Strange Design

Obfuscation masquerading as Clarity

This past week I had the opportunity to read some document created by the military, and I was stunned by how wordy and unclear they were.  The documents proclaimed themselves to be simplified texts, and further claimed that they would demonstrate how command processes are the result of clear, repeatable systems.

Systems are my stock in trade, so I read hoping to find some interesting parallels between military design systems and game design systems.  But sadly, there were no systems laid out in the document at all.  After all of the acronyms, disclaimers, diagrams, and examples were dispensed with, there was no substance to the document at all.

Naturally, I was disappointed.

But after further digestion time, I realize that there is a reason that the military documents were all jargon and no substance.  Military decisions generally boil down to a few common-sense judgement calls, and sometimes a blind guess or two.  Operations are divided up into such small components that nobody really has anything like a prisoner's dilemma going on.  So the people making these command decisions, for the most part, don't really need to apply anything more complex than their own good judgement.

Game systems, on the other hand, are a realm in which our intuition frequently fails us.  The game mechanics may seem lifelike, but they don't respond the way reality does, and so we need to be very careful when examining our expectations.

But the more important difference between military systems and game systems is the fact that game decisions are easy to change for the most part, while military decisions don't get additional chances.  When I make a bad call about a game system, we just change to another available option.  But when some commander's military decision doesn't pan out, there might be an inquiry to verify whether or not the decision was rash.  That's when jargon and obfuscation come into play - making everyone feel that the commander's wrong decision was unavoidably wrong.

So in that light, the documents on military decision making provide exactly the information their audience needs.

Considering: Moonbase Commander

Like a lot of people, I first became aware of Moonbase Commander when Tycho raved about it at penny-arcade.  He described the game as an almost mythical object - a project created by some passionate people as a side project to their regular game development duties - playtested for thousands of hours, and improving upon the classic gameplay of Scorched Earth.  And that's more or less what we got.

But for some reason Moonbase Commander never caught on.  It was published by Atari during a difficult time for that company, but I still managed to get a free copy for some reason (I had just finished G:DAMM for Atari, so there was some channel through which I could beg for swag.)  Since then it regularly sells for $1 or $2 on Amazon, so I've picked up a few extra copies to give to friends.  There are some free utilities which make hosting and starting online games a breeze, so I've been able to enjoy both the single-player and multiplayer in abundance.

Since the community has always been small, I've never seen much in the way of in-depth strategy for the game.  So I thought I'd correct that oversight. 

Gameplay in Brief:

You begin with a single building - "hub" - which offers a modest field of view around itself, and 7 energy.  You may fire a single shot from your hub, and then your opponent fires a single shot.  Passing is an option.  If you run out of energy you automatically pass the rest of your turns.  Once all players have passed is succession, the turn advances and players generate 7 more energy plus extra energy for each collector they have built, if any.  Because you initially have no idea where your opponents are, the first several turns are generally spent building up your infrastructure and exploring space.

Bombs and other offensive shots fly in an arc from your hub, with variable power you control.  With skill and practice, you can be quite precise with each shot, though random variability does rear its head.  But the really unique aspect of Moonbase Commander is the additional buildings which form an inter-linked web of targets.  Each building is connected by a power cord to the hub which fired it - cords may not touch or overlap, so at most a single hub can have about 10 structures around it, and often only about 6.  But hubs can fire additional hubs, which then fire additional structures and more hubs - so the layout of a base becomes quite complex very quickly.

Power cords are directional - they "point" to their hub of origin, and destroying a building detroys all buildings "upstream."  This means that when you lose your original hub, the game is lost as everything you own explodes in sequence.

But the best designed element of the game must be the 18 shot types, which are lovingly arranged into three tiers - 1 energy shots, 3 energy shots, and 7 energy shots.  The rest of this guide will be a close analysis of each type.

Tier 1 - 1 Energy Shots:

Bomb: the classic workhorse

The Bomb deals 3 damage for a single energy, making it the most energy-efficient way to damage a single target.  It has a very small splash radius, so near-misses still deal 1~1.5 damage.  The better you are at Moonbase Commander, the more Bombs you will use, in lieu of the more expensive fancy-pants options which are to come.  Ultimately, winning games means gaining an energy advantage over your opponent, and bombs are the efficient engines which drive those advantages.  Two Bombs destroying a 7 energy structure gives a net advantage of 5 energy to you, minimum.

Cluster Bomb: quantity over quality

Cluster Bombs provide three projectiles for a single energy, making them the most energy-efficient way to create targets for AA Towers.  Cluster Bombs deal 1.5 damage each, so theoretically they can deal 4.5 damage per energy.  In practice however this is never the case, since finding three targets spaced exactly right is nearly impossible.  If targets are placed closely together, you generally are looking to take out a hub, and blow up all upstream buildings that way, rather than hitting them each with damage.  Cluster Bombs deal splash damage like Bombs, but only about .5 damage to each nearby target.
Cluster Bombs have two primary situational advantages over Bombs - better scouting ability, and negating up to 3 AA Towers.  Scouting with Cluster Bombs means taking high-powered shots into the darkness, where their spread will reveal far more terrain than a single Bomb.  Negating AA Towers simply means firing towards a cluster of 2 or 3 AA Towers, and watching all of them work to destroy your tiny projectiles - leaving you with a clear shot the following turn.  Note that Cluster Bombs do not split into 3 parts until they reach the apex of their flight - it is entirely possible for a single AA Tower to destroy a Cluster Bomb before it splits if you fire it such that the apex occurs within the AA Tower's range.

AA Tower: your most important line of defense

AA Towers are incredibly important, and you will probably fire more of them than any other building type.  Because you want so many of them, however, AA Towers are also the shot type which depends most upon "hub space" - which is the concept of how many buildings can be reasonably fit in and around a single hub.  Multiple hubs able to protect one another really makes your defense stronger, mostly because the increased "hub space" allows better placement of AA Towers.
AA Towers have a range, and will attempt to shoot down the first thing that comes within that range.  Moonbase Commander applies velocity to all projectiles, and although AA shots are very fast, it is possible for a shot to hit the ground/its target before the AA projectile can reach it.  It is rarely beneficial to have more than one AA guarding a hub, because multiple AA Towers simply encourage Cluster Bombs, and add little additional benefit.  However, throwing out a second AA Tower in response to a shot which was blocked by a tower the previous turn is a classic move.  Since AA Towers take 1 turn to reload, your new one takes up the slack.  And if your opponent fires and gets blocked by that new tower, your first one is ready again - creating a nearly impenetrable defense.  Opponents need to skip a turn, then use a cluster bomb to exhaust both towers.  That's when you drop a third...
AA Towers have 3 health so they are destroyed by most projectiles, or two Cluster Bomb hits.

Towers: when you really want to know what's going on around you

Towers provide the best line-of-sight in the game, and all for only one energy.  The drawback of a Tower is the fact that it is a building, and thus competes for hub-space with your other potential buildings.  Firing a Tower at maximum range gives you unparalleled visibility, but it also provides a glowing trail right back to your hub.  Towers have 3 health, so they are easily destroyed by opponents.  But since Towers are an investment of only 1 energy, you aren't falling behind energy-wise when they get destroyed.  Your bigger fear is that during the midgame someone will use your Tower cord as a way to get a Virus or Spike into your building network.  Towers can be very useful as a way to "check" to make sure a given location can support a building and cord.

Repair: not something you want to do every day

Repair heals a single health to an object for 1 energy.  Since shots tend to do 3 damage each, repairing your buildings is generally a losing proposition.  However, Hubs and Energy Collectors have 5 health each, so repairing them twice can sometimes mean letting them survive another shot.

Bridge: only for crossing water?

Bridges create large square platforms which sit on top of water.  They are easily destroyed, but are essential for allowing your cords and buildings reach across rivers and other bodies of water.  Bridge sections have only 1 health, and losing a section of bridge generally means losing part of your network, so I advise that you only cross water when it is absolutely necessary.  However, bridges have a second non-obvious use - as scouting devices.  When bridges hit anything besides water, they disappear without effect.  This makes them invaluable for testing a particular shot angle or power setting when you cannot afford to make a mistake.  Also, bridges fired onto an underwater crawler will bring that crawler to the surface, where you can more easily track and destroy it.
It is occasionally useful to create a Bridge where you need to land an EMP - especially if your opponent is putting shield up in the water.

Second Tier - 3 Energy shots:

Missile: the inaccurate general's best friend

Missiles deal 3 damage, and track the first target they see.  Missiles are probably the most popular type of shot for players who have yet to master the aiming system in Moonbase Commander.  Used in this way, Missiles are like 3-cost Bombs.  But even for a player able to drop a Bomb with pinpoint precision, Missiles are useful tools.
Firstly, Missiles are great for shooting down enemy Balloons.  It is possible to take down a Balloon with a Bomb or other projectile, but it requires superhuman reflexes and a healthy dose of luck, so Missiles should generally be considered the only option.  Missiles, once they start to track, also fly faster than most projectiles so they can be used sometimes to skirt through the edge of an AA Tower's range and hit their target before being shot down.  And sometimes a first-turn Missile can blindly find an enemy Hub.
But the most awesome use for Missiles is when you use them to shoot around corners.  All other shots must travel in a straight line from the firing hub to their eventual destination, but Missiles can be fired askew, and then allowed to track back to their target in an L-shaped pattern.  This takes advantage of the fact that Missiles do not begin to track for targets for a moment, so they can be fired at very high power past an AA Tower, and then "wake up" in time to see the Hub beyond, and bend around to strike it from behind.

Balloons: your eyes in the sky

Balloons are by fare the most popular scouting choice in Moonbase Commander, even though they cost 3 energy and reveal significantly less area than Towers.  The reason they are so popular is the fact that they float above the battlefield, and thus take up no hub space.  Balloons may be shot down by AA Towers while they are being deployed, but once in the air they cannot be shot down except by a tracking Missile.  (Or a thousand-to-one Bomb shot)
Since Balloons require 3 energy to destroy, losing a Balloon doesn't put you behind energy-wise.  Balloons and Missiles are so closely connected mechanically, that it even makes sense sometimes to place Balloons around your base specifically to attract any Missiles which might try to shoot around the corner of your defenses.
Balloons as a first-turn move are very popular.

Spike: attacking a cord directly

Spikes are probably my favorite shot type in the game, because they attack Cords rather than actual buildings or units.  They deal 3 damage if they hit a building or unit directly, but they can also be targeted at cords.  When a Spike hits a cord, it sends a pulse in both directions, dealing some damage to the building at either end of the cord.  The damage done seems to depend on the distance the energy pulse has to travel - generally you can expect 1.5~2 damage in each direction for a medium-length cord.  You can hit a building directly for 3 damage and still send a pulse "upstream" for additional damage.
Spikes are incredibly easy to hit with, since they are effective over tons of area.  They also allow you to strike otherwise protected regions.  Spikes even scout for you a bit, since you can follow the energy pulse through unexplored areas, to find out where the hub is.  Spiking Towers is great fun - I'll actually heal up a tower so that I can Spike it again and again without destroying it.  People hate that.

Mines: your first unit

I've occasionally used the phrase "building or unit" when describing things which can take damage.  There are lots of buildings, but many games can go by without a single unit being deployed.  Mines are your first type of unit, and Crawlers (at 7 energy) are the only other type.
Mines fly and deploy exactly like Cluster Bombs, and they deal similar damage if they impact something directly.  Mines seem to have a larger splash area than Cluster Bombs, though this is only very marginally useful.  When Mines strike bare ground however they become explosive units.  Any shot which lands on or near a mine is destroyed.  Mines have 1 or 2 health - it's hard to say exactly.
Mines can be used to discourage opponents from expanding in a particular direction, though that's a rare use for them.  Mostly I use them when I have to make a tricky shot towards multiple targets.  A Mine near a building can later be exploded for additional damage - making your Bombs seems more powerful.  Mines also eat up hub space, preventing opponents from deploying more AA Towers to defend themselves.  Mines can make good walls against Crawlers, too.

Reclaim: removing objects for a rebate

Reclaims are deployed exactly like Bombs, and they deal 3 damage.  But in addition, they increase your next turn's energy generation by 1 for each point of damage they deal.  This means that Reclaims can ideally act like free Bombs.  I say ideally, because each time you miss, or fail to hit a target with all three HP remaining, they act like more expensive Bombs.  And even under ideal conditions they are costing you extra energy this turn for a long-term savings.
Interestingly, Reclaims deal 5 damage to your own buildings, but still net you a maximum of 3 energy.  But they allow you to remove any building in 1 turn, which is occasionally useful.
Reclaims only give a rebate for each point of damage they actually inflict, so reclaiming an AA Tower with 1 health gives you only 1 point of energy next turn.
One of the most important advantages of the Reclaim is the fact that it does not detonate explosives.  Mines, Crawlers, and Energy Collectors all explode violently when destroyed.  But reclaiming them prevents the explosion.  This makes Reclaims one of the most important defenses against Crawlers in the late game.

EMP: resetting the battle for a turn

EMP bombs deal about 1.5 damage when they hit directly, but you almost never aim them directly at a building, because it is the blast radius which really matters.  Everything within the gigantic blast radius of an EMP becomes disabled for one turn.  Disabled Hubs, Offenses, and AA Towers cannot fire, disabled units will not explode if destroyed, and disabled Shields will fall.  EMPs are one of the only good solutions to enemy Shields, and the blast radius of the EMP is only slightly larger than the range of a Shield, so a careful shot is required.
A rare but sometimes critical use of the EMP has to do with defending against a Virus attack.  A building shut down by EMP cannot support a Virus, so any Virus attempting to move through that building will be terminated.  It is also true that disabled Energy Collectors will not explode when destroyed.  Generally you would want to Reclaim a threatened Energy Collector, but if you don't trust your aim an EMP will absolutely substitute in a pinch.

Third Tier - 7 cost shots:

Hub: your most essential asset

You start with one Hub, and you'll make many more in a typical game of Moonbase Commander.  Hubs allow your network of buildings to grow - giving you access to more Energy Pools and giving you better angles of attack on your opponents.  Since your opponent must destroy your Main Hub to win, you really want to make sure that they encounter another well-defended outlaying Hub rather than stumbling upon your Main Hub.  Since your Main Hub generates 7 energy a turn, you always have the option of laying out a new Hub - but you should probably scout out the location first, as a 7-cost mistake can be deadly.  Hubs have 5 health.

Offense: long-range warfare

The Offense is a special Hub which can fire weapons but not new buildings, and fire them twice as far.  Sometimes an Offense is exactly what you need to break a stalemate, as you fling Spikes, Crawlers, and Mines over those front-line AA Towers and Shields in search of the opponent's Main Hub.  This is not generally something you should be considering in the early game, although occasionally a first-turn Offense can fling Missiles to destroy your opponent on turn 2.  Note that Offenses have only 3 health, making them much more fragile than Hubs.

Energy Collector: gaining energy advantage

Energy Collectors give you one additional energy each turn, for the significant investment of 7 energy up front.  If placed on or near an Energy Pool, a collector will generate a total of 3 energy each turn.  Learning to get 3 Energy Collectors on a small pool is the sort of trick you need to practice (hint - use more than one Hub!) to really excel at Moonbase Commander.  Exposed Energy Collectors are very tempting targets for opponents, because they create a long-term energy advantage.  You will almost always want to place a new Hub near a group of Energy Collectors, and ideally you would ring them with 2 or more Hubs for maximum defense.
Energy Collectors explode in a nuclear conflagration when they take lethal damage - which means that if you lose one, you might very well lose all nearby buildings as well.  Reclaim damaged Collectors, or use EMP to render them harmless.
Sometimes you can use the explosive nature of a collector to your advantage, buy throwing one into the middle of an enemy base and then detonating it with Spike attacks on your own cord, or by reclaiming the Bridge supporting its cord across the water.  Your enemies will hate you.

Shield: perfect defense costs a pretty penny

Everyone loves a shield - in one fell swoop you completely nullify all attacks over a large area.  But Shields are very expensive, and sometimes almost trivial to eliminate.  So you need to be very careful with how you use them.  Your opponent's first thought will be to drop an EMP near the edge of the Shield, so getting an AA Tower up to intercept an EMP is almost a necessity.  Placing an AA Tower behind your shield, to intercept any attack one a successful EMP finally hits is also a good idea.  Shields have just 3 health.
Remember that Shields also block your shots, so it can be very difficult to heal buildings or eliminate units under the Shield.  Placing your Shield on water makes an EMP attack very hard indeed, but it further increases the difficulty of actually defending against Crawler attacks.  Remember - a dedicated attack on your Shield will eventually succeed.  The key is to discourage attacks with your Shield, to refocus your opponent elsewhere.

Crawler: the endgame doomsday weapon of choice

Crawlers are the second type of unit, along with Mines.  Crawlers land like Mines, and face the direction they were fired in.  They then move about half a screen's length after each turn, and explode when they hit a building, cord, or unit.  This explosion deals ~4 damage to everything in a wide area, and can set off a chain reaction of multiple Crawlers.
Crawlers will move underwater, where they can be very difficult to see.  By the time you see a Crawler coming at your base, it's often too late to do anything about it other than maybe a perfectly-aimed Reclaim.  The blast radius on Crawlers is enormous.
Crawlers have 3 health, which means that you need a direct hit with a Bomb to eliminate them.  On the other hand, it means you get full rebate with a Reclaim.  Damaged Clawlers explode for just as much damage as health ones, so you don't get any advantage from half-measures.  Disabled Crawlers can be destroyed safely, for what that's worth.  Mean opponents will fire Crawlers at short range, and follow up the next turn with a longer-range Crawler, so that both arrive at approximately the same time.  If you see a Crawler in the distance, a Missle will ensure you hit it decisively the first time.

Virus: subtle scourge of large networks

The Virus is a strange weapon, and never a first choice.  The Virus hits any building for 1 damage, and then spreads to all adjacent buildings the next turn - eventually dealing 1 damage to every building in the network.  In addition, a building is disabled during the turn it contracts the Virus - allowing players to disrupt careful Shield and AA Tower networks from anywhere in the building network.  1 damage is a pittance, but it does place all of your buildings within the range where a single Crawler could detonate them.  It also puts Shields and AA Towers into the range where a nearby Spike hit will destroy them.
In an early version of Moonbase Commander, damaged buildings would smoke through the fog of war, so a Virus could be used to scout an enemy's entire network.  This exploit has since been patched away.
I believe it is technically possible to Virus a Crawler, disabling it.  I can't imagine why this would be worth 7 energy, but there you go.
There is also a subtle interaction if you get multiple instances of a Virus into a single network - when the Viruses "cross" one another, they reset their list of infected buildings, allowing them to go back and re-infect everything they have already touched.  In theory a few Viruses could criss-cross endlessly and destroy everything.  As noted, the only way to eliminate a Virus is to EMP an infected building.  This is often worthwhile if you can do it right away.

Hardcore Players have Tunnel Vision

I've said many times that people, in general, do not know what they want.  Or more precisely, they do not express what they want clearly.  For many people, this is a frustrating inability to explain why something is unsatisfying.  For hardcore gamers it takes a very different and much more dangerous form - it takes the form of specific demands.

I've seen hundreds of forum posts demanding that developers make some change to their game - maybe increasing the cost of a Siege Tank, or speeding up the Flight power in CoH.  Maybe someone's decided that the Lightning Gun is only for campers - or that throws are cheap.  Maybe they even have a lot of people agreeing with them.  Maybe they have math backing up their claims.  But fundamentally, none of what they say matters - because none of it is really on topic.

I went to a really interesting talk about unit balance back in 2000 or thereabouts at the GDC.  The best part of the lecture was a specific anecdote the presenter shared about one of their games - I believe it was a Command & Conquer game.  (I had not played the specific game, so the unit names might be inaccurate.)

The Anecdote:

People complained that the Russian units were way too weak - specifically that the Russian Tanks were too weak when compared to the American Tanks.  We hadn't found that to be the case at all, so we pressed to find out what they meant.  Russian Tanks are stronger and slower than American Tanks - but they appeared to be well balanced for their cost.  It turns out that people thought they were too slow.  We could have sped them up, but that would have ruined the balance we had tried to achieve.  But players kept complaining that, specifically, Russian Tanks were too slow to catch American scout vehicles.

We thought this was fascinating, because of course the slow tanks couldn't catch the nimble scout vehicles - who would think they should?  The Russians had a Terror Drone unit which was fast and perfect for chasing American scouts - players should just mix those into their Tank groups.

But this suggestion led to some stunned expressions - Terror Drones they asked?  We never ever build those things because they are useless.  Terror Drones are too weak and too expensive to field.  So we looked at Terror Drones again and discovered that the players were exactly right - they were much too weak for their cost.  So we improved the Terror Drones, and magically the complaints about Russian Tanks disappeared.

The point is that although players were responding to a real problem in the game - they couldn't identify it.  But because these were clever players who were used to clear communication, they came up with a solution anyway.  A bad solution - one which maybe didn't address the real problem.  We were lucky that we pushed so hard to find out what was really wrong - because it made the game a lot stronger.  If we had had less time or less effort to put into the problem, we probably would have ended up just increasing the speed of the Russian Tanks, and never addressed the root problem.

World Leaders Rankings

I really enjoy playing diplomacy at worldleadersthegame.com.  It's a very clean implementation of Diplomacy, and I love the messaging system.  It has a simple point system for winning or performing well.  Each game gives some number of points, and they are summed together into a Player Score.

Some players (many, even) complain bitterly about the scoring system - because they dislike the fact that mediocre players who play many games can have higher (in some cases much higher) scores than excellent players who have only two dozen games under their belt.  These vocal players seem to want a scoring system which will identify the best players to all, and expose the improper play techniques of the people who currently have the highest scores.

The immediate problem, of course, is that winning a large % of your games is not necessarily a better goal than playing in a large number of games.  In fact, Diplomacy is such a social game I wonder if winning at all is a reasonable measure of how skilled someone is.  The data as presented now makes no claims to represent anything other than "player score."  By normalizing and crunching that data into a ranking system, we eliminate any sort of interpretation, and make score a function of rank only.

Of course, score can already be used as a function of rank - that's exactly the position these hardcore players are taking.  But the fact is that score is a dimensionless quantity - it can be interpreted many ways.  I'd love to see even more data associated with score, such as finished games, dropped games, number of e-mails sent, player-votes for best ally - all the sorts of things that might be interesting in a game of Diplomacy.

Now this isn't just a post about how we can all ignore requests - it's also a warning about how we approach problems.  The error here is not stupidity, or zeal, or ignorance.  Rather, the error is finding a solution before you've really spent time to identify the problem.  That's not just something you find on forum pages - it's something you see in politics, in debate, and in many face-to-face arguments.  Problems are solved though careful identification of root causes - not clever math proving that one solution or another is justified.

Development Stories #2 - Simon lied to me, the Bastard!

One week while Pipeworks was working on our first Godzilla game (Godzilla: Destroy All Monsters Melee) a strange debug error started blocking people's work.  The game would periodically crash, and up on the screen would be the words:

    "Simon lied to me, the Bastard!"

For obvious reasons, people came to me to find out what the root of the crash was.  But I had no idea.  Everyone was befuddled for a few days, and the crash came up not infrequently, which made it hard to test the stability of the game.

But after a few days, Mark Santesson came back to work, and immediately took credit for the unorthodox debug message.  Here was the story he told of how it came to be:

In our game, giant monsters would fight one another and generally wreck the city.  Making the city crumble, burn, and explode well was a major focus of our time.  To that end, we had defined certain buildings as "exploding" buildings - since instead of crumbling into dust and chunks they would explode when taking lethal damage.

When a small building exploded, it was determined that the resulting fire should alight the responsible monster's foot.  We had a "burning foot" animation that was pretty funny, and this was a great chance to use it.  Mark was implementing this feature.  All he had to do was determine which monster (if any) was responsible, and then attach a temporary flame to the foot, and call the proper reaction state.

But Mark had a problem - he couldn't tell exactly where to put the fire.  All out monsters were very different in size and shape, and Mark was working hard to examine all the skeletal nodes and figure out which one had been closest when the building was damaged - it was complicated, and not easy to figure out.

So I helpfully suggested that we just always attach the fire to the left foot.  All monsters had a node named Lfoot01, which was the foot all monsters held while hopping in the "hurt foot" animation.  Mark was skeptical, because relying on data in that way can be dangerous and lead to difficult to track down crashes - but I assured him that it wouldn't be a problem.

Well, it turns out that one monster (Rodan) had named his node FootL01 instead of Lfoot01.  Whoops.

So any time Rodan would step on an exploding building, the game would crash.  Once we realized what was happening, an animator just re-named his foot node, and everything was fine.  But from that day on, many people around the office would affectionatly refer to me as "a liar" and would always point out how untrustworthy I was.  Only a few people actually understood the problem and subsequent fix, but everyone remembered that error message.

Why management decreases the quality of the product

Before I even begin let me offer the following disclaimer: it's not necessarily bad to decrease the quality of a product.  The reduction in quality I'm going to explain is simply the cost of increasing reliability and transparency.  That might be a very reasonable goal for a given studio at a particular time.

So with the understanding that there can be very reasonable pragmatic reasons for hiring management staff, let me explain why they always reduce the quality of the final product.

Reason #1 - Managers are not experts

If a lead programmer needs to express something to an art director, it's most efficient if those two people simply talk to one another directly.  Putting anybody in between them simply causes a game of telephone.  It's like having your lawyer present for every conversation you have with someone else - except that a lawyer has specialized skills which may help in the conversation.  Managerial intermediaries are not experts in programming or art design, and so they cannot represent the people in the conversation.

Reason #2 - Managers create intermediary goals

Most projects have milestones which need to be achieved.  Managers often track progress of a project with granularity much finer than the milestones - to make sure that all developers are on track.  But this process creates additional internal goals which the managers set.  If developers are not given the broad picture, then they lose time by working on smaller bite-sized goals instead of working for the larger goals of the project.  This is an inescapable conundrum, because if the managers do _not_ create smaller internal milestones, then the managers are simply acting as publisher bandwidth again, and we have problem #1 again.

Reason #3 - Managers are too nice

Most managers are decent people, and they realize that a lot of their interactions with developers waste time which could otherwise be put towards the project.  And as a result, managers often try to minimize their interactions - which frequently allows them to miss or gloss over burgeoning problems on the development side.  This is another devil's choice for managers everywhere - you can reduce the amount of time you drain for the team, or you can increase your drain on the team in exchange for a better chance of identifying problems.

In Summary...

So is there any way to protect yourself from these pitfalls?  No. 

Small development teams escew management in fovar of greater autonomy and agility.  As the teams grow, they eventually require some investment money from some source which requires that there be records of how and when money is spent on the team.  To track those costs, managers come in to observe and record the processes of development.  Quality goes down, but investments go up.  Eventually developers become so large that they collapse under their own weight.

Like dinosaurs.  Because bigger is generally not better.

Considering: Desperados 2

I absolutely adored Commandos: Behind Enemy Lines back in 1999, but I was very sorely disappointed with the two sequels.  (The stand-alone expansion to the first game was excellent if a bit short.)  I'm pleased to inform people that Desperados was just as good as the first Commandos game, and Desperados 2 makes everything better.  So why has nobody played it?

For those unaware, Desperados (and Commandos) are isometric tactical games where you command a squad through various mission objectives.  Mostly the game is about moving and eliminating enemies undetected, but sometimes you get to cut loose and blow things up.  The major mechanic of the game is keeping out of sight, which is visualized as a series of animated "Vision Cones" for each enemy.  You are able to see exactly where the enemy can and cannot see you as they turn their head and patrol in a regular pattern around the map.  At some point, nobody is looking at point A, which is when you run across one or more of your squad to achieve some objective.  Most of the time, there is no such opportunity, so you have to eliminate one or more of the  guards - without alerting the others.

Both games gave you a maximum of 6 operatives to control, though most missions had you controlling fewer.  What made the original Commandos so excellent was the orthogonal design of the six commandos.  Each had unique strengths which played off the strengths of the other commandos.  Tiny, for example, could run very fast and kill an enemy silently with his knife.  The Sapper could lay a bear trap, which would silently kill an enemy who walked over it.  Fins had a harpoon gun, which could silently kill a single enemy at up to 8 meters.  Any of those guys could take out a single enemy, but you needed to use all three in parallel to take out larger groups.

These sort of games involve lots of (failed) experimentation, lots of saving, and lots of excited glee as you successfully take out another sentry.  They are also certainly hard.

Commandos 2 went off the rails because it homogenized all of the abilities.  Each commando still had a few unique skills, but the skills were less distinct from one another than in the original game.  Worse, every commando was given the ability to punch, leave cigarettes (as a distraction) and hold an enemy hostage - which sort of eliminated what made Tiny and the Spy so exciting in the first game.  The game also moved to a sort of faux 3D - which never really worked.

Desperados 2, on the other hand, avoided all of those pitfalls, and managed to put together a truly excellent package including stealth, action, subterfuge, humor, and crazy wild west tropes.  Desperados 2 looks just like its isometric predacessor, but the whole game is actually 3D, and the gameplay encourages you to occasionally shift into third-person to get a better view of the action.  This is especially important in the gunfights, but it's useful in many situations.  The keyboard commands for shifting are very well implemented, and the transition seems natural.

Technically, Desperados 2 also did a great job of organizing the data for each level.  It takes a minute or more to load a level, but once you've loaded it moving between saves from that same level is a breeze.  The game even organizes your save games by level, so you can easily jump back to old levels to replay them - without getting too cluttered in the save/load menu.  That's excellent design, because you'll be loading old saves every time you die - which is a bunch.

When I get a really tasty bit of design goodness in my head, I'll review it for pleasure while I'm riding my bike around town - things like the unit attributes of Starcraft, or the spells in Astral Tournament.  The character abilities of Desperados 2 fall into that category.  So please indulge me while I lay out my musings on each character GameFAQs style.

 

John Cooper

John is the leader of the band, and one of the characters you'll end up using the most.  John is quite fast, and strong enough to knock out almost any enemy with one punch.  His stealth is average, and he has the ability to carry a body to a more discrete location.

Weapon: Colt Revolver.

  John's revolver is probably the best all-around weapon in the game.  It has decent range, fires six shots before reloading, reloads quickly, and produces average noise - which is a lot of noise.  Best of all, John aims very quickly, so you'll get accurate shots quickly with him - making him the Desperado of choice for gunning down multiple enemies.

Ability: Throw Knife.

   Throwing John's Knife is the only way to kill an enemy silently from range, and you'll be using it in every mission where you want to kill, rather than stun.  John has to walk up to the body to retrieve the knife after each throw, so you generally only get one shot per group.  Drawing and readying the knife takes a second or so, so this is something you'll need to anticipate.  It is also possible to miss with the knife - though rare if you take the time to throw when your target is stationary.

Ability: Use Knife.

   John can also use his knife to cut things in the game - usually ropes and saddles straps.  He can also run up and kill an enemy with the knife - but punching is usually a faster option, since you don't need to ready your fist first.  John can even kill an unconscious enemy with the knife - though if he's already unconscious why would you need to?

Ability: Musical Watch.

   John can place his watch on the ground, set the volume, and start a timer.  10 seconds later the watch will start to play music - pulling attention from anyone around.  Enemies will sometimes move to investigate the sound, but usually they'll just turn and look in that direction for a few seconds.  This is great for disrupting enemy sight lines.  The watch does not trigger an alarm, even if a guard finds it, so it's a great tool in your stealth missions.  John must retrieve the watch before he can use it again - but he'll get it back between missions even if you leave it somewhere stupid.

John's Role:

   John is your go-to guy for running up behind someone to knock them out.  He can't tie up unconscious guards, so you'll need to follow up with Kate, Doc, or Sam.  His watch means he can isolate guards, and carrying bodies means that he can clean up after himself.  You can use John almost exclusively in some levels and succeed - though as you become more familiar with the other members of the crew you will use him less.

Kate

Kate is the classic dangerous dame.  She goes to work in a showgirl's dress, fishnet stockings and high heels.  She is often the fastest member of your team, but her strength is poor.  Her stealth is top-tier, which means she can slip into places the rest of the team can't.  Kate can also tie up unconscious enemies.

Weapon: Derringer.

   Kate's tiny gun holds only three shots, and does very little damage if you don't get a perfect shot, but it's the quietest weapon in the game by a long shot, which means Kate gets to engage in gunplay much more frequently than many of the other desperados on your team.  You'll often need 2 or 3 shots to take down a single opponent, so don't get her into a real gun battle!

Ability: Makeup.

   Kate can blow makeup into a guard's face - dazing and disorienting him for a minute.  Kate rarely has more than 2 applications worth of makeup, so this ability needs to be used with discretion.  It acts like a silent, non-aggro, short-range area effect stun.  If I had unlimited makeup, I could win every level using only this ability.  Makeup does not cause guards to lose consciousness, so they don't fall down and thus alert other guards.  Best of all, Kate pulling out makeup does not cause panic the way pulling a knife or gas bomb or dynamite will cause panic.  This means that Kate can use her makeup while standing in plain sight of enemies, and they won't ever notice that anything is wrong.

Ability: Seduce.

   Kate can act seductively, which means that most enemies who see her will not raise the alarm, but instead walk over to her.  Their attention will remain fixed on Kate as well, so it's easy to walk up behind them and knock them out.  Kate can even pull out her makeup and stun them while they are flirting with her.  This ability does not work on soldiers inside of a military base - presumably because the sight of any civilian, even one as comely as Kate, triggers a hostile response.  Kate's walking speed is slowed significantly when seducing, but she can move - leading enemies around a corner, or into a shady alley.

Ability: Faint.

   Kate can cry out and faint, drawing focus from everyone around her.  I never once used this ability to good effect - it's just too similar to John's watch or Hawkeye's War Cry, plus it puts Kate at risk.  Kate is at her best when she's unseen.

Kate's Role:

   Kate is a freaking ninja, despite her ostentatious attire.  Kate doesn't have the strength to knock opponents out as well as John, Hawkeye or Sanchez, but since she can tie them up afterward she gets the nod as my solo infiltration operative.  Kate's quiet steps, quiet gun, and silent disorient attack make her the team's stealth assassin.  But when there is no choice but to be seen - Kate is also your girl.  She's the best of both worlds.  Keep her away from all-out gunfights and she'll prove to be a valuable member of your team over and over.

Doc

Doc is a many of many talents - he can pick locks, stun enemies with gas bombs, heal people, and get dirty in a firefight.  The second level of the game is a solo mission where Doc needs to single-handedly infiltrate and rescue two other members of the crew.  Doc is the slowest member of your crew, has poor stealth capability (though not as poor as Sanchez), and hits very slowly - though heavily, since he swings his entire toolkit bag as a weapon.  Like Kate, Doc can tie up unconscious enemies.

Weapon: Bluntline.

Doc's weapon is similar to John's, but with slightly longer range, longer reload time, and more delay between shots.  Doc also doesn't get the aiming bonus Sam does.  Doc's weapon also wins the award for the loudest firearm in the game.  So Doc is great for starting firefights, but will need backup to finish against multiple opponents.

Ability: Snipe.

   Doc can add a longer barrel to his weapon, and turn it into a sniper rifle.  This weapon has a very loud report, so it will draw aggro from a huge area.  But it has range enough to shoot across the entire map, and is perfectly accurate.  This is a great weapon to use over the shoulder, when you have multiple targets to take out.  Doc has a very limited number of bullets for sniping, so you need to pick your targets carefully.  That said, I often get near the end of a mission and discover that I have a few shots left, and I use them to fast-forward through the last few guards.  Attaching and removing the extra barrel takes quite a while, so you'll need to anticipate using this ability.

Ability: Gas Bomb.

   Doc's Gas Bomb is totally fantastic.  He can lob it quite a long way (low ceilings can be a problem!) and when it lands all enemies within the modest radius fall unconscious.  Enemies thus affected get red stars - which means that they remain unconscious for the entire mission, so you don't need to tie them up.  The Gas Bomb is one of the most consistently useful tools in the entire game - many missions are accomplished by locating groups of enemies you want to gas, and then looking for other gas canisters you can pick up along the way.

Ability: Heal.

   Doc can use bandages to heal himself or the rest of the squad by a decent amount.  This ability is only useful if you take damage - and generally that shouldn't happen.  But it is good to have insurance, I suppose.

Doc's Role:

   Doc is all about sniping and gas bombs.  Consider how many of each he begins with, and then try to work your way around the map to collect any more which are available.  Some missions are all about putting Doc in a good vantage point so he can snipe problem enemies as they show up, and other missions are about keeping Doc close so he can pick locks and gas enemies.  Doc is a very poor choice for a melee combatant, because he is slow and loud compared to most of your crew.  But his attack does a lot of damage, so he won't fail you in that regard.

Sam

Sam is the member of the team I use the least - because he has almost no stealth-related abilities at all.  His only contribution to a stealth mission is the fact that he can tie up unconscious enemies.  But when things get hairy, Sam is a powerful force of destruction.  His dynamite attack alone lets him rack up the sort of bodycount in a few seconds which most of your team will never match over the course of the entire game.

Weapon: Rifle.

   Sam's rifle has about the same range as John's revolver, it's about equally loud, but it fires up to 12 shots before it needs reloading.  This makes Sam a good choice for guarding a corner - taking out patrols as they come around.

Ability: Dynamite.

   Make no mistake, this is Sam's signature move.  He can throw a stick of dynamite, which will bounce and roll and eventually detonate into a gigantic explosion which will kill or disorient most enemies on the map.  Enemies who see the dynamite in the air will begin to panic and flee - so precision aim isn't usually a huge concern.  This can NOT be used in close quarters at all.

Ability: Nitro Can.

   In the original Desperados, Sam had a snake in a bag, which he would leave as a trap for the unwary.  Well, now his trap is a can of nitroglycerin.  Sam can shoot the can to make it explode - killing all nearby enemies.  This allows greater precision than the dynamite, but that precision is only really useful in stealth situations, and this sort of attack (a rifle shot, plus a huge explosion) breaks stealth right away.  If you can pull this off, you could just shoot the guy instead - which would cause less commotion.

Ability: Bomb.

   This is an ability which can only be used in very specific missions, in specific areas where you are allowed to place the bomb.  Bombs trigger scripted destruction, such as collapsing walls.  You'll be told when to use this ability.

Sam's Role:

   It's too bad that Sam has almost no generally useful skills - he's all about loud + messy solutions.  You just don't have enough tools to use Sam at the beginning of a mission, so he's generally relegated to a few choice explosions at the end of the mission, or just sneaking along in the back of the pack uselessly.

Sanchez

Sancez is the strongman of the group.  He is a bit slow, super unstealthy, but strong and tough.  Sanchez can wield gatling guns, carry bodies, and knock people out better than anyone else.  He also has a few versatile stealth options.

Weapon: Shotgun.

   Sanchez uses a double-barrel shotgun.  It has the shortest range of all weapons, and gets only two shots.  But in exchange it does the most damage of any weapon - killing almost any enemy in a single shot.  If you need to fire on tough opponents, Sanchez is your man.  The two shots can be fired in rapid succession, and reloading actually doesn't take too long, so Sanchez can certainly hold his own in a gun battle.  His high health also makes him the perfect point man, since he can take a few bullets without much trouble.

Ability: Throw Rock.

   Sanchez can throw a rock to stun an enemy.  The effects of this vary - but usually the enemy remains standing with 2 or 3 stun stars over his head.  Other guards will notice the rock, and raise an alert - so this is about disabling a single guard, who has his back to a wall or is otherwise inapproachable.  Generally you follow up by knocking them unconscious with a melee strike, but you could choose to sneak past.  The rocks follow a physics model, so in some cases you can miss, or strike multiple enemies with a single rock.  Rocks are limited (like most abilities) but tend to be plentiful compared to makeup or gas bombs.

Ability: Tequila.

   This is a pretty useful ability, but not in the way I think it was supposed to be.  Sanchez places a flask on the ground, and any enemies who see it will walk over and pick it up, drink it, and quickly become drunk.  If a few enemies see it, they will fight over it.  That's the theory, anyway.  In practice, I found that I would place it on the ground, and then knock out anyone who approached it, while their attention was focused away from me.  It's like Kate's seduction ability.  Or, in missions where lethal solutions are acceptable, I would use the tequila to lure enemies into the beartrap.

Ability: Beartrap.

   This is self-evident, and awesome.  Sanchez places a beartrap on the ground, and the first enemy who steps on it dies silently.  Sanchez then moves the body, re-sets the trap, and repeats.  This makes Sanchez your go-to guy for patrolling enemies, though unfortunately there are only a few missions where you really get to go beartrap crazy.  In the original game Sam had "Snake in a bag" which did pretty much the same thing - but that's because Sam was designed to be an analog of the Demolitions man in Commandos, and Sanchez gives better thematic flavor to the Beartrap here.

Sanchez's Role:

   Sanchez's best attributes are related to his strength - he knocks enemies into red-star unconsciousness, so you don't need to tie them up afterwards.  He picks up and moves bodies much quicker than John or Hawkeye can, so he can clear a scene better than anyone else.  He has very high health so when you need to do something dangerous, he can survive it.  Those attributes already make him a valuable member of the team, even before you factor in his excellent and versatile abilities.  Unfortunately, Sanchez is your least stealthy member - the sound of him running is louder than many gunshots.  And he's not especially fast, so it is difficult to use him to knock people out when they aren't distracted.  This by itself can incline players to use John and Kate over Sanchez, but once you learn his quirks you'll find that Sanchez can often get the job done faster all on his own.

Hawkeye

Oh Hawkeye - as the new member of the team I had such high hopes that your abilities would be interesting.  Hawkeye is not available until after the 9th mission, so you really have to work to get him.  But all you get for your trouble is a super speedy + stealthy character with a hodgepodge of abilities you already had.

Weapon: Bow + Arrow.

   I don't know about this weapon yet, as I haven't used it.

Ability: Tomahawk.

   Don't know.  Probably a knife-like ability?

Ability: War Cry.

   This is Hawkeye's primary use - it's essentially a copy of Sam's musical watch ability, but it centers on Hawkeye, and triggers instantly.  Enemies turn and face Hawkeye, and occasionally move to investigate, but eventually lose interest.  So it's good for turning heads, and little else.  That's good enough though in many cases, because Hawkeye is excellent at getting into strange places stealthily, so you can pull attention away from whoever is going to knock the guard out.

Ability: Heal.

   Hawkeye can heal using herbs.  This is of limited use, just like Doc's heal.

Hawkeye's role:

   Hawkeye is pretty much a slightly faster and stealthier Kate, but with a melee attack on par with Sachez (He red-star stuns most enemies) , and none of the distraction abilities which make Kate and Sanchez so versatile.  He's not bad by any means, but he feels unnecessary.

Working towards the mutual good

I've heard people say that selling goods and services to meet demand cannot be an evil act - that the market is fundamentally amoral.  But I don't believe that.  Because people desire things which are, fundamentally, not good for them.

Now obviously I can't speak for any individual person.  But collectively, humans make mistakes all the time - and very frequently those mistakes are enabled by people looking to make money.  This symbiotic relationship quickly turns parasitic, as those entities motivated by profit begin to find ways to market and distribute their products more effectively - changing from a passive to an active element in the decision making process of their clients.

I recently tried to participate in a very juvenile discussion of prostitution.  Someone asked "What's (morally) wrong with prostitution?"  Their essential point was that it is legal to give money to people, and it's legal to have sex with people - so having sex for money should also be legal.  The great majority of people seemed to agree with this logic.  And, indeed, prostitution has been an enduring institution in all civilizations for the past few millennium.

But just because one can imagine a victimless form of prostitution does not mean that prostitution is generally victimless.  In fact, data about actual prostitutes tends to show that prostitutes are forced into their line of work through poverty, abuse, drug addiction, or some combination thereof.

Looking at a different subject - I can easily imagine seeking a position where I lived in a fancy estate to work on my novel.  I would get free room and board, but receive no direct financial compensation.  In exchange I'd work on my writing, and help keep up the house some during the days.

I might really enjoy that sort of lifestyle, but in most of the world we call it slavery.  Not because working without pay is always evil, but because once you allow work without pay, you quickly end up on very shaky moral ground.

Even if you aren't clearly evil - you're not a pimp, you don't traffic in human beings, you don't sell drugs, and you're not in marketing - these are still important considerations for a lot of people.  Because many of us have the power to work towards the mutual good, as opposed to passively working for a morally-blind marketplace.

Do you prepare food?  Is it food that actually sustains people?  Do you use local ingredients?  Are you able to adjust your menu to fit demands in creative ways?

As a game designer are you encouraging thoughtful, expressive play?  Are you creating something that will maintain interest in an intellectual way?  Does your game give players anything to discuss?  Does your game promote social interaction?  Is your game honestly educational?  (Games don't have to be aimed at children to be educational - Civilization, for example, contains lots of educational elements)

I really want to enrich the discussion around games and game design, and I think a big part of that is acknowledging the cultural relevance of games.  Games have an obligation to promote the mutual good, just as film does.  That doesn't mean that we don't have room for the occasional mindless shooter or tasteless fighting game - but I like to think that we can promote goodness in our audience through the choices we make as developers.

Statement of Intent: I want to lecture at the GDC

Someone asked me a few months ago what drives me - what do I want to accomplish with my life?  My answer was that I wanted to contribute to the general ability of people to discuss games and game design.  That's pretty nebulous, but it means that I should at least be out there, with at least a few people listening to what I have to say.

And here's a more concrete goal - I want to lecture at the GDC.  I've been to lots of GDC presentations, and I have a good sense of what makes a presentation good or bad.  I have lots of insights I'd love to share with people about game design, and I have a certain amount of dramatic flair which I would apply to my presentations.

I remember going to Clint Hocking's first GDC presentation, which was titled "The interesting thing about bishops..."  I was tremendously impressed and started corresponding with Clint, among other people.  Clint had just recently entered the game industry from a writing background, and based on his ability to clearly express himself and compose questions, he rose quickly in the Ubisoft heirarchy.  His GDC presentation almost every year is not to be missed.

My point is that just a little bit of exposure goes a long way, and the GDC is a great place to get exposure as someone who wants to help shape the way games are discussed.

So, starting this year I'm going to try my best to present something at GDC.  I'll try again every year regardless of my level of success the previous year - for at least 10 years.  Mark your calendars.

I know many people who have had lectures rejected from the GDC, and I think a key element to getting your talk scheduled is to make it very accessible.  (Alternatively you can be famous, or a lead developer on a top game from the previous year.)  So "orthoginal unit differentiation" might not get you the nod from the planning committee, but "six easy tips to insure your project makes money" might be a popular lecture to attend.

What should my lecture this year be?  I'm sorely tempted to do something on the different types of system equilibrium - but I think I'll stay away from that unless I come up with a better soundbite to describe it.

Turning the World

I have two children, ages 9 and 6.  Not surprisingly, they've been exposed to lots of video games.  But they don't really play many games on their own yet.  I've been teaching them some vital skills, and it's been very interesting to see what comes easily and what doesn't.

The first vital skill they had to learn was moving a character with a control stick.  My daughter mastered this one first, and she and my younger son would play Twilight Princess together.  She would hold the nunchuk, and run around with the stick, and my son would swing the controller (to attack) whenever an enemy came near.  It was awesome, and it reminded me of how my sister and I used to play the space sequence in Ultima 1 together - I would control the cursor (it took both my hands to control the arrow keys) and she would hit the space bar to fire.

I've been finishing up the achievements in Prototype this week, and my son has learned to run around, jump and glide in that game.  He's not especially skilled, but just holding the R trigger and running around looks pretty awesome, and he likes to jump off high buildings for a few minutes now and then.  I let him generally walk us towards our next objective, but sometimes he can get pretty turned around.

So today he gets turned around, and he says to himself "I'll just turn the world" - and he proceeds to adjust the camera with the R stick.  Camera control on a separate stick has been one of the big sticking points for my kids - because when I say "camera" they don't understand me at all.  They don't think about the camera as an observer.  But all of a sudden today, my son was turning the camera around like a pro.  But that's because he learned that the R stick "turned the world around."

I feel like my son has discovered the equivalent of non-Euclidean geometry, because the fifth postulate of video games - that you watch the world through a camera - was too complex for him to take into account.  (Did I lose too many people with that reference?)  But honestly "turning the world" actually works pretty well as an isomorphism of orienting a camera - at least when the camera is close to your character.  And once you've learned those skills, the specific metaphor you used to teach your muscle memory probably won't matter.

Considering: Prototype Movement Controls

I'm a huge fan of the Incredible Hulk, and so I was pleased to play through "Hulk" by Radical, which came out with the Ang Lee movie in 2003.  It wasn't a great game really, mostly because the hulk was fighting in tiny rooms with indestructible walls, floors, and ceilings.  There were also the dreaded Banner stealth missions.  Not the finest moment in gaming.  But the game's saving grace was a great story, including really fantastic bosses which came straight from the best of the comics.  When I started fighting Madman (from the amazing "countdown" four-issue storyarc) I knew that someone on the development team really cared about Hulk.

Two years later, Radical came out with Hulk: Ultimate Destruction, which I believe stands as the best Hulk game ever.  I went to their talk about their design process at the GDC that year, and their philosophy was simple - if the Hulk is moving, everything gets out of his way.  And it was really gloriously true in the game.  Some people hated watching the Hulk sprint up buildings, but I thought it was fantastic - really an inspired bit of gameplay.  The jump on release scheme was also great, because jumping is the Hulk's signature movement style, and it was great to charge jump after jump.

The story in Ultimate Destruction was not fantastic, but clearly somebody tried really hard.  Paul Jenkins - who was a regular Hulk comic writer for about a year, was credited with writing the game.  This was a poor choice, as Jenkins wasn't really a traditionalist.  His story arc on Hulk was pretty bland and forgettable.  Shockingly, Peter David (the best Hulk writer ever) was comissioned to write the in-game hidden texts, and he wrote a four-part comic explaining the Abomination's new origin story, to come out simultaneously with the game.  If only Peter David had been writing the main story, instead of only the extras...

Radical did not make the Xbox 360 follow-up "The Incredible Hulk" - but it's clear that game was based on the same engine and assets as Ultimate Destruction.  Unfortunately, that game failed because the story was totally bland (they tried to give an "expanded" version of the movie plot, which just didn't hold together well at all.) and the Hulk's movement powers were weakened.  Now the hulk climbed buildings by punching holes with his hands - which makes more sense conceptually, but isn't nearly as fun.  Even more damming was the fact that I got the hulk caught on the corner of a park bench while sprinting.  I held the hulk running in space for several seconds - not believing that Radical's core philosophy could have been so completely thrown out.

But now on to the matter at hand!

Since their 2005 release of Ultimate Destruction, it seems that Radical has been working on an amazing follow-up of their own design.  Something to free them from the design constraints of the Hulk - let their imagination really run wild.  And unfortunately, that game turned out to be Prototype.

Prototype

Unfortunately, Prototype is almost exactly a clone of Hulk: Ultimate Destruction, with only a few new features added to the mix.  Unfortunately again, all of these new features turn out to be less cohesive than the feature set in Ultimate Destruction, and the game feels muddled and directionless as a result.

Let's start with the story - such as it is.  Our story is told in two ways: through between-mission cinematics, and through 130+ videos from the "web of intrigue" which you unlock by hunting down 130+ individuals who each know a piece of the story surrounding you and the game world.  The cinematics are the worst type of "Stop where you are!  Get the fuck away from me!  Gwuahhh! Alex, we've got to get you out of here!" style scenes.  They make the mistake of trying to tell us about what we're doing next, rather than contextualizing our actions.

But the biggest sin of the game are the movement controls.  Ultimate Destruction had a really great system and clearly Radical remembers how great it was, because it's all right here just as it was.  Triggers are sprint and lock-on, A is charge and jump-on-release, B is grab/throw, and X and Y are light and heavy attacks.  Sprinting in the air does an air-dash, and through unlocks you eventually have two air-dashes, bigger jumps, and other enhancements such as grab while running.

But Prototype adds a new element - gliding.  Gliding is a great idea, and a lot of fun.  But unfortunately it throws the balance of the old system out of whack.  In the old system Hulk had a certain amount of air control during his jumps, and his jumps were huge.  Prototype gives you somewhat smaller jumps, because you can glide to cover the horizontal distance.  But since you have an all-horizontal tool at your disposal, air control is less important during the jumps, which makes jumping overall less important.  Similarly, series of jumps charged with the A button in the air are totally unnecessary.  But those jump sequences were the whole reason for jumping on the release of A, rather than the press.  You held A to charge up your jump.  And now gliding is activated by a second release of A, which never feels natural at all.  (If jumping and gliding were both activated on press, I think the game would have scored a full point higher in reviews) 

Finally the air dashes, which were so essential for Hulk, are similarly mitigated to near-irrelevance by the glide.  The resolution in Prototype was to make air dashes quicker, but hard to orient, while gliding has a very slow turn rate.  So instead of one all-purpose tool we have two, neither of which really does what we want.  

Sigh.

Finally, Prototype takes the concept of "weaponization" from Ultimate Destruction, and just ruins it completely.  The Hulk could grab an object and turn it into a weapon he could wield.  Cars became iron fists, buses became shields, cranes became whips, tanks became clubs.  It was great fun, and more than just a little useful!  I loved searching for a bus when the helicopters would come after me, and discovering a crane was always a great bonus, because the whip was incredible.

But Prototype simply gives you all these weapons as mutations of your body.  So instead of making weapons part of the world around you, there is a new menu (and more buttons) to fiddle with to set your weapon at any given time.  That sounds awesome on paper, and I'm certain it was a key part of the pitch for Prototype, but in practice it ends up being much less compelling.

Now I don't mean to say that Prototype is a bad game - it's not.  Most significantly, I think it's a fair bit better than Infamous, which came out at the same time and was directly competing for market share against Prototype.  But it's clearly not as good as Hulk: Ultimate Destruction, and that makes me sad.  It makes me worry that their Hulk games were excellent because of luck, rather than skill.  And I really want to believe that good designers can be consistently good.