enough to appear in E3 demos.
Call of Duty: Black Ops
This pistol is a backup weapon in the Call of Duty world. As such, this revolver is serviceable, but not flashy enough to encourage you to use it when you aren’t forced to (like this claustrophobic tunnel). It’s hard to say much from two bullets, but the muzzle flash was nearly invisible and it didn’t light up the surroundings as much as I’d might expect in a dark tunnel. The audio was also very basic, a clip we’ve heard many times in the last 10 years… I hope by ship they can get some reverb to make it as deafening as I might expect gunshots to be in a space like that.
The effect on the enemy is hard to gauge since this was really a set-piece. There was a reasonable bit of impact blood and a reasonable hit reaction, which hopefully will be true in all combat situations.
Medal of Honor
This footage is from DICE’s multiplayer mode, so it leans closer to Bad Company 2 in overall feel. Compared to COD:BO’s period-authentic revolver, it has the advantage… semi-automatic so that the player can unload quickly and go for a headshot, so its effectiveness appears pretty high. The bulk of the muzzle flash is a classic single frame but it does the job, mainly because it is coupled with a satisfyingly “crunchy” sound effect… not unlike Indy’s awesome pistol in Raiders of the Lost Ark.
Stray bullets create a subtle puff of smoke with environment impact, but there was little impact on the character himself visible beyond the view-obscuring muzzle flash sprite. This is often okay with pistols, as long as they have their role.
The Agency pistol is pretty good, although it’s hard to judge it completely fairly because it appears that the game view appears a bit stretched, making the pistol stubby and awkward-looking in the player’s hands. (I double-checked the footage to verify that this was the way it showed.)
The muzzle flash is very red, just a couple of sprites… it’s got an okay punch to it, but red doesn’t get it looking “hot”, and the sprite never reaches full opacity, giving it a “flat” feel. Perhaps it is an art style choice… this game is more colorful than most. The weapon seemed somewhat effective, but during recoil, the weapon obscured the target denying me the ability to see what happened. On the other side, the sound effect has a nice report and trail-off.
Rage‘s pistol has an okay muzzle flash, although its edges are fuzzy that looks a little “old school”. In this environment the world lights up well in response. From the audio present in this footage, sounds “crashy” and needs punch, but it’s hard to judge because it was not recorded direct-feed.
The biggest disappointment I felt was the weak sense of stopping power with this weapon. The interview toted the enemies’ ability to react to shots realistically, but I think it could still benefit from a more prominent effect on impact. In addition, the pistol just didn’t seem to be taking down that enemy effectively. Bungie was the first to more or less prove that longer-lived enemies appear more intelligent, but I hope they can use their hit-reaction tech to give you a better sense of doing something to your foe by ship-time.
To me, shotguns are the most satisfying weapons to use. They have a strong, easy-to-grasp tactical role (lotsa damage when close), but they also can deliver great pyrotechnics in the process. I only found a couple at E3.
It’s difficult to get a true sense of the Crysis 2 shotgun in the midst of all the chaos in this demo, but I do like the strong, directional array of “hit spangs” rising off the surface from impact. That really speaks “multiple pellets” to me. The muzzle flash is nice and hot with a one-frame “punch” to it with just a wisp of smoke after. The audio was pretty boomy, but it was hard to single out in the footage.
I can’t figure out what’s going on with the shotgun against the enemies here. Perhaps that is because of the situation, but the effect of weaponry should stand out in the chaos of combat (that’s the point). I hope the weapon does more to the enemies than it appeared to there.
Medal of Honor
This shotgun is a big guy, serious business… its size might set it apart from models in other games, but I’ll say that it looked menacing in the player’s hands, and had a satisfying muzzle flash and “booming” sound to it.
It also looked like it did a number on the enemy, sprouting multiple bloody impacts, and they didn’t get lost behind the muzzle. I would have liked to see more of what happened when a surface was struck, however, it seemed fairly subtle.
Now we’re down to business. Just about every game has one of these, and I define them loosely: “automatic medium-damage weapons”. They may be assault rifles or heavy submachineguns… subtle variations on a theme, so hardware-heads shouldn’t get their panties in a bunch about terminology.
This one felt a little silly to preview since the beta has been available for a while, but nonetheless I wanted to compare it with other upcoming titles. I was not a fan of the original Halo assault rifle because of its anemic audio and flash, but with this new version I’m feeling it more. The “teletype” sound effect now has more rumble to it, and the muzzle flash has less of a cartoony feel to its “star shape”, and looks reasonably hot.
The weapon is still of the “hold down the trigger until the enemy soaks up enough bullets” variety, so in multiplayer it helps to have XP popping up to know that you did something. Luckily with their newest animations, the enemy reacts more satisfyingly to the impact, even though there really isn’t much of an effect. The world impacts are unfortunately subtle for this weapon, basically invisible from a distance… It’s hard to be objective because historically this rifle deliberately lacked punch as a common weapon, but as one of the first things in the player’s hands it’s pretty important to feel good. Luckily, it now seems to have evolved to no longer feeling like a second fiddle.
Medal of Honor
A typical sighted assault rifle, this weapon has a great “train chug” sound to it, and it does some nice damage, throwing dusty impacts all over the environment and struck enemies. It definitely looks like an effective weapon.
The muzzle flash is nearly invisible however (doubtless matching the real-world weapon?)… the in-view weapon’s subtle shake is the main visual cue that something is happening with it at all. More importantly, the animation result on the enemy is still poor. It can be hard to play proper hit animations while an enemy also is moving, but here it felt like they kept running until they died with a minor twitch. This is a common challenge when shooting at multiplayer opponents, since they continue to run at full speed while taking damage.
What a difference from the previous MOH footage… without the background din of combat, Breach seems silent by comparison. As for the weapon itself, it has a “tiktiktik” sound that makes me think of a biplane’s machine gun. You still hear it sometimes in WWII games, but it sounds archaic because of its low rate of fire. I believe this is supposed to be a modern assault rifle, so there are definitely options that could be used to punch it up.
There’s lots to do here, in muzzle flash, enemy and world impact… I’m even not sure how effective the weapon was against the enemy given the feedback. Also, I appreciate the trade-off offered when zooming the weapon obscures a lot of the view, but thanks to the Call of Duty series, so many people play this way that it seems ridiculously punishing. This game could just get a bit more lively in ambience.
Blacklight: Tango Down
Another lesser-known title, Blacklight has a bit more tightness to its machinegun. It’s still a light weapon, with a fairly teletype-sounding audio report, but the very high rate of fire (with a nice envelope as a burst is fired off) definitely makes me feel better about it. The environment lights up well enough when firing, but the muzzle flash (from the footage) seemed very static and soft. The tracers flying out to the target could have looked good but these are thin and hard-edged… making them look like a rendering artifact beneath the sights. The big impacts (see below) helped a bit, but the weapon could use even a little bit more of a sense of danger.
The game seems to have a fictional dressing that allows for some sci-fi-looking elements sprinkled into the realistic environs… They used it well with the strong blue hit impacts when an enemy is struck, but they are also very big and distracting… you can only tell that you hit “something”. I could an impact that big for a sniper rifle, but for a bullet-spewing weapon like that, I’d dial it down.
Once again, the weapon appeared a bit stubby because of the odd aspect ratio, but in general it hit the Call of Duty mark pretty well. Audio-wise, they went for a strong gunshot for each individual bullet along with a lower fire rate. It makes the weapon seem heavy and mechanical (good), but over the course of a burst it grows pretty monotonous, like the old Doom minigun that just played the pistol sound over and over. I’d love to hear a more defined start and end to the bursts.
The weapon itself could use a bit of shake or something to tell me that its doing something. From first-person I can’t see the muzzle flash and it appears like the player is just holding down the weapon on its target. It does do the job functionally, as an effective weapon. Also, the environment impacts are nice and sparky… a bit random but I did get a sense of the player tracking their target with successive shots.
The muzzle flash in Rage’s assault rifle looks okay, but the old practice of fading out “star” flash sprites reduces the feeling of a “fiery” effect, and I see it here as I did in the pistol. Unfortunately, the audio from this footage is poor making the assault rifle sound “crashy”, but I’m sure it’s got the “pop” in person.
The weapon does light up the environment, but in this environment it can only be seen at a very close range, so I didn’t see it much. The weapon recoiled well with nice shells being thrown up visibly… However, I still wanted to see more feedback to where I was shooting. The video quality makes some of these effects darker than they really are, but given that this is a dark-looking place it could be a general issue. I wanted to see the results of my gunshots on the walls, and the floor. Finally, I also didn’t see a real threat to the mutant again… it’s like the pistol. No stopping power visible. Crank up the reactions with that reaction tech… make sure the impacts are visible.
These guys are obviously about showing off as much as their game is, and I feel it here. The rifle here felt dangerous, with a bright, fiery, and non-repetitive muzzle flash. The gunfire audio is has a crunchy mechanical sound to it, and that worked great (unsurprisingly similar to Gears of War‘s).
The weapon’s seems very functional and effective in taking out enemies. It was interesting to see the cartoony projectiles, almost like a scrolling shooter’s round bullets. On issue was that there was so much going on at my muzzle and with the bullets traveling outwards (this is an unapologetically noisy game) that I wasn’t always clear on what was being hit. I also didn’t see much in the way of impacts at a distance, although there was other footage against a close-up boss that was more tangible. The game’s about chaos, and I imagine it’s a lot of work to make sure the player knows what is being done to whom, and where.
This footage isn’t great (to be fair, the video is wayyyy oversaturated), but I wanted to include it because of the blinding amount of feedback from the rifle. The weapon has a huge star-shaped muzzle flash, which probably looks good in screenshots, but doesn’t have a lot of variance to make it vary over a sustained burst. It also manages to cover up a ton of what is going on at the impact site.
The game shows a Crackdown level of feedback, with orbs flying out whenever an enemy dies… The world impacts are also flamboyant at times, with some crazy cascading sparks. Unfortunately the chaos meant that once again I wasn’t entirely sure what was going on when I hit an enemy. I like “cartoony” as much as I do realism, so I’d have to reserve judgment until I get to see it without blurry footage.
Vanquish is clearly not a traditional FPS, but I wanted to include it anyway for contrast. This machinegun has an extremely high rate of fire, with a somewhat “lighter” sound effect than blends into the background din. This is the sort of machinegun that Japanese sci-fi games have exhibited for the last 10 years or more. Whereas western games put the machinegun in a place of honor as a powerful workhorse, Japan casts them as easy-to-target weapons that chip away health slowly.
The muzzle flash and impacts flamboyantly throw around sparks, but this is still a backup weapon, like the MG on a tank or jet fighter. They leave room for the more “cinematic weapon”, which may be a large explosive or melee attack. I love these sorts of games to death, but it’s interesting to see how very different the influences are.
I don’t generally include sniper rifles, but this time I figured I would. The category includes bolt-action or single-shot automatic scoped rifles, so I didn’t want to get the hardware nuts mad.
This particular weapon is Halo‘s marksman or battle rifle, rather than their sniper rifle, but aside from stats they seem similar. I’m surprised at how little feedback you get when firing while zoomed… there is a tiny recoil in the view, but aide from that, the sound and the reaction of the enemy is about the only result. The sound, however, is sharp with a great decay and the hit reactions are great.
The muzzle flash isn’t too important, but I would love to see something in the world. A faint tracer, or a more clearly visible impact effect. While there are caveats to this (too much pyrotechnics will render snipers too visible), it could use a bit of punch.
Blacklight: Tango Down
Blacklight‘s modern-seeming game goes far further in the “spacey” effects sometimes than the far-future Halo… Their sniper rifle fires what appears to be a solid laser, with a large blue impact on the enemy. This is fine for their vibe, but given that capability (freedom from “realism”!), I’m surprised to not see something more “impactful” going on. The blue flash is a round cloud of sparks, and enemies more or less fall down in response to being killed. The “laser” projectile could benefit from variance over its lifetime like a “railgun”, and the impacts are crying out for some sort of directional component, ricocheting away from the surface, or piercing the enemy somehow.
The audio is fine, but has less of a punch than Halo’s seems to. A gunshot from a sniper rifle is never, ever work being subtle about.
The sniper rifle in The Agency has gunshot audio that is slightly muffled for some reason… it has a punch, but the decay is perhaps a bit less satisfying than Halo’s “throatier” sound. The weapon itself does not move when fired, which is a disappointment, since there is no muzzle flash to see. Since the zoomed view appears sensitive to control, it is even more important to deflect or disrupt the player’s aiming since the overall vibe is that the weapon is “light” and can get jarred more easily.
I found their distinct tracer interesting, but unfortunately the effect is so solid that some rendering artifacts (resulting from rendering it as a flat polygon) appear when shooting. I like the idea of a tracer, especially for the “James Bond modern fantasy” world like this one, but work needs to be done to break up the line and make it seem less like a laser beam. In addition, I didn’t feel like I saw the result of my gunshot on the surface or person being targeted, because the tracer obscured it entirely.
This is E3, and anyone who has prepared something for E3 knows that someone picking it apart can be infuriating. Worse yet, the quality of E3 footage varies, so my comments may be influenced by lighting, background noise, or bad focus.
Not all the weapons or features or effects I’ve examined here may be final, but on the upside, just about everything I mention is remediable with a small amount of audio, art or balance work. You can disagree with me too, that’s what comments are for. The same goes if you are working on one of these.
It’s fun to look at these items regardless… I think I got bit by the bug again. In the coming weeks I’ll take another stab at getting that Weapons of Awesome Power section at the top going, and we’ll see where things go.
Last time I talked about co-op games and the need to tweak the difficulty in the name of a better experience. I mentioned that inflating the total number of hit points can work, but it can lead to the trap of The Numbers Game, where your existing health and damage are scaled together, resulting in a constant game challenge that only shifts when you fall outside of what the game deems “fair”. This can work for RPG’s, but is troublesome when used on shooters.
This argument isn’t particularly constructive for the FPS designer, the poor soul who is scrambing for anything to keep things balanced and exciting in the wild wooly west of random online pick-up groups. We can’t really blame them for falling back on pure numbers when necessary (full disclosure: X-Men Legends was an unabashed example of the Numbers Game…). Solving this problem through other means is a tough one, but it is worthwhile to pursue avenues other than the venerable Bag of Hit Points.
When looking for ways to increase difficulty, game developers often wish the answer was as simple as adding more enemies to each encounter… “No problem! Double the players equals twice the enemies! BAM!” Sadly, very few game technologies give developers the luxury of unlimited enemies onscreen… It’s a yoke that just about all of us constantly struggle with. Back in the sprite days of Doom and Heretic it wasn’t a big deal; those games had extra enemy spawns that were triggered at higher levels or during coop. In these days of high-fidelity characters, however, most technology still keeps the population somewhere in the single digits.
There are tricks to give a feeling of greater foe count… Placing additional enemy encounters in the “dead spots” between the “standard” encounters works, if you never overlap the encounters to keep the maximum enemy count low. It can be tricky to pull off but it can give the player a sense of a relentless onslaught, requiring him to manage ammo and health more carefully even though he never sees more than a few at a time. Unfortunately, by leveraging this method you can ravage any sense of pacing, leveling out those tension peaks and valleys into a steady drone, and changing the game in perhaps unintended ways.
Another population trick is to deliver more enemies to the battlefield through respawning. This allows enemies to repopulate places that were previously cleared out, denying players a safe retreat and keeping the pressure on. This approach can be reviled by players because it denies them a sense of “completion” within a level (it was a contentious feature in Soldier of Fortune, for example), but it can be used to great effect. My first experience with respawning as a gamer was Doom’s Nightmare mode, where all enemies regenerated after 1 minute or so. (Damn that was hard!) More recently it was embraced in Left 4 Dead, where zombies can pour out of just about any opening, inaccessible fences area, rooftop, etc. Careful use of this mechanic can help a game level feel as though it is packed with hundreds of foes, even if your tech can’t handle it (although L4D did a damn good job with the population too)…
Up to 11
In contrast with developers, a majority of gamers wish that higher difficulty was just a matter of making the enemies smarter… perhaps implying the existence of a Jules Verne-era dial with “IQ” scrawled on the side that can be cranked past 10. Unfortunately game AI is never so easy that you’ve got unused slack somewhere that you’re not using. Generally it’s not decision-making that challenges the AI programmer or designer, but rather environment response and navigation. Sure, humans can walk across an obstacle-ridden field without even thinking about it, but an AI that does this naturally requires tremendous work whether the difficulty level is baby-like or insane.
Even if the leap from floor-traversing mouth-breather to devious mastermind were easy, I’ve repeated many times that “smarter enemies” don’t always pave the way to “more fun” anyway. Aside from major boss-level fights, the opponents in many games don’t last long enough for you to really experience their brilliance… showing off would require for the player to see them, which means they’re probably already busy killing them. Sure, you could make the enemies tougher so that you might see their amazing kung-fu, but then you’re back to turning them into HP bags.
Some games emulate enemy “smarts” by simply increasing their shooting accuracy or giving them uncanny perception. Let me be clear, this is not fun. Most games with gun-toting enemies need them to be inaccurate at long ranges. If an enemy pops into a play space a long distance from the player, chances are the player is not aware of their presence immediately… a bit of warning is needed before he gets a bullet in the head. If an enemy’s accuracy is simply dialed up arbitrarily, the player suddenly starts taking significantly more damage from enemies a few pixels in size… you’ve turned the game into a walk through a room full of snipers. Statistically the enemies are just doing more damage to the player each second anyway… so that’s what you want, just crank the bullet damage and be done.
There are some better models out there for bullet inaccuracy that can help remedy the fun, but they are still aren’t used enough (that would be fun to talk about in another post).
Phases of Death
While technically it is a variation on adding health to opponents, but one way of amping difficulty is giving them additional damage states, each of which has to be “killed”. Examples of this are the humans that become “tentacle heads” when headshot in Resident Evil 4, or the aliens in Blacksite: Area 51, that can sever their torso after being “killed”, crawling after the player. These imply non-realistic opponents or a heavy art burden, but it can also be done by having the foe walk with a limp in phase 2, or switching weapons each phase.
This may sound familiar because it is the way that just about every classic boss encounter works… and for good reason. Bosses are “tough” and hence can soak up a lot more damage… without small player victories and new behaviors coming into play, defeating them would be a tiresome exercise in shoveling damage until it goes down (and many action games still use this model).
Cleverly-designed damage stages also have the advantage of not always being used. A 3-stage robot can skip stage 2 and 3 in easy difficulty, last for a second stage in later levels, and die only after all three in harder difficulties. Depending on how an enemy is designed, it might even be possible to regenerate them to a previous state if not killed in time for even more challenge.
A distant relative of the staged model is a mechanic where an enemy must be in a certain state to be killed. For example, a foe may have a “stunned” state which provides you with an opening to slice off their head. By this I don’t mean “Japanese boss-style” where every 15 seconds the monster opens up his eye-dome and can be hurt, but rather a system that gives the player the tools to introduce a vulnerable state. Another example is a robot who must be immobilized with an electric shock, before you step in to disable his control chip. Or a soft-bodied creature that must be frozen first and then shattered (similar to how the freeze-wrench combo works in Bioshock). Games like modern military shooters are more limited in this respect, but if you have the luxury to play with your enemy ecology (that is, their function and response to various stimuli), a large number of additional options open up to you. As above, you can scale difficulty by sprinkling in these new mechanics more aggressively later on (say, early robots don’t have an “overdrive chip”, but all of them do at higher difficulty).
Other tricks aside, there’s still a value in delivering a smarter enemy… or at least one the gamer will recognize as “smarter”. One example of that is to add some coordinated behaviors between multiple foes. A simple trigger that makes several enemies charge at once or synchronize multiple grenade tosses can add difficulty in a hurry. Throw in an audio cue to draw attention to it and suddenly the gamer has a newfound respect for your AI mojo, baby. More complicated relationships like an enemy that lays down covering fire while the other charges can be good also, but the more complex they get, the more likely they are to fail or get lost in the shuffle. Sure, it makes things harder, but look for ways that are going to make the player feel the added pressure.
Left 4 Dead in particular leveraged this synchronization through their “AI Director”. While not technically an issue of “smarts” (the zombies are, well, zombies) when the director decided that the experience needed a massive assault for added danger, the player knew that the game was turning on the heat. The perception of challenge is perhaps as important as the actual difficulty increase.
Adding enemies with interrelated behaviors is also a method to give a feeling that they are “conspiring against you”. A prime enemy that adds challenge to any group is the buffing opponent. The Shaman in Diablo II and the Arch-Vile in Doom II could resurrect fallen opponents, making the whole group for more deadly and interesting to fight. Other, more mild examples of this type of enemy are enemies that have an aura that heals or increases the strength of their nearby friends.
A player must completely change his tactics and adjust when fighting a group that has a buffing foes added. Even one can add considerable challenge, and multiples can be devastating. Beyond just spawning extra ones, their behaviors can be tweaked for difficulty by increasing the buffing effect, casting time, or the radius of their ability.
Looking closer, any type of environmental object that enemies can utilize can add challenge and an extra feeling of intelligence, no matter how simple. Enemy buffs could be provided by emplacements rather than spawned foes. The health stations in BioShock kept you from leaving your nearly-dead foe alone, because he might return fully charged. Emplaced totems such as those in World of Warcraft are similar.
Reflecting on what I’ve written above, I see that there is no “magic bullet” for scaling difficulty (although I hoped that writing about it might shake one loose). The techniques that must be used will certainly vary from game to game… My main suggestion is to look beyond the traditional D&D numerical “crutch” when faced with systems that need to scale. I understand why we do it: it’s not just our inner fanboy screaming to get out… numbers are a necessary part of our job and getting results. However, I love designs where the player doesn’t have to be steeped in numbers in order to succeed.
You might also come out of this thinking that I turn my nose up at RPG’s. On the contrary, it’s one of my favorite genres, and Diablo II is still my favorite game of all time.
Finally, I certainly mean no disrespect for what Resistance 2 has accomplished in its multiplayer mode. Along with Call of Duty 4, I think it has laid a foundation for great semi-persistent online experiences for shooters. What others might build upon these concepts in the future makes my mind reel.
I’ve been playing Resistance 2 lately, taking in its new cooperative mode on the recommendation of a couple of friends. Co-op in shooters has a long but spotty tradition, so it was neat to see Insomniac deliver a non-competitive experience with a different feel. This one gives the player one of three classes that can be leveled up independently by matching up to play short missions. Each class has a different loadout and responsibility when played: grunt, medic, supply. In addition, there is a resource that can be gathered during matches in order to purchase upgraded abilities and weapon packages.
It was a fairly addicting experience, as I pushed to each successive level like I might grind an MMO. The matches themselves were entertaining on their own, with tangible sense of achievement every few rounds. Unfortunately, I eventually ran out of gas, not because the game itself wasn’t fun enough… but rather because each creature had 5-10 times the health of their single-player counterparts. The campaign was a well-balanced shooter with crisp control aim and a great sense of power, but coop had me holding my machinegun on what should have been “popcorn” enemies for several seconds, watching XP pile up as I waited for each bag of hit points to fall down. Quite simply, it didn’t feel at all like the shooter than I played when not online.
I cannot fault the concept of jacking up the time to kill each enemy… it’s a time-honored tradition from many classic games. In Resistance 2, it was clear that they needed to extend the experience and increase the effort required to bring each one down. It’s probably a useful excercise to discuss why this might be.
The Numbers Game
One reason for enemies to soak up bullets in coop is to match the level of difficulty to the number of players. If 10 players entered a level that was intended for just one, they would slaughter everything with their added firepower. Some games ship this way, whether due to limited resources or to reward the effort that used to be involved to connect multiple PC’s for a coop session in the first place. These days most do their best to notch up the challenge when new players join a session, and sometimes this is done by increasing the enemy toughness. Diablo did this in the most overt fashion, by reporting to the player that “the enemy forces have grown stronger” when each additional player joins the game. Behind the scenes it was increasing the experience level equivalent of all the enemies in the world.
Another reason for enemies feeling more invulnerable are in games with a progression track… ones that reward the player for each hour of play with upgrades in ability and potency (like Resistance 2 does). Unfortunately it’s easy to forget that in order to build a game where the player feels more powerful over time, you must hold back some of the most potent player abilities at the start of the game. The player certainly feels a sense of achievement as he gains all the new kick-ass stuff over the course of the game, but sometimes that means he also feels anemic at the very beginning, where he has the least capability. The damage potential for his weapons are less, his health is less. His weapons aren’t as flamboyant. Even the most basic enemies might take far longer to kill than they will later on. This is a dangerous practice that must be handled with care… these are the crucial early hours where a player should be falling in love with the product rather than feeling emasculated.
Somehow this is an acceptable practice in RPG’s… Those games are almost entirely about progression and acquisition, so it is expected that the player will evolve tremendously over hundreds of play hours (or thousands in an MMO). But because of these incredible progression arcs, most RPG’s play what I call The Numbers Game. You’ve doubtless seen it… when players start off their game doing tiny amounts of damage to wimpy rats, but eventually grow to deal thousands of points while fighting giant dragons. In these situations there is a continual arms race between the damage you deal and the health of your enemies.
For example: I’m playing an MMO and start with a character who can deal an average of 20 points of damage to an enemy who has about 100 health. About 5 hits will take him down. After some play time, I level my character to level 20, and can now deal 100 points of damage. Good for me! …except now most of the enemies have around 500 health. I guess it’s still 5 hits to kill one. Finally, after months of investment I reach the coveted level 50, and I’m clobbering opponents with 1000 points per hit. Of course, you guessed it, my enemies have 5000 HP (or more).
This shouldn’t be a big surprise, because as you gain power, it would be anticlimactic to see a lessening difficulty… we all want to grow up to finally beat that huge, scary thing that we fled from many hours of play ago. Keep in mind what this means, however: In a combat-centric game, a player’s primary metric of power is the number of hits per kill. (This abstracts to “the ratio of time investment per reward”… but that’s fodder for a later post). But when I play the Numbers Game, do I really feel better about taking down a Level 50 Hoary Drake with my Level 50 character than I did taking down the Level 1 Scrawny Rat with my Noob? The answer in RPG’s is often “yes”, but in shooters you can get in a lot more trouble.
The difference is in the essences of the genres… RPG’s deal with skill advancement primarily on the character itself, as he “levels up” and increases his capability through higher numerical stats. The player himself has less pressure to hone his actual playing skill, aside from juggling the new options presented to him when new abilities are unlocked. He makes choices about how he wishes to advance, working with figures like “strength”, “speed” and “willpower”, even though they often just present different ways of hurting an enemy. These various axes of advancement give the player something to aim for, a vast possibility space that he can explore and achieve in.
Since advancement is so tied to how the player’s capabilities are represented, the player keeps a much greater awareness of the numbers and how they affect him. He understands and accepts that an enemy that is 5 levels above him is extremely dangerous, because the numbers say so. This is totally fine, because most RPG’s are not about combat… they’re about advancement, acquisition, and a bit of exploration. (If you really believe that you played Diablo for the click-and-kill combat, I think I have some real estate you might be interested in…)
Shooters by comparison leave skill advancement largely to the player’s mind and body. Your manual aiming ability is your primary “accuracy stat”, and timing, dodging and area management are all critical traits that don’t live in the game itself. In most shooters the player’s character is just as effective with a pistol at the beginning of the game as he is at the end… even though the game progressively demands more of the player himself with larger groups of foes and challenging level layouts.
Shooters also are tuned for action experience, living and dying by their weapon balance and ammunition management… Killing one enemy is usually 1-3 shots, and an FPS starts to instinctively know which weapon is best for each situation. This is the biggest reason that shooters are so seriously wounded by the Numbers Game. By increasing enemy health arbitrarily, the choice of weapon eventually becomes less important. Enemies that used to be demolished by a shotgun blast take several hits, causing players to switch from surprise or flanking maneuvers to attrition tactics. Pistols go from being the standby for taking out weak enemies with minimal ammo investment to becoming basically useless. Different skills and sometimes abhorrent tactics are adopted in order to succeed because the game becomes increasingly “unfair”. Players might even start to think in terms of DPS (Damage per Second), a major metric in MMO’s and a strong symptom of the Numbers Game.
More Than Just Digits
So this argument does nothing to help out the intrepid FPS designer, who still needs to solve these difficulty issues… He’s willing to do anything to make the game as fun for 10 players as it is for one, and to make gamers feel increasingly awesome for each hour he plays. I don’t blame folks for falling back on the numbers when they need to; Diablo II is still my favorite game of all time, and when working X-Men Legends I personally applied the Numbers Game to near-excess (more on that another day).
Solving this problem through other means is really hard, but going the Bags o’ HP route should be a last resort. I’ll see if I can scrape up some alternatives in the next post.