Her Story is a short story told in a pleasantly unorthodox way. Unlike Gone Home, which was touching but straightforward, Her Story is actually architected in a way that ties its mechanics to its premise.

The simple, central conceit of the game is piecing together the story of a crime, by keyword searching through archived interrogation clips. What makes it a "game" is that each search is limited to five results, so even if you know a keyword that will reveal the whole plot, you won't get very far until you can formulate a more specific search. So the game is a genuine quest to forensically ascertain key story details, by gradually looking farther into whichever rabbit holes you're able to find. And this process of organically stumbling upon story beats is satisfying in a really unique way.

Unfortunately, Her Story doesn't conclude on a high note; actually, it doesn't conclude much at all. At the risk of spoiling it, the point at which the game prompts you to "end" is pretty arbitrary. And even after this point - for that matter, even after viewing every clip in the archive - there are a couple of tantalizing subplots that aren't adequately addressed. Sure, it's fun to speculate about the main mystery, but I feel like-- if this was a real police investigation, then a lot of important information has gone missing.

Although the end is a bit of an immersion-breaking let-down, the (brief) ride there is a fresh, fun take on the mystery game. I think it demonstrates a lot of potential for more, and more sophisticated, investigation-driven games in a similar vein.

Better than: Gone Home
Not as good as: BioShock Infinite, I guess?, I don't fuckin' know what's comparable.
One specific complaint: The "Delete Session Data" function deletes more than just "session" data. That was annoying.

There is a very short list of complaints I have against The New Order: The two-timelines mechanic is not compelling enough to make me start a second playthrough; collectibles and upgrades in previous chapters can be missed forever (which, again, not quite compelling enough for a replay); and two or three encounters felt unfair, requiring a specific set of tactics and maneuvers to survive.

Otherwise, and in general, this is an incredibly well-made and captivating game that just happens to be a first-person shooter about killin' Nazis. It's one of the finest story-driven games I've played, with an impeccable presentation all throughout. And the gameplay, from knife-throwing to rocket-launching and everything inbetween, is consistently tight and rewarding.

This is the caliber of game everyone should play. Except those who get squeamish about graphic evisceration and dismemberment, I guess.

Better than: Spec Ops: The Line
Not as good as: BioShock, if only because of that game's more-open game world.
I've never felt quite so emotionally attached: to a game character who just killed hundreds of people.

Progress: Completed on "Bring 'em On" (normal) difficulty.

In 2013, I glogged about playing 87 games across 151 posts. And in 2014, 112 games in 145 posts.

In 2015, I only managed 29 games in 53 posts.

Sure, I'm in the best shape of my life, have an exciting new job, and even rearchitected this blog. But at what cost?

If there's any semblance of "theme" in my past year's gaming activities, it's re-treading old ground. Despite my low game count overall, several of them were games I'd already played -- revisiting Red Faction: Guerrilla, the remaster of Shadow Complex, the PC version of GTA V, even SpaceChem for some reason. Not to mention Broken Age's second half, and Legacy of the Void finally finishing off the StarCraft II saga.

And then there was Assassin's Creed Rogue, which felt quite a lot like I was just playing Black Flag again. And Saints Row: Gat Out of Hell -- but now I'm falling down a rabbit hole, aren't I.

Breakdown of games played last year:

• PC: 26
• Wii U: 2
• DS: 1 (still only Mario & Luigi: Bowser's Inside Story left)

The good:

• Grand Theft Auto V on PC finally fulfilled the grand crime sandbox's destiny.
• TIS-100 (fucking hell, still in-progress) provided me with a whole new way to hate my stupid monkey brain.
• Wolfenstein: The New Order (in-progress) has, so far, been more satisfying than I expect a "simple" shooter to be.

The bad:

• Assassin's Creed Rogue, a clearly-unfinished attempt to spin off of Black Flag, utterly disenchanted me with the entire Assassin's Creed franchise.
• Broken Age, though good overall, was a letdown in its second act.
• The Magic Circle (demo), which purports to lampoon game production cliches, was itself a game production cliche.

The ugly:

• Batman: Arkham Knight was a great game, but a technical travesty.
• Oceanhorn: Monster of Uncharted Seas aped a lot of good Zelda features, but was also a bit of a technical mess.
• StarCraft II: Legacy of the Void was mechanically great, but made a mockery of its story.

Things I'm looking forward to in 2016:

• Majora's Mask 3D, and all the other stuff I just didn't get to last year.
• Borderlands: The Pre-Sequel, Mad Max, and the, like, ten million other on-sale games I just bought on Steam.
• Upcoming releases of The Witness, Rise of the Tomb Raider, Uncharted 4, hopefully some kind of immortality serum.

Prime factorization really took it out of me. I admit it! I need high-level programming. Ugh. That was painful.

That being said, without the benefit of seeing my scores above my friends' on an easy-to-read graph, the prospect of revisiting my solutions for optimization is not quite appealing enough. I guess my optimization ego is not as much of a "pride" thing as it is a "jerk" thing. Oh, well.

Progress: Finished all levels, not most optimization challenges.

One "trick" in SquareCells was shifting from a fill-based to a clear-based system, i.e. whereas most picross games allow free marking of non-filled squares, SquareCells is the opposite. Unfortunately, since SquareCells lacks a "try it out"-style system, it's difficult to distinguish between squares that are definitely filled and squares which are only hypothetically filled based on some deductive guess -- making it impractical to play such hypotheses, as opposed to just marking stuff and seeing if the game allows it.

Some puzzles made really good use of combinations of clues, hopping back and forth between clue types to deduct how a block of squares was supposed to fit together. But other puzzles, especially in the last set, were over-reliant on playing so many moves ahead that it was indistinguishable from randomly guessing and seeing if it worked. Which isn't very satisfying.

And in the end, with only 36 puzzles, the good in SquareCells wasn't really enough to outweigh the annoying.

Better than: Gemsweeper
Not as good as: Picross e
And another thing: Leaving the Unity default logo as your application icon? Pretty amateurish.

Progress: 108/108 stars.

I spent... oh, wow. Thirty hours? On these puzzles. That's a lot of puzzle.

No regrets.

Better than: Picross DS
Not as good as: Well, it doesn't have as many puzzles as Picross DS, but I still think Paint it Back is better due to its ultimate puzzle complexity.
What would be even cooler: Is DLC, and/or Steam Workshop integration for community puzzles. Yeah, that would be pretty dope.

Progress: All puzzles complete, 329 medals.

I never got into the "pure" FPS sub-genre -- your Dooms, your Serious Sams, what have you. So it's a good thing that Wolfenstein: The New Order actually has a lot more than mere "purism" going for it. Sure, yeah, it has assault rifles and linear levels and you fight a bunch of Nazis. And the gameplay extras don't really add up to much -- some health upgrades, some skippable weapon enhancements and perks. But to me, the real story in New Order is ... uh. The story.

Even in the game's introductory WWII sequence, its narrative feels like the star of the show. B.J. Blazkowicz has a burdensome past (...which I had to research) and his compatriots build up personalities of their own very quickly. The characters B.J. interacts with throughout story missions show real senses of humanity, and the events - in both historic backstory and real-time gameplay - are genuinely interesting and well-written. And the soundtrack is just phenomenal, a great combination of ominous ambience, thrilling action riffs, awe-inspiring environment cues, and chilling alternate-history Imperial Germany versions of 1960s pop music.

The fact that the gameplay sequences inbetween New Order's expertly-directed cutscenes are all about killin' Nazis is just icing on the cake.

On normal, "Bring 'em On" difficulty, I'm not usually having any problems with said Nazi-killin'; although I am worried about how many retries it took me to get through my most recent bout with armored supersoldiers. At this point though, it's still worth it just to see what's next for Captain Blazkowicz.

Progress: Just stole a train.

More so than TIS-100, Human Resource Machine makes me appreciate the fact that C is a high-level language. Because while TIS-100 (as with SpaceChem and Infinifactory before it) upturns programming fundamentals to build a game about concurrency/parallelism, HRM is actually more like a real, tiny but fully-functional computer. And it will remind any computer science student of what it's like to take conditional statements and variables for granted.

(Specifically: HRM provides an unconditional jump, and jumps when zero and negative, but not for a positive case; so creating logical blocks of code for this can make for a very messy jump diagram. And while most of HRM's puzzles include memory in the form of addressable locations, managing all of the requried "copy from" and "copy to" instructions manually gets pretty exhausting.)

So while Human Resource Machine is firmly in the category of "metaphorical programming games" - which I'm incredibly happy about - it's still fresh in the context of more abstract games like TIS-100. Ignoring Tomorrow Corporation's surreal semi-dystopian setting, this game's levels feel a lot like real-world assembly programming challenges.

But there are some problems with HRM that bring the experience down somewhat, mostly in its UI. Contrary to TIS-100's limited space for writing instructions, HRM has a scrollable area for lengthy programs; but this doesn't scale well with the later levels, making it difficult to navigate jump destinations. The game supports adding comments to code and labels to memory locations, but forces you to write them with a mouse, instead of just typing letters on the keyboard. And while the "run" speed can be slowed down and sped up, even the fastest setting is way too slow for the last few levels, taking whole minutes to run through nested memory-scanning loops.

Still, despite these flaws, Human Resource Machine is a pretty fun programming game. What interferes the most with its value proposition is that it's fairly short. Although I'm totally stumped on an optional level for prime factorization - which is just, I don't have to tell you about that - the rest of the game only took me about three hours to plow through.

I'll be spending some more time on that, though, and on trying to meet the optimization challenges in other levels. This may not be as much of an ego-stroke as literally comparing my performance numbers to my friends' (as the Zachtronics games do), but, my pride requires I do it anyway.

Better than: Little Inferno
Not as good as: TIS-100 -- but maybe just because I'm a masochist
And the ending: Is a satisfyingly humorous send-off.

Progress: Finished all but Prime Factory, missing a bunch of optimization challenges.

Did I mention that I fucking love picross? Because I do. And that's what makes SquareCells so confusing: It's sort of like picross. But it's different in ways I'm not really sure about yet.

SquareCells mixes up the traditional nonogram ruleset in some pretty fundamental ways. It shows the total number of marked and unmarked squares for each puzzle, so you know how many there should be in the end; it tosses in numbered blocks, which indicate how many marked squares are contiguously connected; it throws in bracketed row/column numbers, which indicate how many squares are marked, but not in what arrangements; and as a result of these new rules, sometimes row/column markers are blank, leaving you to figure the context out from other surrounding clues.

On paper, these sound like challenging and fun twists to traditional nonogram rules. But in practice, I've had to rely on guesswork once or twice already. And while trial and error isn't exactly anti-picross, it's a disappointing crutch to lean on in smaller and simpler puzzles. Of course -- maybe I just haven't gotten the hang of the new rules yet, and how they interact with each other. Maybe my guesses were actually unnecessary.

What makes this more frustrating than it might otherwise be is that - unlike Paint it Back - SquareCells penalizes you for unmarking/removing squares that are supposed to be part of the puzzle. So in addition to reducing the number of stars you earn from completing the puzzle, it goes ahead and reveals what you should have done instead. Which seems like a futile exercise, since if you miss a square you're obviously just going to come back later and re-do the puzzle for full stars. It discourages careful analysis of the puzzle board in favor of, well, trial and error.

I don't know. I'm still settling into it.

Also -- this is an unabashedly short game. Only 36 puzzles! It's hard to get excited about that, in the face of Paint it Back's deluge of huge fucking puzzles.

Progress: 19 of 36 puzzles complete, 57 stars

I fucking love picross! Nonograms! Shit is tight. And apparently I'm not the only one -- yes, we are a cult in very high esteem/notoriety/infamy.

Anyway. Paint it Back's simple aesthetic belies a stunningly rich amount of puzzle content. This ain't no bite-sized Picross e slapjob; there are, uh, 152 puzzles? That I know of? Plus a mode that throws out random puzzles, which, I don't know how many of those there are. Maybe an infinite amount? No, that's not mathematically possible. But probably a fucking lot.

And PiB doesn't shy away from complex puzzles, either. At about the halfway point, I'm already doing puzzles larger and more intricate than the endgame boards from Picross DS, my previous record-holder for most intimidating nonogram. There are some 40x30 boards I can see in the back. My brain is salivating all over itself for those.

These puzzles are at their best when they're sprawling and meticulous, forcing a patient approach to inductive logic and pattern recognition. And this game is picross at its best.

Progress: 79 puzzles complete, 151 medals