The combat is a little more sophisticated than is apparent in the demo: there are a few weapons to swap between - notably, the warhammer's slow, brute strength changes the melee mechanics somewhat - and situationally-available guns that are pretty great, like a laser cannon and a heat-shotgun. And some enemies do actually demand different tactics, ... but this is where the game can get a little messy.

As the campaign marches on, you'll encounter enemies who can do increasingly ludicrous damage from range: devastating mortar launchers, disturbingly strong machine guns, lightning laser beams, and grenade spam will annihilate you handily unless you're well covered. Granted, this isn't "supposed" to be a cover-based game, but while it's easy enough in the early chapters to charge into a horde of enemies and start mopping them up, later on this tactic is suicidal.

It doesn't help that Space Marine's execution-system, which is built to allow you to stay in the thick of melee combat, is fatally flawed in the presence of enemies who can do great damage, and/or who can attack from range. Captain Titus can refill his health by stunning and then executing enemies, but he will still take damage - and can even be killed - while the (lengthy) execution animation is playing. So while you're cutting one guy's head off and waiting for the health bonus, another nearby enemy might shatter your health bar with a single large attack, or a half-dozen dudes with guns might just whittle you down unopposed.

In the last wave of the waves-of-enemies-based final boss battle, I actually died while executing an enemy because another enemy just dropped a grenade at my feet. Since the execution takes so long, and there's no way for me to move, or cancel it, there was absolutely nothing I could do. (And since the game rarely saves checkpoints between waves, I had to start back at the first one again. Which I did about fifteen times, before giving up and restarting the chapter on Easy mode.)

When the game works, though - that is, when the number of enemies you're fighting is manageable (or when you have a jetpack and can just dive-bomb them) - it can be pretty fun. There is a little depth in terms of choosing which enemies to toss bullets at, when to charge in, and how to balance melee combos between stunning and finishing enemies. But in general it derives its simple joys from running headlong into foes and smacking them around.

Also, the campaign is pretty short, which is really to the game's benefit -- it already feels like it's stretching its premise a little thin. As long as your expectations aren't too high, you can make a fun weekend out of it.

For a studio that's only known for its real-time strategy games, this is a fairly competent first attempt at an action/shooter formula, and it's interesting to think of how they'll evolve their ideas in the inevitable sequel. But within the action and shooter genres, Space Marine is generally unremarkable. Its controls are solid, but it too frequently avoids genuinely interesting combat encounters by throwing out wave after wave of enemies. And while the guns are fun to play with, the lack of a proper cover system significantly limits your opportunities for dealing damage without dying instantly.

Better than: Conan (2007), Dark Void
Not as good as: Darksiders, Vanquish
Launching without co-op: is kind of stupid, right? Who's going to care about this game anymore in October?

Progress: Finished on Normal (except the final boss, on Easy)

I'm going to rattle through a few bullet points:

• A broad and ambitious production
• It is pretty cool how open-ended the mission structure is
• The narrative, with all its player-choice twists and foibles, is interesting but comes across as incomplete
• The melee system is begging for some added depth
• Making my initial accuracy shitty, and keeping it shitty unless I invest in it, isn't okay
• The enemy AI is a little baffling
• The mouse is beyond the comprehension of most of the game's menus
• The hacking minigames ... are pretty neat
• Boss fights ... [are] retarded

All of these things are true about Deus Ex: Human Revolution -- but I actually wrote them back in February, regarding Alpha Protocol. I found myself recalling Michael Thorton's adventure frequently throughout DXHR. In many ways, the games are extremely similar: generally speaking, because they are both broad-reaching games with great potential, which is fulfilled enough to make the game fun - under the right circumstances - but is also obvious in its deficiencies.

More precisely, both games purport to give the player complete choice over how a mission is tackled, while at the same time leaving many mechanics (specifically, gunplay and melee) unbalanced and underpolished enough that it isn't really much of a choice. Stealth works okay in Human Revolution, and upgrades throughout the game make it work better, but straight-up shooting is almost never a good idea; even with his Dermal Armor fully upgraded, Adam Jensen is no better at taking bullets than Batman is. And unlike the Dark Knight, Adam Jensen has a pretty difficult time escaping once he's been spotted.

Also unlike any other video game protagonist, Jensen is really not built for melee combat. He's got a pretty sweet takedown, to be sure, but it uses a full bar of energy each time -- energy which maxes out at five bars; and once a bar is depleted, it won't regenerate unless it's your last one. So the energy economy makes Jensen's robotic arms mostly useless.

As for non-metaphorical guns, weapons like pistols, shotguns, and rocket launchers are upgraded not with character/experience points, but with upgrade kits which you'll find and purchase throughout the game. Without a laser sight, most weapons are atrociously inaccurate; this especially hurts in the game's introductory mission, before Adam has any augmentations, where you must take on a strike team using only an assault rifle that is basically incapable of hitting anything from range. As the game goes on, you can upgrade and get better weapons (a fully-upgraded pistol, with armor piercing and a silencer, is almost game-breaking); but early on, it's depressing how ineffective your armaments are.

So the stealth is alright, but doesn't become really impressive until it's upgraded; gunplay is terrible at first, but warms up later on; and melee isn't an option at all. There are a great deal of alternate paths in each mission area, between multiple entrances and exits, secret tunnels and vents, and safe rooms that skirt patrolled hallways -- so there are a lot of ways to accomplish a task, even if there aren't a lot of different mechanics to exploit in that process. I would say that, in terms of how you play the game, the choice of route is more meaningful than the choice of augmentation or load-out.

On that note, while some of the augments are pretty cool, too many of them function only as keys. That is, if you want to hack high-level consoles, you need to invest points in your hacking "capture" level, or it won't let you. If you want to finish a side-mission that blocks your progress with a gas-filled chamber, you need the lung augment that makes you immune to the gas, or you'll simply die trying. If you want to reach a high ledge and there are no crates around, you need the leg augment to increase your jump height. This is sort of like a choice of playstyle, but too often it's not a choice of how you tackle an objective, but a choice of whether you tackle it or don't.

Of course, if you go to the effort to do side-quests, and uncover secrets which yield experience rewards (e.g. by hacking), you'll end up getting almost all of the game's upgrades before you're done. And many of them, maybe a third or as many as half, are so situational that they're borderline useless. Increased sprinting speed and duration are convenient, but don't help the gameplay at all; the social-skill aug to peer into an NPC's personality is not really any more helpful than just reading and listening; there is an augment to use a takedown on two adjacent enemies at once, but this only happens a handful of times throughout the entire campaign. I only got these augments because I ran out of better things to spend points on.

What I would like would be for the starting augments to be more powerful, so you can see more of the game's mechanical diversity in the beginning; and for experience-based upgrades to go deeper into these mechanics, so that a player could become a master ninja, or a master soldier, or a master brawler, depending on how Praxis Points are allocated. But high-power augments that would support the idea of "mastery" don't really exist in the game.

One thing that Deus Ex really knocks out of the park (especially as compared to Alpha Protocol) is the overworld, in the sense that it kind-of has one. Although the serpentine pacing of the game's story makes their "hub" nature somewhat debatable, Human Revolution's two big cities of Detroit and Hengsha (near Shanghai) are big, complicated, and dense -- filled with interesting environments, hidden items, shops, and side-quests. There is so much to do in Detroit, especially; I suspect that I spent more time exploring the city than I did in the story missions set alongside it.

I do wish that there were one or two more cities, and although it may sound a little ungrateful given how rich the existing ones are, I think some of the level design resources allocated to the game's later missions could easily have been taken for it. Mission maps late in the game are, frankly, excessive -- if your goal is to explore every nook and cranny - since most of the game's nooks and crannies have hidden items to collect - these missions can take hours. Setting aside alternate paths, most of this geography just seems wasteful.

As for the story: it starts slow, builds up some intrigue, takes a retarded twist halfway through, and ends somewhat ineffectually, regardless of which of the game's four endings you select. The stupid twist isn't really the game's fault, except in that it had to build itself into the existing Deus Ex mythos; but the ending, while not exactly bad, doesn't feel impactful at all. It doesn't give further insight into the game's world, and it doesn't make any profound point. At the same time, it isn't offensively preachy or a brazen sequel setup, so at least there's that.

There are a few other nitpicks I want to mention: character animations are awkward, and clearly came from a time before ubiquitous motion capture; voice acting is pretty good for primary characters, but variably dismal for the supporting cast; the game's handling of variable dialog selections is admirable, but still imperfect; handling inventory space is pretty ridiculous, with some big items (like the rocket launcher) taking up entirely too many slots, and some unstackable items (grenades) being too difficult to keep around to really be useful; the automatic saving frequency is far, far too low for how often you'll really want to reload a checkpoint.

Ultimately, though, what I want to say about Deus Ex is this:

• It's initially disappointing that you don't really have that much freedom in the game;
• It's fun -- once you learn how the game wants to be played;
• It's lengthy -- most reviews quote about 20 hours, but having done all the side-quests I could find, I came closer to 40;
• The premise shows a lot of room for improvement, which I can only hope is used in future iterations;
• If you really want to go nuts and start pulverizing civilians, you can absolutely do that.

Better than: Alpha Protocol
Not as good as: Batman: Arkham Asylum
If you enjoyed this: I really think you should try Alpha Protocol, because it does some things much better than Human Revolution does

Progress: Finished on "Give Me a Challenge" (normal)

Not unlike Space Marine, Bodycount's promotional media did a good job of convincing me that this wasn't just going to be another rote genre entry -- that it would do its shooting so well, and combine it with enough unique, flavorful twists, that it would be set apart from FPS juggernauts like Call of Duty and Battlefield. But based on the demo, it, to be blunt, doesn't.

The controls are surprisingly complicated for how typical most of the mechanics are; one notable exception from the norm is its lean system. When you press the left trigger to aim, if you hold it all the way down, you'll be rooted in place and can lean left or right (whereas, holding it halfway will allow you to strafe as is customary). The lean can be pretty useful, but the control is baffling -- when I'm in a tense moment and I jam on the trigger, I don't want to stand still and get shot at.

Codemasters has also been hyping up Bodycount's varied color scheme, which replaces the typical grays and browns with yellows, greens, and blues. And while it's true that the game draws from a rich palette, the high-detail art direction - combined with muddy texturing, which is unfortunate - means that rather than using color for contrast effect, game elements just end up blending in with one another.

As for the destructible environment, this is no Geo-Mod -- only certain things can be destroyed. So rather than a real mechanic, it ends up feeling like a context-sensitive opportunity.

Of course the big question is, since Bodycount is a first-person shooter that goes mostly by the numbers, why is there no PC version? The demo's default aim sensitivity is down way too low, which may or may not be related. It gives me the impression that Bodycount's developers are more familiar with GoldenEye and Halo than with Counter-Strike and Unreal Tournament -- because if they were PC players, they would surely know that the platform has already perfected a control scheme for this kind of game.

Progress: Gave Up -- Finished the demo level

Yesterday when I mentioned that I was putting "about" as much time as I could into Human Revolution, this demo was what qualified the about. The promotional media for Space Marine has made protagonist Captain Titus out to be the ultimate badass: covered in plate armor, bristling with heavy guns, waving around a chainsaw sword, and also, uh, JETPACK. With all that pure, unadulterated kickass just bouncing around all willy-nilly, does it matter if the game is totally retarded?

Well, the demo doesn't make a great case for it. It shows two levels -- one where you fight a bunch of orcs, then a bunch of orcs, then a bunch of orcs, then some orcs, then more orcs, then wave after wave after wave of orcs. The level lasts less than half an hour, but becomes painfully repetitive almost immediately. In the second level, you get a jetpack and can dive-bomb down into orcs, so basically the same thing but a lot shorter and easier.

The high concept still sounds amazing, so I'm not ready to write it completely off just yet. But if the demo's level designs are really representative of the full game, I think I'll stick with watching it, rather than playing it.

Progress: Gave Up -- Played both of the demo levels

I've been putting about as much time as I can into Human Revolution since Monday night, and there are a bunch of things I want to talk about. Most of my thoughts regarding the game are still pretty unorganized. In a nutshell, though, I would say that I'm having fun, but there are a million little imperfections that keep it from being really great. And while I'm not all that disappointed in the game, I am (yet again) pretty disappointed in the publishing-critical complex.

Particularly, there have been several reviews for the PC version which state, quite assuringly, things like "the PC version does nearly everything right." Well, that's a lie -- I would say it's a stretch to call this PC version anything more than adequate, and even that's only in the context of today's AAA PC games being ported from console versions. Loading times are always unreasonably long -- even on my 6-core 3.7 GHz machine, where the game is installed to a SATA-3 SSD. Both myself and a buddy who have ATI graphics cards have had a huge issue trying to change video settings: once we changed the antialiasing or texture sampling at all, the game refused to run until they were turned down to the lowest possible values -- the game complains that our video hardware can't handle it, even when the settings are returned to their defaults, which were running just fine.

And perhaps most fundamentally, mouse controls in menus and other point-and-click situations (like the hacking minigame) are undeniably broken. Hovering over a button is rarely enough to put it in focus; I usually need to hover out and back on again, or click on it a first time - like a pre-click - to make the button selectable. And as a result of this focus problem, clicking on a button doesn't always click that button, which can be especially frustrating when I'm trying to hack a node, but the game ends up using one of my special items on it instead.

The depressing point about all this is, it's really no worse than, say, Mass Effect. That is: while the PC version was very clearly not well-taken care of, and the PC experience does suffer somewhat as a result, it is something that the modern-day PC gamer has become accustomed to. Publishers can say with a straight face that they paid special attention to the PC controls, but it doesn't matter, because we stopped listening to those assholes a long time ago.

Anyway -- I'm going to be complaining about nit-picky issues like this a lot more over the next few days, but the game is still basically fun. I spent my first several hours in Detroit just exploring the city and breaking into people's apartments, and now as I'm progressing in the city's missions, my challenges of finding information on this guy, or locating this gang's secret hideout, are practically already done. That's pretty cool.

Progress: Dilapidated factory

I didn't try the original Deus Ex until way too late, and as such its features and technology just didn't impress me. The choice-heavy gameplay it pioneered has become usual, and its mechanics simply couldn't measure up to modern games. But I can appreciate that, back in 2000, it would have been a pretty amazing experience.

So when I watch trailers and developer diaries for Human Revolution, and it looks so good, I really hope that it can live up to that idea of pushing the medium forward, again. I am slightly worried, though, that it will just be a barely-playable mess, the latest in Eidos's and Square-Enix's last few years of overhyped AAA garbage.

Fingers crossed -- we'll find out on Tuesday.

Dear Goichi Suda,

Please take my money.

Yours forever,
Mirkon

I still come back to TBOGT pretty frequently when I need a GTA fix. Recently, I realized that the extra stuff to get 100% completion was actually fairly simple (at least compared to the ridiculous amount of pigeons, stunt jumps, and other crap in the base GTA4). And, yeah, it was a relatively small amount of stuff, but the majority of it was pretty underwhelming as anything other than notches on the completion percentage.

The drug war missions were neat at first, but become intensely repetitive, since there are so few different mission types. Base-jumping was surprisingly fun, but there are far too few of those, too. The underground fighting matches got stupid once fighters started carrying weapons, and while the special club management missions are pretty cool, actually getting to them - by standing around and watching things for minutes at a time - is painfully boring.

I can't complain about the game's overall quality, given that the core campaign experience, and basic dicking around, is still so fun. Nevertheless, it's hard not to hope that GTA5 will have more meaningful and entertaining extras.

And it's a particularly interesting situation in light of Saints Row 3, given that Volition's franchise is built around those extra activities, and their third iteration looks to have a more engaging (if not serious) narrative than its predecessors.

Progress: 100%

I know, I know, late to the party again. That may go some length toward explaining my lack of interest in Psychonauts. Not unlike Brütal Legend after it, Psychonauts has a really cool theme and some stellar writing, but the gameplay just isn't compelling at all -- with its boilerplate platforming mechanics and countless collectible pickups, this really just feels like yet another Super Mario 64 knockoff. And while Banjo-Kazooie succeeded as a collect-a-thon with imaginative worlds and nonsensical items, it was also brimming with personality; Psychonauts has good characters, but the levels themselves just aren't very engaging.

And, again, maybe I would have been more tolerant of this in 2005. But I simply don't have the patience anymore to wade through tepid, unusually-lengthy platforming in order to learn more about the game's clever world.

Progress: Gave Up -- Didn't finish Basic Braining

I feel for the Humble Indie Bundle guys. It's a good thing they're humble, because most of the bundled games aren't very good -- of course I enjoyed VVVVVV, and Braid which was in the last bundle, so I already own those. And most of the others - And Yet It Moves, Crayon Physics, Cogs, Machinarium, Osmos - are games I've demo'd before, and just didn't find very interesting or good. But I threw some money in for the charity angle, and Steel Storm was a pleasant surprise.

Not that's it's great: the gameplay is simple, the campaign difficulty spikes unexpectedly, and the level design is intensely repetitive. But the bullet- and laser- and explosion-filled chaos makes for a pretty fun time, even when I'm dying as a result of it.

That being said, I don't know that I'll bother coming back to the game for Episode 2 ("Burning Retribution" proper), because even in the first episode's short campaign the primitive mechanics and lackluster level design got stale by the end.

Progress: Gave Up -- Finished Episode 1