Some arcade games make it past location testing without issue; occasionally they need some tuning, or a redesign, but nothing prohibitive. Others fail to meet expectations and get canned. A select few survive this stage despite not passing muster, and that's possibly what happened with Donkey Kong 3. It's a sad yet mundane story of a product unable to meet (let alone surpass) its precursors, all the while eclipsed by the genre competitors it was aping.

Our tale begins further back than the original Donkey Kong, however, as this sequel owes much more to Nintendo's ill-fated vertical shooters of the time, particularly Radar Scope and Space Firebird. Both ancestors offered technically competent but uninteresting variations on the ubiquitous Space Invaders formula. To the latter's credit, the developers had a solid grasp on sound and graphical feedback by this point, with players feeling that crunchy impact while mowing down alien hordes. But the lack of risk-reward features like the degrading shields in Taito's game, or the cleverly evolving raid and bullet patterns of Galaga a bit later, meant that these post-Sheriff shooters were just too spare and derivative to stand out. I can definitely see why Nintendo's last attempt to break into the shooter market went for a mix of platforming, puzzling, and blaster action. And they would have had a winner if not for some critical problems in the pernicious primate's latest title.

The dastardly donk's back at it again…in a greenhouse? I suppose it's better than Congo Bongo's colonialist chase through the jungles. Genyo Takeda and Shigeru Miyamoto help provide the customary trappings of an early-'80s Nintendo arcade release, with fairly responsive controls accompanied by a very readable set of graphics. I rarely struggle to figure out what's happening on-screen even way late into a playthrough, and the SFX ring clearly in my ears. Nothing here feels like an evolution beyond Donkey Kong Jr., though. The likes of SEGA and Namco were already starting to produce more audiovisually stimulating stuff like the former's DK clone and the latter's Mappy. Also lacking here is much of the series' humor—the most I can recall is my amusement whenever DK gets his head stuck in the beehive. It's a rather sober outing for not-Mario vs. not-Kong, contrary to expectations that these developers had previously set. I'd be fine with all this if the game design itself was strong, but, well…

Simply put, the game's a bit too easy, with its difficulty level evolving mainly from faster attacks and a few additional enemy types. Our protagonist Stanley may have a measly little pesticide pump to work with, but the admittedly clever goal of scaring DK to the top means it's possible to ignore the bugs and just go for broke. Arguably the most fun I can have with Donkey Kong 3 is exactly that: speedrunning each board to get as high a time bonus as possible, ensuring I can regularly extend my lives. That sounds fun on paper; now imagine this game throwing you a frequent gun power-up which trivializes combat entirely. Sure, you have to actually lose a life for the upgrade to reappear, and there's still some hits on DK to go before it tumbles. But the relative ease of getting 1-ups via fast stage clears means it's often too easy to get this item. One has to play several stages before there's much threat of bees successfully stealing away with your flowers…which means perfect bonuses become a formality rather than a reward. Combine all these flaws and we've got a game loop which struggles to encourage more skillful play and leads to unsatisfying high-score runs.

The echoes of mediocre starfighting action from Nintendo's pre-Donkey Kong bombs reverberate throughout this final numbered entry, contrary to the heights their R&D teams would reach via many Family Computer releases to come. I feel bad for Genyo Takeda especially since he's so often singled out as the one responsible for playable yet dismal software such as this. To me, it doesn't seem as though releases like this suffered from a lack of foresight, ambition, or no-nonsense design. He later directed the StarTropics games, simultaneously within Nintendo's house style and quite distinct from anything Miyamoto's crew were involved with. (Let's also remember that Miyamoto was involved with Donkey Kong 3, yet the besmirchment's sent elsewhere.) Something managed to go wrong with each and every one of the company's arcade shooters, regardless of circumstances, and it's no surprise that they later played it safe with Solar Striker. If there's anything I'll give Donkey Kong 3 credit for, it's that shooting-heavy action platformers were a rarity in 1983. Add this distinction to the game's brand power and I'm unsurprised that it still sold reasonably well to arcade operators in Japan, no matter the lack of staying power. Still, I'd rather play Moon Patrol; even the best PCBs of '83 struggled to match that one in quality and replayability.

Save for its quick appearance on the Famicom, Donkey Kong 3 fell into footnote status much quicker than the average Nintendo hit of the time, and I attribute that to its shortcomings more than just its small delights. There's still a competent action game in here whenever you're unable to immediately trivialize a stage. I had fun timing my shots to avoid blocking myself with stunned snakes. The way Stanley gets wiped out by an ambush of insects is morbidly amusing. And while it becomes too damn straightforward after 15 minutes, I hardly mind this kind of accessible arcade experience after playing some of the period's more punishing stuff. But in exchange for that ease of completion, I expect way more variation in levels, enemy characteristics, and types of traps to deal with. It's because the game settles for so much less than it promises that I can't help but rate it so low. And as 1984's catalog would show, Nintendo themselves knew that the future belonged to increasingly complex and extensible game loops, even for cartridges as constrained as Excitebike. (Hell, it's a wonder that Takeda's own Sky Skipper gets more retrospective praise than anyone ever expected.) In the end, this gassed-up-the-ass gorilla had reached the end of his usefulness; it would take something altogether new and wild to bring back the beast.

The neat thing about Doom's classic mod scene is how easily you can observe the evolution of game design over time, as well as a community's priorities and preferences; it becomes clear what ideas had lasting power, and which didn't. As Doomsday of UAC approaches its 30th anniversary this summer, this early fan level retains its iconic status despite how unkind time has been to it. The minute I exit a cleverly sculpted wreck of an 18-wheeler, greeted by an industrial park teeming with demons and "realistic" features, that's how I know I'm playing something special.

Doom modding in its nascent days amounted to either (a) poking around in hex editors, trying to create a cogent WAD file through trial and error, or (b) wrangling the initial batch of pre-Windows map editors which loved to crash and/or corrupt your hard work! It's a miracle that something as fun, well-paced, and innovative as this map from Leo Martin Lim came together at all. (The other miracle is that levels this old are still preserved in their original archived form, and we have Ty Halderman and his successors in charge of the /idgames FTP archive to thank for that.) All college kids who played some version of id Software's original Doom in its launch days had many ideas for mods, but so little experience and precedent to build from. Lim wasn't even the first to release anything beyond a series of sketchy boxes with monsters and weapons—Invasion…: Level 1 - Contamination beat it to the punch by about a month, replete with special effects and new assets like music and textures. But Doomsday of UAC proved that you could make a similarly cinematic experience using just the base game and a host of magic tricks exploiting Carmack's engine. It helps that this map has solid combat and exploration in its bones, too.

After crawling from the wreckage, you're pressured from all sides by a trickle of imps, pinkies, and shotgunners ready to pounce before you can assemble an arsenal. This prompts a mad dash through a midnight maze of boxes, trailers, and enemy groups one might want to use against each other. We're far from the clean, abstract but believably efficient spaces that Romero designed for Doom's Episode 1…nor is this anything like the trap-driven dungeon crawling one finds in Sandy Petersen's levels. Doomsday of UAC marked the beginnings of what FPS modding circles call "Doom-cute": heavily kit-bashed replicas of architecture and objects like cars, toilets, etc. using only the original game's resources. The tipped-over truck with a spinning wheel looks impressive already, nothing like what id's crew displayed in the shareware and payware episodes. It gets even more exciting when you realize it took a lot of sector geometry manipulation and a well-placed texture animation effect on certain lines for this to function at all! There's just enough breathing room to admire the scenery while gunning down monsters and collecting the necessary ammo and key cards to proceed.

I shan't spoil the rest of Lim's one and only wonder-mod since it relies on a clever twist or two, but just know there's some trickery afoot, creeping up on players as they head further into this corrupted corporate complex. Hidden usable doors guard access to visible yet seemingly unobtainable power-ups. A conference of powerful baddies lies deeper within the offices, guarded by hapless imitations of the salarypeople who once roamed here. Even the bathrooms and parking garage aren't safe! Considering that Lim and other authors also had to build around the Doom engine's strict limits in this pre-source port era, the level of detail and scope in Doomsday of UAC matches and sometimes outdoes OG Doom maps like Mt. Erebus. I don't know that it exceeds the best parts of id's game, particularly Computer Station or Containment Area, but the ingenuity on display here always brings a smile to my face. Best of all, there's never too much going in the player's favor, nor too little. Weapon and encounter balance feels spot on, the secrets are rewarding to find (nor essential to a fault), and the sequencing of incidental combat into traps and back rarely feels awkward.

Doomsday of UAC is still a 1994 FPS mod, though, warts and all. It's very easy to get through for a modern Doom fan, even those who have only played the official games. Texturing and level of detail is mostly sparse aside from the aforementioned set-pieces. The famous "crystal sector" room can just end up feeling gimmicky or frustrating if you haven't kept a backup save ready. Nor is the original Doom's bestiary and set of player options as ideal for these large open spaces as Doom II's equivalents. I've warmed up to the original's emphasis on cacodemons, rockets, and copious cannon fodder thanks to later WADs like Beginning of the End and Doom the Way id Did, but the general experience for early Doom mods can feel underwhelming if you've played anything much newer. The best moment arguably comes right before the end as you deal with an elaborate cyberdemon trap to nab the red skull key, which involves tangoing with barons of hell and lost souls in the process. This would have been an intimidating puzzle for players of that period, and I get a kick out of it now. But in the back of my mind, all I can think is how much crazier I'd redesign this into, using modern tools like Ultimate Doom Builder and such. At least the transition from "invaded office space" to hellscape remains evocative today.

Overall, I'd say Doomsday of UAC more than deserves its lauded spot in the history of DIY world-crafting and FPS fandom. It features prominently on Doomworld, both in its 10 Years of Doom feature, with its close rival Invasion…: Episode 1 unfortunately absent. [1] Lim's mod was further recognized fifteen years later via the site's Top 100 Most Memorable Maps retrospective, the only map predating Doom II's release to rank in the top 10! [2] And if that's the consensus from community veterans, so often locked in debate over what classic mods and maps truly influenced what, then who am I to downplay the quality and significance of this one? Running through the infested UAC corporate park has become a rite of passage for many players seeking entry into the depths of Doom modding madness. I won't deny it seems quaint and clouded by nostalgia nowadays, yet even Romero himself has highlighted this as an example of the game's impact on future developers, if not id Software themselves. [3] The early success of mods like this and Slaughter Until Death paved the way for id (and competing developers) to hire these amateur designers, or simply license community projects like TNT: Evilution for commercial release as shown with Final Doom. My heart goes out to the unsung pioneers like STONES.WAD; it's simply hard to compete with a milestone like this.

[1] Tropiano, Matthew, and Not Jabba. “Top 100 Memorable Maps 10-1.” Doomworld, December 9, 2018. https://www.doomworld.com/25years/top-100-memorable-maps/page10/.
[2] Watson, Mike, and Andrew Stine. “The Top 100 WADs Of All Time: 1994.” Doomworld, December 10, 2003. https://www.doomworld.com/10years/bestwads/1994.php.
[3] John Romero (20 January 2015). "Devs Play Doom." YouTube. Retrieved 21 January 2015.

"On the other hand, however, there are people who still only use stone. How do they catch game and defend themselves from outsiders? Using only their strong jumping power and stone spheres. They use those skillfully to defend themselves. Here, we introduce two such people. It seems their names are Mario and Luigi. Will they ever learn about sophisticated culture?" These wry words from the game's manual point to the absurdity of Mario Bros., but also the predicament that Hudson Soft's version of the game found itself in. Spread over so many Japanese PC platforms in October 1984, it's a seemingly simplistic variation on the arcade original in a time of rapidly evolving competition from developers at Namco and elsewhere. What more can a few overburdened programmers crank it out before the window of success for this type of game disappears?

Punch Ball Mario Bros. is fun enough despite its platform limitations and unwillingness to go farther with its deviations from the source material; it's also a benchmark for the systems it got ported across. Hudson knew how to get the most from their partnership with Nintendo, bringing conversions like this everywhere it made sense. The weaker but cheaper PC-6001? Yep. The graphically powerful but expensive Sharp X1? Why of course! And it's fascinating to see how much the coders wrought from less powerful PC hardware vs. the best of the bunch, too. I was expecting major sprite and color clash going into the PC-6001mkII version, yet it felt hardly less clean and responsive to play than the Sharp PC varieties. (Considering the X1 has its own sprite-ing hardware, its stiff performance had me underwhelmed.) I expected Hudson to struggle with the FM-7's poor keyboard I/O, however, and that bore true.

See, there's a problem afflicting so many FM-7 games, at least the ones reliant on predictable inputs: the system doesn't fully poll keyboard inputs. This means you can tap anywhere on the numpad (or "tenkey") to move a direction and the game will act as if you're holding that down. But while that first input's still active, no further repeats of it are counted; this doesn't enable a soft autofire, basically. FM-7 keyboard play manages to wedge itself into the worst of both worlds, and I find action games for the system less pleasant for it. The only way to mitigate this is building up muscle memory, tapping a neutral key (ex. "5" on numpad) to reset your movement or switch to another constant action. This habit does help with other versions of Punch Ball Mario Bros., thankfully, since returning to neutral means conserving momentum and, thus, narrowly avoiding critters, fireballs, and other hazards. Sure, the other NEC and Sharp PCs don't suffer anything like the FM-7, but it's ultimately helpful to internalize these controls and work around the lag.

Judging ground and air momentum matters even more here than in Nintendo's game because your main trap-em-up mechanic is throwing a ball, not bumping enemies from below. It's a neat lil' twist on Mario Bros., incentivizing more tactical movement to line up shots and then recover your weapon afterward. The ever-present pipe in the middle provides a route back to the top-center of the action; dropping the stone below resets it atop the POW block. Since the space key fires at neutral and lets you jump when running, the player has to act more deliberately than ever. Your ability to fight at the ground level adds a feeling of empowerment missing in the arcade game, turning this into an action-platforming hybrid without any use for hammers or other power-ups. Where Miyamoto and co. saw their design as leaving players exposed to threats depending on the level layer they're on, Hudson figured their PC audience would be familiar enough with that paradigm to want something different…something aggressive. I could chew right through this game with extra practice, playing faster than the original generally allows.

Compared with Mario Bros. Special, this later port hews closer to what Nintendo must have expected from their new bedfellows, yet the game design comes out subtly different in a manner reflecting different hardware. Punch Ball Mario Bros. doesn't need to employ the myriad gimmicks of its cousin because it's already diverged from the jumping-first model. That's a problem because, as satisfying as it can be to chuck dodgeballs at turtles, the combat never gets any deeper or more explosive. In trying to find the balance between speed, quantity, and quality for their Famicom-to-PC schedule, Hudson ended up compromising most on ambition. Someone's going to accuse me of being unfair, given the obvious additions and reductions in these PC versions, but I just feel like the developers could have gone farther. Awkward co-op via a shared keyboard means these releases needed a lot to distinguish themselves from a trip to the local cabinet, as demonstrated by the excesses of Donkey Kong 3 and Super Mario Bros. Special. Everyone at the company knew that the future wasn't strict arcade adaptations, but remakes like what Nuts & Milk received when it jumped from PC to Famicom.

I won't hold that quality of spareness too hard against Punch Ball Mario Bros.. It's disappointing how slight in new content and mix-ups this version has, yet it still plays well within its constraints. Switching from bumpin' to ballin' no doubt circumvented the rendering and physics problems of the former on these PCs, and it's a small miracle that the end product looks and sounds this consistently everywhere you can run it. Producing this many multiplatform ports is never easy, even back in the bedroom coding era, so it's cool to witness Hudson working their magic in that regard. Most importantly, I think it's telling that Mario fans poo-poo the mid-'80s J-PC ports while conveniently ignoring this one to fixate on SMB1's misfire instead. PC users of the time had seen considerably worse efforts, even from Nintendo's original PC porting partner Westside. The Famicom had undeniable sprite-ing and scrolling advantages over contemporary PCs and Hudson still proved you could bring similar experiences to those machines largely intact. Arcade perfect wasn't the name of the day, but it's surprising and awesome when talented coders got close. Perhaps the worst I can say about this game is that no amount of fidelity would have elevated Mario Bros. beyond the system exclusives and genre revolutions to come.

The Original Sworn Brother Simulator: hours upon hours of chaos, hordes upon hordes of hapless soldiers as blood sacrifice for the honor, glory, and jealous one-upmanship of the Three Kingdoms' ruthless, sexually repressed macho men. Sure, I've heard of the Peach Garden Oath…but how about that whole Shu roster looking like the future winners of Eurovision Asia? Or the way Lu Bu and Zhang Liao glance at each other while riding out to stop Cao Cao? And why do I gotta keep pounding on these guys to get meat buns and stat upgrades outta them?! I take back last month's comments about *Romance of the Three Kingdoms VII—this one's the real Bromance. It also just happens to be a major turning point for Koei, a "safe" brand that ventured out of its RPG, sim, and strategy comfort zones to finally break into the mid-range action market. The big cuddly bear of Japan's game industry started laying down the fight; few knew where musou would go.

I had my reservations going into my Normal skill playthrough of Dynasty Warriors 2, having heard about its jank, slim content, and overall difficulty—these fears were validated, but I'm liking it more after trying part of it again on Hard. Faithfully covering the Three Kingdoms in any action game up to this point meant focusing on the set-piece battles and character stereotypes derived from Luo Guanzhong's classic novel, something which Koei's then youngest division Omega Force (OF) had to translate into mainstream 3D polygon game terms. It's natural that they'd combine the arcade beat 'em up with a light strategy layer. What's special about Shin Sangokumusou are the ways in which it recycles the original Dynasty Warrior, an average versus fighter reborn as the unlikely usurper of trends in beat 'em up design started by Dynamite Deka and SpikeOut. This game ultimately saw much success for OF, its talented staff led by Akihiro Suzuki and Kenichi Ogasawara, and an odd first impression for PS2 adopters expecting something more substantial. These days players tend to view DW2 strictly as a curiosity, and hardly without reason. I just think there's a bit more going on here.

| Yellow Storm |

Romance of the Three Kingdoms looms high in East Asian cultural history, hence all the different ways it's been retold across comics, film, music, and games. (Arguably only Water Margin aka All Men Are Brothers supersedes it in influence—see every shonen manga ever.) In the video game realm, Capcom was first to take a more action-oriented approach to portraying the Three Kingdoms conflict, first with a dramatic post-Dragon Quest series called Tenchi wo Karau (Destiny of the Emperor), and then the pre-Final Fight belt-scroller Dynasty Wars in 1989. By contrast, Koei's grand strategy treatment, initially restricted to Japanese home computers, traded out flash for in-depth systems, letting players role-play as provincial rulers far away from the battlefield. How could these divergent angles meet to make a flashy but accurate, strategic but skillful rendition of the saga?

Omega Force started as Koei's fourth internal division (hence "force", which reads in Japanese as "fourth"), and it saw Suzuki and Ogasawara attempting to solve the aforementioned problem. Though decently regarded in 1997, Dynasty Warriors didn't become the phenomenon they hoped, crowded out by other weapons-based fighters like Soul Edge and Battle Arena Toshinden. (Still better than the results for Destrega…) The team would find more success with WinBack, proving they could bring innovations to 3D third-person action and engage with arcade stylings without trend-chasing. While Suzuki and other veterans from the early-'90s brought their historic knowledge and development experience to the group, Ogasawara represented the new batch of Koei recruits ready to build something console players could uniquely enjoy, rather than PlayStation owners seemingly getting the scraps of PC releases. He saw his colleagues playing RTS games during lunch breaks and was inspired to make a prototype of the archetypal musou game, with single fighters now accompanied by tiered ranks of followers all tied into a morale system which players manipulate to win the larger battle. [1]

It had already taken a nascent Omega Force around a year and a half to make just their first game…once they got their hands on PS2 development kits, there was no time to lose if they wanted the sequel out around launch! (The main team under Koei head Yoichi Erikawa were already wrapping up Kessen for that crucial March 4th, 2000 launch date.) So it goes that Dynasty Warriors 2 narrowly became one of the the system's first 25 releases and proceeded to sell very well. New adopters quickly got a taste of the opening level, a congested assault on the Yellow Turban Rebellion which brought the ailing Han Dynasty to its knees and decorated the careers of many warriors old and young. Pounding heavy metal music, loud and chunky martial arts ambiance, and a tight balance between difficulty and power fantasy all came together despite the odds. It'd be just a year until Koei & OF promptly hurried a sequel to market, knowing they had a new franchise in the making. Critics were impressed that the company had produced anything so unlike their typical history software, feeling more like the work of arcade-based developers rather than home-oriented ones [2].

| Killing Time |

Dynasty Warriors 2 gives you 28 generals from the Wei, Wu, and Shu forces (plus "rogue" players like Lu Bu) to play across 8 battles, with most characters sharing similar story mode progression. Not everyone's available from the start, unlocking as you play with fighters from all sides or achieve feats like earning a thousand kills at Hu Lao Gate. This amount of content hardly seems light until you realize just how many fighters share the same movesets. In musou/fighting game jargon, these are clones with only secondary differences like stat assignment and different hitboxes. Each character's got health, attack, defense, and "musou" stats affecting their abilities in combat, and each variety of character can combo their square attacks up to four moves, or follow that regular combo with a charge move (ex. a "charge 3" (C3) = 3 □ + ∆).

Fighting itself is something Dynasty Warriors 3 and its successors would greatly improve, though I find it fine in this early form. Chaining moves doesn't take a whole lot of practice, or even that much mashing once you've internalized the animation timings for attacks. Actually hitting foes can sometimes be fickle with certain weapons, but that's usually just an issue when trying to C1 (juggle) a gate captain or general. Swap guys often, though, and it'll quickly become clear how busted certain combos of moveset and weapons can become. Let's put aside Lu Bu and talk about Dian Wei. His big axe should slow everything way down, yet he runs about the same speed as most other guys (poor Xu Zhu) and has a broken AF C3 where the axe becomes a veritable boomerang. When a fighter's only flaw is this AoE move not always working as it should (the axe can't fly behind the camera frustum and hit enemies), Now That's What I Call Imbalancing! The gulf in performance between crowd-control gods and the weaker units like Zhuge Liange wouldn't be this pronounced in the series for some time.

Archery and mounted combat are also introduced in this installment, albeit with issues that would plague them up until the later PS2 games. The former involves stopping still and holding R1, transitioning to a slow and clunky first-person view where you can periodically string single arrow shots, regular or charged. In practice, bow play gets the most use against guard tower turrets and bosses you shouldn't approach until low on health (again, Lu Bu). It's too inflexible and awkward to use for anything else vs. closing in. Meanwhile, cavalry charges go haywire half the time you try. Enemies will gladly jump to knock you off, with varying success due to hitbox weirdness. Running into large groups will often provoke your steed into rearing up and losing momentum, which can lead to someone dismounting you. And killing enemies via trampling doesn't net you kills, because only weapons kill people apparently! This sucks for characters like Xiahou Yuan and Ma Chao, masters of archery and equestrian respectively, who don't get to shine despite debuting in this game. At least jumping is still useful for level traversal, even if there's a surprising amount of invisible walls preventing you from hopping onto stairs from the side, among other things…

Grinding's a dirty word in the musou fandom, insofar as no one quite agrees on how much padded-out gameplay exists in these titles. Dynasty Warriors 2 easily earns the accusation. Ho boy can this one become a grind if you're unaccustomed to a couple of things. For starters, the game's mostly designed for players to take a character between their Musou Mode playthrough and Free Mode runs—usually from Easy or Normal difficulty depending on playstyle—before being able to try Hard mode comfortably. This becomes a trend throughout the PS2 games, though they've solved the pacing issues by Dynasty Warriors 5 with mechanics like out-of-battle weapon tuning. So that's fine, it's just more kickass brawling action, right? Maybe, but there's a pernicious little thing that officers major and minor love to do: repeatedly heal up before they die. Every non-generic officer (plus the gate captains) gets a limited but scaling supply of dim-sums and other power-ups, used randomly but frequently when recovering from knockdown. The closer they get to dying, the more likely they use an item! This uses the typical combo flow against players, rather than to their benefit, and can throw a lot of people off. I don't like this system in practice, despite how it cleverly encourages cheesy tactics like weak arrows or refusing to finish combos. It's something I haven't missed going into later games, but there's the seed of a clever idea here.

I'm happy to say the animation quality and kinaesthetics haven't aged much at all, notably when using the game's special musou attacks. Should you get stuck on the other end of a brutal assault, a timely musou can change the outcome in just several seconds. Sword wielders slash through the crowd before launching everyone around them with a single swipe; ax and club battlers pummel unlucky soldiers into a pulp with plenty of knockback; and those cultured fan users basically unleash black magic in a flash. The spectacle's always exhilarating, as is the freedom these specials give players when dueling the nastier opponents. While this game has a subtle auto-lock tendency, influencing your character to hone in on a single enemy at a time, it's much less restrictive than how Omega Force implemented this in the following two games. Dynasty Warriors 2 cares more about making each link of hits satisfying to pull off than keeping the player down for the AI's sake. It becomes tempting to leap around, sucker-punching guys from the air just to watch them fall to the floor as you prepare a deadly finisher. When all else fails, one can always guard attacks by holding the camera face button (L1), which works as stiff but reliable as ever.

Put it all together and the magic starts to happen. Charismatic leaders advance across plains, canyons, farmlands, and fortresses with retinues in tow, with enough numbers to make this world feel alive and bustling. The iconic moments are all here: crunchy impacts as combo finishers send troops flying, crashing into each other like ragdolls; rivals appear from the fog of war to challenge you, either to win and end the match or fall and herald your victory; gates open as officers retreat, traps are sprung right upon your allies' feet, and the cadence of battle hits climax. It's hard to explain how this game, with its aforementioned core flaws and priorities, manages to enthrall me so much. Sure, I could whinge about the genre's perceived repetition or this game clearly being a prototype of things to come, but OF really understood how these games can and should be paced. I can have fun going full 1v1000 with OP fighters like Lu Bu and Dian Wei, but also get a kick out of managing upsets with sub-optimal choices like Sun Shang Xiang or Jiang Wei—the "what if?" attraction of future titles clearly existed this far back. Role-playing beat 'em ups weren't new, but this variation sure was.

| Divine Wind |

Musou games live or die based on their game loop, and strong combat doesn't mean as much without a similarly worthwhile campaign that challenges players, maintains variety, and delivers that sense of martial adventure. Again, here's where Dynasty Warriors 2 shows its budget and design limits, as Omega Force needed to spend precious resources on getting all those scores of units to render at the expense of level design. It arguably wouldn't be until Dynasty Warriors 5 that Koei found an ideal ratio of asset and SFX complexity to level density and novelty. What we do have in DW2 is quite good regardless, as shown in the multi-layered, ambush-laden attack on Zhang Jiao's peasant rebels. Allied and enemy forces always start from their own forts and hidey-holes, accompanied by reinforcement points leading off the map. Most mission goals revolve around defeating the main enemy commander, but raising your side's morale (let alone deflating the enemy's) often becomes most important of all.

There's a common line of debate about old vs. new Dynasty Warriors (and musou in general) about how important or relevant it should be to fight within your army, rather than steamrolling ahead with nary a consequence. I tend to prefer the former myself, something which this game promotes by having very aggressive AI on Normal and Hard difficulties. Later games make the 1v1000 meme real largely by empowering player characters and dumbing down AI fighters, but almost all the PS2-era musou titles preferred to have peons and corporals interrupt your combos or even use special attacks to weigh you down. It's truly dangerous to wander away from friends and get surrounded; I've lost a couple times while playing Yi Ling and Wu Zhang Plains on Hard this way. Getting juggled sure isn't as fun as juggling them yourselves! For lack of something more complex like holding/sieging bases or keeping supply lines established the further in you go, this mix of active AI (without gimped movesets) and limited on-map resources like health or save points hems the player in from trivializing campaigns.

Taking out officers and large groups of foes, preferably via long combo chains, yields item drops affecting either temporary stats (e.g. current health and arrow stock) or permanent ones. Some of these also pop up from busted crates or pots lying around settlements, but it's often viable to forage these from the killing fields as you go. Something I really like are the limited-time stat boosting power-ups which any musou fan will recognize: armor and attack buffs, drinks to restore your special meter, etc. Encouraging players to keep moving and plow forward, skipping from one skirmish to another while making optimal use of item drops…It Just Works™. Another thing enriching the gameplay is the battle UI, displaying informative chyrons whenever you hit an "X units defeated" milestone or when an ally/enemy unit's morale rises or declines. Information travels faster to our screens than could possibly have reached officers' eyes and ears all the way back in the past, and that's okay. It's yet one more decidedly unrealistic element which brings players into the stressed, frenzied state of mind that anyone fighting on these fronts would have had. There's always something going on, for better or worse, and late-game stages can really test one's endurance.

And even with its teething problems, Dynasty Warriors 2 makes good use of its large stages and flashy art direction to compensate for things outside of Omega Force's budget or expertise. I can list quite a few set-pieces and vistas that'll stick with me all throughout the series: the claustrophobic gauntlet at Hu Lao Gate, Zhang Fei's bridge standoff at Changban, the fire attack raging across Yi Ling, and the dread expanse of death at Wu Zhang… All these levels would eventually improve and expand over the years, with more events and intricate pathways to explore, but it's cool to experience them in this formative state. It's just a shame that textures are comically low-res and that the engine can't yet gracefully handle level-of-detail loading, leading to a lot of pop-in and disappearing combatants which adversely affects the game loop. Thankfully you're always able to see your fighter, one of many flamboyantly designed heroes balking at RoTK's conservative character design ethos. Even the most conservatively dressed fellas here look amazing, yet still believable enough (if ceremonial outfits count as wartime or official attire). The cast only gets more colorful and distinct from here on, referencing famous legends about individuals like Xiahou Dun or Gan Ning to embellish them.

| The Boundless Ground |

Maybe the best example I can give for Dynasty Warriors 2 remaining relevant today is its soundtrack. Omega Force's then novice musician MASA (aka Masayoshi Sasaki) banged out one of the most era-defining hard rock soundtracks in the PS2 library, all without having played a Koei game before. [3] From the opening march upon the Turbans, across the splattered red cliffs at Chi Bi, and at the bitter end of generation-spanning warfare, MASA's percussive guitars, synths, and rhythms attune every play session to the pulse of confrontation. The grunge and Y2K electronica of everyman PlayStation games in years preceding gave way to a gentler mix of genres with improving production values, even from inexperienced developers like him. These songs are so fun and memorable that Dynasty Warriors 3 imports almost all of them with slight rearrangements. Nor is the foley any slouch, either. Hit effects, hoof-steps, and rancorous environmental sounds pervade every inch of the game's missions, all without blaring over the tunes.

I've been giving this game a lot more praise, or at least apologizing for its shortcomings, compared to most opinions I read in the current year. It's a hard one to get into now because it's been so thoroughly obsoleted by its sequels, save for hacking potential (seriously, DW2 has some of the most absurd cheats and custom fan levels I've seen from a musou game). For new Warriors fans, I'd say there's a bit more here to experience than just a history piece. Oddities like cinematic kill cams, on-map save slot collectibles, and musou canceling (denying a unit's special with your own!) give this entry a more arcade-y feel than would become the norm. The sheer moxie of infantry doing anything to hitstun you outdoes the harder PS2-era games like DW 3 or DW 4. And while it's low on overall content, the unique level designs for stages like Yellow Turbans and Chi Bi keep me coming back. Bonus points for carrying a smaller filesize footprint than the following games, too! I guess that's what happens when you've got minimal voice acting and cutscenes (both of which are surprisingly less awkward than in Dynasty Warriors 3 for whatever reason).

Dynasty Warriors 2 set a standard for console-bound 3D beat 'em ups to follow. While SEGA and Capcom eked out some '90s arcade-style examples of the genre up until the HD transition, even the latter ended up creating Sengoku Basara to meet Koei's new cash cow on its own terms. The newfound emphasis on influencing a larger world at war, rich in unlockables and storied presentation, made for a compelling combination of elements that Omega Force did better than anyone for quite some time. Eventually there'd be competition from the likes of Ninety-Nine Nights, Kingdom Under Fire: The Crusaders, or the ill-fated Demon Chaos, but Dynasty Warriors ended up having to worry more about OF's own Samurai Warriors series more than anything. The musou genre lost its freshness fast, too, with criticism most often coming from jaded fans or critics perceiving stagnation within these core series. Yet it's still a hugely successful part of the AA+ gaming market today, and it's meaningful to trace these ups and downs back to their source. If you've bounced off DW2 in the past, maybe consider a retry; there's some very nice PCSX2 cheat codes to enhance graphics and playability if desired, too.

| Sources |

[1] 任天堂ホームページ. “社長が訊く『ニンテンドー3DS』ソフトメーカークリエーター 篇|ニンテンドー3DS|任天堂.” Nintendo Co., L. Accessed January 22, 2024. http://www.nintendo.co.jp/3ds/interview/creators/vol16/index2.html.
[2] “鈴木 亮浩 | 社員インタビュー | 新卒採用2017 | コーエーテクモ ホールディングス.” Koei Tecmo Holdings Co., Ltd., January 4, 2016. https://web.archive.org/web/20160104172459/https://www.koeitecmo.co.jp/recruit/graduate/2017/member/vol-01/.
[3] “佐々木 優嘉 | 社員インタビュー | 新卒採用2017 | コーエーテクモ ホールディングス.” Koei Tecmo Holdings Co., Ltd., January 4, 2016. https://web.archive.org/web/20160104172514/http://www.koeitecmo.co.jp/recruit/graduate/2017/member/vol-03.

Fellas, if your game has platforming, comical traps, and a giant angry gorilla attacking you from above, that's not Donkey Kong, that's Crazy Climber!

The similarities are plain to see, yet there's a good reason why Miyamoto swapped out wall-climbing for jumping. Rarely does an arcade game use controls this unique and vexing, a great fit for gated joysticks and a horror with analog sticks. Crazy Climber asks you to learn an 8-way movement scheme and scramble up four increasingly challenging towers, with falling objects ready to knock you into the air like Hans Gruber. Imagine trying to learn how to walk all over again, frustrated in your attempts to reconcile how you think movement works with how it really does. This game might as well have pioneered the "difficulty wall" as a concept. It's just too damn happy to end your runs because of a split-second mishap, as was the case for me when trying to scale past closing windows or escape the Kong's punching range. Were it not for how unique and engaging it is just to learn the game's controls, I would have given up a long time ago. Surviving one round is a feat unto itself, and playing fast and efficient becomes a test of skill and endurance like little else at the time.

I'd ask what sick, twisted, and clever minds could engineer this monument of a 1980 arcade classic, but the development history's more complex than I expected. Basically, Nihon Bussan (aka Nichibutsu) had much less involvement in the creation of their earliest hits than the Western world thinks—almost all design, art, and coding work came from the often-forgotten Jorudan Co. [1] Both Crazy Climber and its older cousin Moon Cresta trace their concepts back to the late Mitsuo Honda, a planner working with Nichibutsu's home staff, plus freelancers Kiyoshi Oda and Toshikazu Sato at Jorudan, to bring new concepts into Japan's arcade market. Like other ambitious post-Space Invaders titles from the turn of the '80s, this one saw a rushed release, with concepts like clambering away from fires (a la The Towering Inferno) scrapped for time or avoiding bad press [2]. Whereas the publisher's space shooter innovated via a Voltron-like power-up system, Honda and co.'s second game practically reinvented the action game overnight. Only Space Panic was even toying with the idea of gravity-based threats and vertical progression by this point.

My enjoyment here came from narrowly dodging seemingly everything and the kitchen sink while I played human fly. You're able to straddle across and ascend up the windowsills not just in cardinal directions, but between two adjacent columns when positioning your arms just right. This all happens via a dual-stick control setup mimicking climbing motions, which takes up plenty of mental bandwidth before dealing with the towers themselves. All four stages vary in layout, alternating between wide and narrow columns sectioned by different kinds of traps. Sometimes it's burger-headed guys poking out and dropping plants, buckets, etc. on your head; one collision dislodges a hand, another send you falling! (And it's a life lost if you're in the middle of your climbing pose/animation, so staying put can be safer than risking a last-second escape plan.) Elsewhere, there's big 'ol birds crapping on you from a considerable height, or the aforementioned King Kong pastiche guarding the way to the roof. These earlier obstacles mainly affect one or two routes upward, but then there's malfunctioning signs spanning half the screen to get around. Yeah, it's a lot.

For creating what might as well have been the very first platformer, Jorudan struck gold. It's criminal that Honda, Oda, and Sato got paid a measly 8 million yen vs. the large revenues Nichibutsu's cabinets grossed [3]. Crazy Climber works because of its difficult controls, where finding a rhythm and consistently staying out of harm's way. The cute flourishes layered throughout, like the public domain song jingles announcing each stage, amount to something more than the sum of its parts. Maybe I'm just going easy on this one because it looks and sounds much better than almost anything from 1980. Blue sky background? Check. Crunchy sound design that would have made Namco proud? Absolutely. The game loop itself manages to find that one-more-turn addictive appeal despite its obtuseness. For example, obstacles that don't immediately drop you will still reduce the end-of-stage score bonus. Looking upward and juking enemies into wasting their shots on either side of you becomes a tantalizing activity for score-chasers. It's also just damn funny to watch my poor lil' guy get electrocuted while launching past an electrocuting banner, lighting up in the process. Just don’t dawdle for too long, though, or the game gets properly mean.

There's not much complexity to the average Crazy Climber—enough for a well-paced 1CC, not quite there for a repeat visit. It's a shame that the developers had to scrap four more levels, for which they'd built out new baddies and things to avoid, but had to scrap and move forward without or risk missing their deadline. Thrills come quick and linger around in this game, from the complications of getting anywhere up and around skyscrapers to the simple joys of finally grabbing that helicopter at the top. As unintuitive and hostile as it seems, I think it ends up becoming a fun, appropriately zany Harold Lloyd simulator (no suspended clock included). The score-run videos I've seen demonstrate the depths of player skill one can reach with this premise, to say nothing of the later sequels. That said, Crazy Climber is still a hard sell for anyone not into the more arcane depths of MAME set-up or who lacks one of the simplified console ports. I personally wouldn't settle for anything but the Famicom conversion since it adds content to compensate, but the home ports largely avoid the troubles I faced with rebinding sticks.

| Bibliography |

[1] 前田尋之. “『乗換案内』のジョルダンのルーツはアーケード開発だった? 前編.” Institute for Game Culture Conservation (ゲーム文化保存研究所) (blog), March 9, 2019. https://igcc.jp/ジョルダン1/.
[2] 前田尋之. “『乗換案内』のジョルダンのルーツはアーケード開発だった? 中編.” Institue for Game Culture Conservation (ゲーム文化保存研究所) (blog), March 16, 2019. https://igcc.jp/ジョルダン2/.
[3] 前田尋之. “『乗換案内』のジョルダンのルーツはアーケード開発だった? 後編.” Institute for Game Culture Conservation (ゲーム文化保存研究所) (blog), March 19, 2019. https://igcc.jp/ジョルダン3/.

Humanity sleeps in the machine. It gurgles for breath, suffocating beneath smoke and gunfire within the netherworld. I grip the joystick with hands like claws; the sweat feels wrong, like oil on water. Heads-up display signals flare all around my vision as I wrench the exoskeletal warrior through warehouses, space stations, and forlorn caverns. When the foes aren't robots, they're pilots just as feckless and desperate as I. The job is king—morals are optional. Captains of industry march us inexorably towards doom, and I'm just trying to keep my head down, chin up against the rising tide. The harder I fight, the deeper I explore, the more I sense the great chain of being start to fray.

Armored Core…that pit of vitality lying within the most veteran of mercenaries, and an apt title for the series to follow King's Field. From Software staff would tell us they bungled their way into developing this game to begin with, but it's appropriate they'd shift from one dark fantasy to another. Both series deal in obscure, arcane worlds, just with divergent approaches to non-linearity and game complexity. They started life as 3D tech demos before unfolding into realms of mystery and danger hitherto unseen on consoles—the kind of innovative experience Sony hoped would set their PlayStation apart from the competition. And for all the nitpicks and missed potential I can (and will) bring up, it's impressive how effectively this studio captured the one-man-army appeal of mecha media versus other developers' outings at the time. From a simple animation test to one of the studio's core franchises, it's a hell of a leap. [1]

| From this point on, you are…a Raven… |

Mecha action games on the PlayStation weren't in short supply before Armored Core (AC) arrived, though I'd forgive you for believing that. The earliest examples—Metal Jacket, Robo Pit, and Extreme Power—all featured some amount of mech customization and variety in scenarios, but always with caveats. None of them had the storytelling emphasis that From Soft's game introduced. At most, Extreme Power let players choose which missions to attempt first, acquiring points to buy new parts if successful. But that still lacked elements like e-mail chains and running a deficit after overusing ammo and/or failing missions. Robo Pit introduced the extensive parts system within a 3D versus fighter context, and Metal Jacket focused on simpler open-field battles a la MechWarrior. (Though the latter remains maybe the biggest influence on so many mecha games to come, it didn't receive a PS1 port until the same year as Armored Core.)

If anything, I see a lot of commonality between the first AC and Front Mission: Gun Hazard, the latter releasing in 1996 with some notability. Combining the series' heavy geopolitical tone and intrigue with a game loop and structure akin to Assault Suit Valken, Squaresoft's game reviewed well and prefigured the genre hybrids they'd produce for Sony's machine. Critically, they also reworked the parts-as-equipment framework from Front Mission, balancing it with arcade-style pacing and more wiggle room for players wanting to test drive multiple builds. The trouble with mecha xRPGs, then and now, is motivating constant character creation (aka editing your mecha) in order to complete stages, ideally while avoiding damage and long-term costs that could ruin a playthrough. I have no way of knowing if the original AC devs were familiar with Gun Hazard and how it elegantly solves these issues via its mix of complex story, set-pieces, and missions designed to reward creativity.

It's hard enough to make a sci-future this dreadful so enchanting and replayable. Armored Core's semi-linear plot and trickle feed of environmental worldbuilding go far in reifying the player's ascension to ace pilot, a new hero of chaos. People are right to point out the jarring, confrontational "initiation" battle, a middle finger to trends of tutorialization beginning in the mid-'90s. Surviving this teaches one to never fully trust the world they're thrust into, be it the obtuse mecha controls or the machinations of agents, corporations, and other Ravens contracting and challenging you. The fun comes from accepting these additive layers of masochism, a reflection of the decaying worldview which From Soft presents without irony or pomp and circumstance. It's on the player to investigate and understand their predicament. Future series entries add fleeting moments of cooperation and optimism to mitigate the grim bits, but the tone here's consistently muted and adverse. Absent are the triumphant flourishes of Gundam or even VOTOMS, replaced by an engaging but ever-present indifference to the erasure of people and elevation of proxy warfare.

| You have the right…the duty to find out. |

Opening missions in this game settle into a formula of scout, destroy, rinse and repeat, followed by a shopping spree. It's never quite as comfortable as you'd hope; browsing for a new radar attachment after gunning down protesters feels ever so morbid. Nor are you interacting with fellow Ravens during the majority of a playthrough, instead fighting or helping a select few through happenstance. Armored Core keeps players at arm's length from the consequences they wreak upon the world, often chiding them through AI monologues and tetchy e-mail chains. This odd pacing and story presentation lets From Soft transition between unusual missions and plot beats without breaking a sweat. The further you work for Chrome or Murakumo to the other's downfall, the murkier the mystery gets, with ulterior motives of anonymous agents pressuring you into service.

Thankfully there's a decently balanced in-game economy to support the amount of experiments and risk-taking the campaign requires, though not without problems. Buying and selling are 1-to-1 on cash return; you'll never enter the red just through shopping. Instead, the way most players wreck their run is by abusing ammo-based weapons and continuing after failing missions with mech damage. Save scumming isn't a thing Armored Core looks down on, but it will go out of its way to promote ammo-less tactics with energy swords and simply dodging past optional foes. Around halfway through the game, it arguably becomes more important to scour levels for hidden parts instead of relying on the diminishing returns from Raven's Nest inventory. I wish this first entry had done better at keeping the market relevant, but it wasn't to be.

My go-to build throughout the story was an agile, energy-focused quadrupedal range specialist dressed to the nines with secret parts. (If the game let me use the Karasawa with these legs, oh boy would I have been unstoppable!) Sure, there's a lot of fun one can have with beefy machine guns and missile options, but getting the most cash out of missions requires plasma rifles and mastery of lightsaber stabbing to play efficiently. While Project Phantasma struggled to balance the economy back towards non-energy offense, it wouldn't be until Master of Arena that the series largely evened out the trade-offs between common mecha archetypes. For instance, tank-tread mecha in this first game are actually damn powerful due to a lack of movement tricks for the bipeds, but it all falls apart when it's time for platforming or quickly escaping. Bipeds often get the class-favorite treatment in this genre, yet struggle to wield a variety of parts and weapons to handle most challenges this game throws at you later on. That leaves quad-legs builds as the most flexible and resilient option at higher levels of play, a flawed but interesting subversion of what's usually seen in mecha anime and manga. (Ed: Yes, I'm aware reverse joint legs exist. No, I don't use them in a game that punishes jumping all the time. Later AC games handle it better.)

With all these incentives combined, the pressure to learn the classic Armored Core control scheme and physics becomes bearable, if still overbearing. I've come from other tank-y mecha games like Gungriffon, so the adjustment period wasn't too bad for me, but I get why many newcomers stick with analog-patched versions of the earlier entries. Memorizing the timings for boosting before landing to minimize lag, or how to effectively pitch the lock-on reticule and snap back to center, matters more than anything in the first couple of hours. Then add on tricks for circle strafing back away from enemies, often while firing guns or launching missiles, and the combat evolves from awkward plodding into a dance of destruction. And there's no arena mode here to let you practice these techniques in a consistent, scaling environment. A veteran Raven or horse of robots can descend upon you in any of the mid-game/late-game missions, requiring quick reactions and establishing a zone of control (or retreat). It's sink or swim in the truest sense. Past the teething phase, it's easy to return to this control scheme and feel one with the AC, even after years have passed. I won't doubt that full dual-stick analog controls will work even better and enable a longer skill progression, but I adapted to the famous claw-grip style quicker than expected.

| "Pledge allegiance to no one!" |

Any problems significant enough to keep Armored Core below a 4-out-of-5 rating or higher must be deep-rooted in the game's loop and structure; that's sadly true for the level and encounter design here. I'm far from opposed to dungeon crawling in my semi-linear mecha action-RPGs, at least when there's room enough to blast around duels (plus verticality to reduce the claustrophobia). Still, a few too many stages in this debut feel like holdover concepts from King's Field II instead of properly scaled settings to wrangle a mech through. The difference between enjoying "Kill 'Struggle' Leader" and dreading "Destroy Base Computer" boils down to whether or not the story framing is compelling enough to justify zooming through non-descript (though nicely textured) hallways for most of their runtime. Occasionally the designers get clever with metal-corroding gas, inconveniently placed explosives, and other traps to keep the spelunking varied; I had a hoot tearing through the insectoid lairs like I was playing an antique musou game! But later series installments improved these confined missions with more arena-like rooms and affordances to players who make it far in and then can't win due to a sub-optimal build.

If I had to speculate, wide open-ended maps are less common here simply due to hardware constraints, be it rendering ACs and other actors in any abundance (regardless of level-of-detail scaling) or the enemy AI struggling with pathfinding in combat on a broader scale. It's a shame regardless since bombarding installations across water ("Reclaim Oil Facility"), going en guarde with a berserker atop a skyscraper ("Destroy Plus Escapee"), and rampaging down public avenues ("Attack Urban Center") offer some of the best thrills in Armored Core. Objective variety and complexity never reaches especially high regardless of mission category, so just getting to rip up groups of MTs, droids, and ACs goes a long way. Defending a cargo train in the desert starts off humble, then escalates to defeating a full-bore Human Plus combatant interceding on the situation. A series of undersea tunnels and chambers, well-defended and secretly primed to implode, threaten to bury you while avenues of escape close off. A select few dungeon crawls also open up in unexpected ways, particularly those set on space stations where vertical engagements come into play. I didn't think mecha and sewer levels could work, but here I am grinning as I pursue Struggle operatives down waterways or methodically undo their bombs within a rat's maze of air treatment tunnels.

Armored Core rarely has bad missions so much as disappointing or overachieving ones, which makes the finale so uniquely odd. By this point, the entropic cycle ensnaring Chrome, Murakumo, Struggle, and adjacent organizations has caused untold devastation across the earth. Now even the Raven's Nest falls, revealed as the illusory sham of governance it always was. Even bit players in the narrative pitch in, waxing over e-mail about the futility of these conflicts and what's really driving it all behind the scenes. So, with all this build-up and conspiracy baiting, I had high hopes for the last hour, wishing for an epic battle and world-shattering revelations to boot. Sure, I got the latter (if in a minimal, trope-adherent form), but instead of satisfying gladiatorial action, I had to ascend the fucking cubes. Everyone's got a horror story about "Destroy Floating Mines", it seems, and I'm just glad to have survived this much awkward, drawn-out platforming using my quad-leg AC. Squaring off against Nine-Ball afterward isn't quite enough to compensate either, not unless you can have an even pitched fight against this iconic rival and win the first couple of attempts. (The penultimate chambers also reflect poorly on the camera's ability to track fast-moving combatants, even if it makes for an exciting sequence.) I can still appreciate how From Soft didn't explicate too much at the end, instead trying to confound players with interesting questions and non sequiturs in the level design itself. It's all a big joke and we get to grimace through it.

| Shape Memory Alloys |

In conclusion, it's a good thing From Soft nailed all their game loop, distinctive mechanics, and interweaving systems here. The original Armored Core is unfortunately limited with how it challenges players, both in level design and mission pacing. Not having an arena to lean on makes completing the missions with maximum efficiency more of a priority, which can lead to excess retries and scrimping on investments in hopes of affording something better later vs. smoother upgrades in the short term. (I do appreciate how only fighting other ranked ACs within missions makes the Ravens' dynamic more hostile and contradictory, but the game does so little to expand on that angle.) These problems sting less knowing that, as a prototype of adventures to come, this game still accomplishes so much with so little.

Not many series strive to reach a profile this high while teasing players with details out of reach and mysteries about its development unanswered. Anyone invested in the wider world Armored Core hints at, from the shadowy groups running these underground beehive cities to the horrors hiding behind Human Plus, has to read through "data books" (artbooks) and track down magazine previews for scraps. We're only now getting English translations of the artbooks and related articles, all of which are coloring the fringes of the AC universe while only letting trace amounts of humanity through the barrier [2]. And as far as these games are concerned, pilots' backstories and white papers on neural augmentation procedures amount to nothing. Heroes and villains drop in and out of history like mayflies—only shocks to the system register on the scale From Soft's using. We're just along for the ride.

It feels like there's still so much else to analyze here: how the studio crowbar-ed their King's Field engine into handling these pyrotechnics, the peculiarities of Human Plus endings as difficulty modifiers, let alone the timely yet appropriate electronica soundtrack. A lot of PS1 releases from this period struggle to make the best use of their developers' skills, assets, and remaining CD space. I wouldn't say Armored Core succeeds at the latter, using only a few FMV sequences at key points in the story, but it's a remarkably lean and appealing game relative to its own premise. Replays come naturally thanks to multiple Human Plus tiers and the freedom to play all missions upon completing the story (plus making new saves to transfer into Project Phantasma). The controls here, though lacking in finesse, carry forward into a good chunk of the later games, with concepts like boost canceling staying relevant even after the switch to analog. Contrast this maturity with all the pratfalls From Soft made during their King's Field days. They'd learned how to not just lead in with a better start, but retain their creative momentum on budgets larger and smaller with each sequel.

Armored Core represents a coming-of-age for the PlayStation as it entered the midpoint of its lifespan, setting a bar other mid-sized studios could aspire to. Its rough edges hardly mar what I'd call one of my favorite experiences in the system's library so far. Maybe I'm going easier on this one due to my enthusiasm for the genre and the myriad themes this game explores, from cyberpunk dystopia to the malleability of history in the post-modern. It could just be that the core game's so, uh, solid after all this time. I chose not to rely on Human Plus for my first playthrough and that might have helped. No matter how you approach the series today, it's awesome to see it debut this confidently, and plenty of players must have thought so too. The Armored Core series became From Soft's backbone for a decade before the Souls-likes came to replace it, and what AC achieved for mecha games (and ARPGs in general) can't be overstated.

| Bibliography |

[1] Alex “blackoak,” trans. “Armored Core – 1997 Developer Interview - Shmuplations.Com.” PlayStation Magazine. 1997. Shmuplations. https://shmuplations.com/armoredcore/.
[2] Reddit. “Translations of Pages 103 to 105 from the Book Armored Core Official Data Book.” Accessed January 14, 2024. https://www.reddit.com/r/armoredcore/comments/x940dj/translations_of_pages_103_to_105_from_the_book/.

"Chicken, fight like a robot!" blares the cabinet speaker as I generously leave some killbots alive, moving on to the next room. There's no end to them: monochrome sentinels of death, all bunching up along the walls, all waiting to fire. I ready the ray-gun and trudge forward, fingers on the trigger, reflected through the joystick. This continues for untold iterations as the numbers go up, the Evil Otto bounces down upon its hapless allies, and the rhythm of combat multiplies. Everyone has their place here, the guards uncaring for their neighbors and the smiley-faced custodian uniquely aware of this realm's absurdity.

Ok, ok, it's Berzerk, and it's not that complex aside from catalyzing the deaths of two hapless players. Compared to both emerging maze games and shooters like Space Invaders, this one's appeal must have been immediate: unpredictable gunfights, a sassy robotic narrator, and more chaos than something as scandalous as Death Race could have imagined. The origins of the run 'n gun style arguably trace back farther to Tank and its clones, but changing out vehicles for humanoid avatars lets the pacing slacken and rev up more granularly. Crashing into electrified walls becomes more of a threat than colliding with far-off aggressors, who now threaten more with split-second lasers than an opposite player careening into you. So the dynamic's similar but different enough from the motion-full Asteroids, with danger coming from less angles but requiring quicker reactions.

What undoubtedly helped Berzerk survive its encounter with Williams' Defender is the touted "sense of humor" seen with Evil Otto and how it affects players and AI alike. After all, why risk visiting the next dungeon screen and quickly dying when one could just stay put, safe and sound? The danger and necessary push of Evil Otto remains one of the golden age's most iconic symbols, both scary and ironic. And it only gets better when a savvy player sicks the boisterous bounding ball on unsuspecting droids, racking up points while kiting Otto around each arena. Counting score by each enemy's death, rather than players' successful shots, was maybe the smartest move to extend the game's longevity. There's an upper human limit of how long someone's going to traverse one nondescript room after another, but surviving long enough to abuse this scoring mechanic? Why the hell not!

I'd much rather try and master Jarvis' side-scrolling shooter if I'm having to choose between these games (that, and Defender technically started location testing and initial distribution at the start of '81). But there's a simple charm to creator Alan McNeil's attempt at crossing genre lines, evoking the first dungeon crawlers without their labored pace of play. McNeil himself came up with Berzerk's premise based on his own nightmare of robots chasing him through hallways, let alone the memory of security browbeater David Otto haunting the designer from his time working with Dave Nutting Associates. [1] People always talk about how this game feels like wandering through a dreamscape they can't escape, and I think there's some truth to that. On the bright side, it's cool to see McNeil take the early micro-processor lessons learned with Midway's Gun Fight (a recreation of Taito's TTL-chip classic Western Gun). We're talking about the guy who later laid the foundation of Macromedia Director and other technology, so it's unsurprising he could get so much from basic hardware.

Juking bots with Otto and luring them into fences calls an earlier game to mind: CHASE, a derivation of a 1976 BASIC game where players have to crash pursuers into objects and each other [2]. This puts Berzerk in the same conversation as other PC-to-arcade transitional software; in fact, it's one of the earliest examples. It's fascinating how much Stern's 1980 smash success treads the line between hobbyist sleeper hit and a throwback to much earlier media like the Berserker novels and publications like Creative Computing. From here, the tendrils extend towards Robotron and its ilk, let alone a direct sequel in Frenzy. I may be moving on from this weird and seemingly simple curiosity, yet it's going to reappear in one form or another as the '80s arcade era marches on.

Bibliography

[1] Hunter, William. “Berzerk.” The Dot Eaters (blog). Accessed January 1, 2024. <https://thedoteaters.com/?bitstory=bitstory-article-2/berzerk>
[2] David, Ahl, and Cotter Bill. “Chase (High Voltage Maze) - Cotter.” Creative Computing, January 1976. Accessed via Internet Archive. <http://archive.org/details/Creative_Computing_v02n01_Jan-Feb1976> Retrieved on January 6, 2024.

And now, the first MAME Mondays write-up of 2024, wherein I subject myself to the coin-op horrors.

Panic sets in when you realize that ladders don’t count as open spaces, meaning aliens can't fall down them when trapped from atop. If I'm going to play a Heiankyo Alien follow-up with emphasis on using verticality (and thus gravity) to defeat monsters, then having only one way to deal with higher-tier aliens (two or more blank pits atop each other) is a problem. Both Space Panic and its predecessor rely on players making the right tactical move to surround themselves with pits, knowing they have limited movement and must guess where the baddies might move. But at least the 1979 game didn't bother with punishing you for not running over to a trapped critter in time; here, that's practically a death sentence since it enrages the monster, meaning now I have to double-put the thing before oxygen runs out.

And there's the rub: with a strict time limit per life and the enemies gaining more advantage from mistimed traps than players, Universal's take on the maze trapping formula feels claustrophobic in ways Heiankyo Alien simply didn't. Sure, I can also fall down the pits I've made to reach safety, and thankfully Kazutoshi Ueda didn't add death from falling here. But the win state's the same as ever. Dodging monsters means nothing if I can't reliably get rid of them, especially if I now have to time not one, but two pit placements against the timer and the AI's whims. Now let's couple this with a rather high time-to-bonus-life factor (3000 or 5000 points depending on DIP switch), plus how mooks will speed up across stages, and we've got an exponential difficulty curve to make long-form play even riskier. I wouldn't say the University of Tokyo's crack game hobbyists had all the acumen of veteran designers, but they at least wanted random arcade-goers to make progress and regularly get high and higher scores. Its slower monsters and time limit meant one could strategically camp at an intersection if desired, or take risks with pit placement to maximize clear bonuses much easier. Compared to Digger, Ueda's game seems downright hostile, and just as aesthetically spare without the gentler ramping-up of its inspiration.

It's impressive how quickly Universal cloned and, in some cases, matched up to outdid the games and developers they copied, but this plan doesn't seem to have worked out with Space Panic. To its credit, the game sold well and influenced the design of Donkey Kong and Lode Runner among others, but those genre evolutions were inevitable. This title's main new mechanic, using physics to defeat and avoid monsters you could previously only corner, feels both under-utilized and skewed against the player at all skill levels. Adding gravity and risks associated with multiple height levels was the logical next step from flat mazes. Other pure maze games like Pac-Man and Lock 'n Chase would find new ways to restrict but also empower the protagonist, adding verbs like power-ups and lockable doors to discourage turtling while reorienting game flow around dashing for safe spaces. In these earliest trap-em-ups, though, the hierarchy's all flat, with only the largest spaces between ladders and intersections having any priority. So there's a disconnect between what Ueda's added to distinguish this from Heiankyo Alien and what designers elsewhere in the industry would do to innovate the maze game genre. Consider that "platforming" came about by accident, the serendipitous result of imbuing the player character with more and better internal logic relative to their environment.

I guess this isn't too bad for 1980, just underwhelming as a missing link between Heiankyo-likes and the complex maze games to come. The concept's just too underdeveloped and unkind to its target audience, even knowing enough people clamored for a new trapping game to plunk considerable coin onto this one. The cleverest thing happening here, I'd argue, is reusing assets to make a unique hi-score entry screen. It feels quite playful to clamber around and dig for the letters vs. just tapping buttons to type my name in. Honestly, I might enjoy that more than the all-or-nothing dynamic of isolating my guy between pits halfway up the maze or running after the aliens, hoping I don't work myself into an ambush. Whereas I can appreciate the primordial beginnings of the genre in Head-On and Heiankyo Alien, this maze game is too small a jump in complexity and too much of a killjoy whenever I'm about to hit a groove playing it. Here's hoping Lady-Bug and the Mr. Do series can make up for this!

SEGA AGES and Arcade Archives have dominated the Switch eShop's retro re-release scene for too long—now it's Project EGG's turn to share the money fun, bringing '80s/'90s Japanese PC classics back into circulation for today's players. It's a bit questionable, then, that despite all the added quality-of-life features, they've kept Game Arts' Thexder just as comically difficult as ever. I'm going to find it hard to play this without save states now, aren't I. No matter what, anyone keen to try an unforgiving but surprisingly rich action platformer, a computer-bound refraction of Super Mario Bros. that same year, should give this a look-see.

I reviewed Thexder earlier this year while trawling through memories of playing localized and untouched J-PC games back in college, wondering what the hell I'd gotten myself into.Ghosts 'n Goblins hadn't yet reached the PC-88 by 1985, but its sheer level of brutality finds a fellow traveler in this title. Thankfully it's not as outright misanthropic and impenetrable as Capcom's infamous side-scroller, yet I wonder if it's not too far off. A studio like M2 or Hamster would have included some kind of easier mode or options to favor players (that aren't just saves), but not D4 Enterprises. They're confident that this will hold up well in the people's court, and I respect their moxie. Let's not act too surprised to see *Xanadu selling way better, though, whether because of the Falcom bump or how it's just slower and affable from the onset.

| Robo-Dexter's Laboratory |

Even the slowed-down DOS port of the game, released by Sierra in 1987, slices you in two with a proverbial laser beam before reassembling your sad self in not quite the right order. This may not have the design of a Mario kaizo hack, but it's very happy to sucker in players with a challenging but reasonable first stage before laying down the gauntlet. Level 2 asks a simple question: to shield or not to shield? One can choose to barrel through enemies, losing a mere 10 energy to replenish the barrier. Someone more daring may try and avoid as many enemies as possible, ending the level with no shield use and gaining a free 150 energy and raised energy cap as reward. Each following level begins to feel less like a robotic defense installation, more akin to a series of puzzles testing your resource management and ability to juke the droids into sticking on walls or each other.

Routing Thexder involves figuring out which stages are best tanked with shields or are possible to clear while taking minimal damage vs. the no-shield bonus. This becomes easier the further the player gets because designers Hibiki Godai and Satoshi Uesaka decided to recycle level layouts towards the end of the main loop, albeit with changed tilesets and enemies to keep it all fresh (no recolors!). So while it's as tough to learn as any memo-oriented shooter, there's some lenience here and there which retains the game's first impression of Mario meets Major Havoc and Macross. The only notably slow bits are "boss fights", which take the form of multi-object structures containing enemies within. It's fun to pick off each robot in just the right order to avoid releasing them from their prisons to attack you. These sections get a lot more use out of the plane morph than one might expect, all because you need a focused stream of fire instead of the auto-targeting projectile in mech form.

| Preservation, incubation, what's the difference? |

I was apprehensive about whether or not Project EGG's jump from subscription-based Windows emulation to Switch would shake out, and my worries were far from unjustified. As much as I admire Thexder in its historic context, the game's a hard sell today without offering more invasive ways to alleviate its ruthlessness. Save states are limited to just five slots here, along with a few game speed options which let you play slower or faster at the expense of audio design. Button remapping works very well, but the lack of turbo functions spells trouble in the future (or right now in Silpheed's case, though I need to play that port to know for sure). Including the manual is a nice touch, yet it also highlights the lack of other contextual materials like magazine articles, production materials, etc. There's a lot more D4 could have added to round out the package without using excessive time and budget they understandably would want for upcoming releases.

On the other hand, maybe five save slots and a minimalist approach works fine for this era of J-PC game soft, not swinging too hard towards purism nor revisionism. I think players should have the final say in how a game's played, with developers ideally accommodating multiple audiences' needs through options vs. forcing a one-size-fits-all approach. And it's hard to tell if Project EGG currently sees an opening in the retro games fandom to attract newer, younger players or if they'll lay the commercial onus strictly on aging dads and salarypeople from the '60s, '70s, and '80s who are already familiar with the J-PC library. When the software you're trying to sell is interesting yet somewhat inaccessible by design—front-loaded with challenge and gotcha moments meant to prolong playtime while encouraging the game's community through shared secrets—what can a modern digital distributor really do the bridge the gap in player demographic mindsets?

The EGGCONSOLE releases of RELICS and Thexder feel tentative and spare, very much testing the waters of how far these ancient J-PC legends can evolve to meet EGG's market needs. But I also doubt one can really do more with a game this honest and straightforward, not without designing basically a whole new game via difficulty rebalancing. M2's able to go that far because SEGA and other large firms back them up on projects big and small, whereas D4 and Hamster only have so much capital to buoy their teams. The mere threat of drastic changes turning off long-time EGG players might be enough of a problem to outweigh creating a full-on modern remake (ex. Thexder Neo). It's a tough call for the company to make, but I think they should still try to push the envelope more than this. Something as simple as letting players keep the end-of-level energy bonus even when using barriers, for instance, wouldn't deviate too far from Godai & Uesaka's design. It'd just speed up the average player's learning process, letting them chomp at the later levels earlier.

| Moonlight Sonata |

D4's imperative with EGGCONSOLE, ultimately, is to thread the needle on fidelity vs. accessibility while escaping the chokehold that their PC service's subscription model's kept them in for decades. I've rarely been able to convince anyone to pay upfront for an account, and then buy access to the games EGG carries, vs. just directing folks to emulation and the Neo Kobe packs on Internet Archive instead. Not only is this release a way better value for returning players who no longer have to lug a laptop with them to play Thexder (assuming the Steam Deck can't run the EGG Player, which is likely), but it's a cheaper buy-in for anyone encountering this and similar software for the first time. (Granted, you've always been able to play EGG games without an Internet connection and/or active sub, but that doesn't remove the stink it seems.) Make no mistake, there are worse ways of handling this big an emulation library—AntStream says hello from the bleachers.

I hope and expect this move onto Switch will get Project EGG out of the rut it's been stuck in, and that more will at least try to crack Thexder's tough exterior now that save states are available. There's quite a bit more that the company could have done to improve their emulation package here, so I hope any broadly applicable add-ons to later releases get backported here. It's hilarious how much they've improved on the basic Windows-bound emulation of Thexder, and I wouldn't be surprised to learn that EGGCONSOLE's grossly outperforming the traditional PC service now as a result. Well, enough musing. I've got a saveless run to figure out when I'm not thinking about ancient Rom—erm, sorry, I meant Bouken Roman and other weirdo platformers confined to PC and arcade obscurity. Thexder's lucky…it's got a sizable legacy, being the game that sold the PC-88 in '85, and that shines on nearly 40 years later thanks to its solid, addictive design. All these titles coming to Switch now, late as it is, just means I get to see more of y'all try and hopefully enjoy these old standards on their own terms.

Only a year after the first wave of indie throwback shooters (Dusk, Amid Evil, etc.), Australian jam band King Gizzard and the Lizard Wizard (what a name) saw the potential in promoting their thrash metal album via a mini-FPS. I'm not about to say it went well, or did anything a more evocative music video could have done better, but the effort's appreciated. Infest the Rats' Nest deserved way better, though. Back in my day [grandpa voice] our free FPS flash games let us shoot Bill Gates for the crime of Microshaft Winblows! This honestly just does way too little with the single's subject matter and feels like a last-minute promo.

We've basically got just one little combat arena available from the jump, with usable but very basic movement and controls. Then there's a super limited bestiary to fight, pulling as much from the iconography of albums like Nonagon Infinity as it does Infest the Rats' Nest. Building even a small demo game for the latter's top single—a roaring banger satirizing Elon Musk and other techbro imperialists' dreams of abandoning Earth—shouldn't be difficult. Just have two or three punchy guns, simple yet replayable levels to blast through, and some of that juicy, inscrutable Jason Galea psychedelia. What could go wrong?

It could just be a big fat nothingburger, that's what. The arena's all you get, as are a bunch of annoying bats to shoot at. Low-end Serious Sam/Painkiller noodling under weed-excited skies isn't all that engaging, sorry fellas. And while the music does the heavy lifting as it should, I just don't think Flightless Records committed any ambition to this beyond ticking a marketing checklist. Seriously, the very idea of a King Gizzard boomer shooter is awesome. The band's variety of genres, their loosely-connected musical multiverse with albums and videos nodding to each other...so much potential, yet it's hardly hinted at here. The most I'll say is that this isn't abominable to play, just an annoying reminder of what could be. Maybe I just gotta work on my Doom mapping chops and start a community project to fill the gap.

A full-bore King Gizzard WAD would have to cover the following key points:
•Has to loop in on itself (last map leads into the first)
•Total conversion of all graphics, monsters, weapons, and power-ups (aka some real GZDoom shit)
•Copious references to everything throughout the Gizzverse, from Han-Tyumi's barf bag to a trip down The River and into the Crumbling Castle
•Game flow that just feels like a series of effortless, chaotic band jams, transitioning between incidental and slaughter combat
•Some breather levels playing off the folk pop/rock songs; similarly, the secret maps could use concepts from Gumboot Soup and the band's demos
•Full MIDI remixes of relevant KGatLW songs, alongside the originals themselves if the player desires
•The band's gotta stream themselves attempting to play the thing on Ultra-Violent at least once

I could go on. Basically what I'm saying is we live in a timeline where masterpieces of music industry multimedia like Frankie, Ed Hunter, Spice World, Tomarunner Vs L'Arc En Ciel, and Wu-Tang: Shaolin Style exist (among others). Something's going terribly wrong if the likes of zero-to-hero Aussie maximalist rockers are settling for this by contrast. They can go harder, cheesier, zanier...more extra in every way, just like their music videos and performances, yet look what the Mars for the Rich online game has amounted to. Sorry state of affairs, or just a mistimed and under-resourced gift to fans excited for the album? I don't know—it just irks me. Maybe I'm underestimating the labor and budgeting a longer ludonarrative experience requires, but I doubt it'd break the bank to make something simple but more involved and replayable than this.

Santa's jokerified and you get to kill Scrupulous Fingore. 11/10 ludopoints!

For a holiday season checklist-ticking release, Daze Before Christmas is rather humble, content with being one of the few enjoyable titles solely dedicated to the season. I'm not sure which was Funcom's first game precisely, this or the seemingly abysmal We're Back! tie-in A Dinosaur's Tale, but it seems like Sunsoft saw something of worth in this one. Going from these dinky platformers to big MMORPGs like The Secret World and Age of Conan—let alone anything as acclaimed as The Longest Journey—must have been wild for this Norwegian startup. It's hardly as if they lacked in other options: most Scandinavian developers stuck to popular PCs like the Amiga during the early-1990s, for good reason. Developing games for cartridge is way more expensive, and riskier due to manufacturers' quality auditing, than putting out mail-order floppy software. So let's give credit where it's due and assume even something like this, which outwardly resembles shovelware, took plenty of effort and tender loving care to create too.

| The presents were mutilated beyond recognition |

With all of the kids' presents missing and a bona fide Gang of Four terrorizing the holiday spirits all over the world, it's now your job to guide Fake Tim Allen through 24 levels of comfortably numbing platformer tropes. Everything's introduced via a cute "T'was the Night Before Christmas"-styled lyric, with each stage laid out via an advent calendar. Honestly, the most memorable parts of Daze Before Christmas are the intro artworks preceding levels, particularly those which look deranged or ominous. Lead artists Ole-Petter Rosenlund had come from the Amiga scene, drawing for works like Psionic Systems' Assassin, so it stands to reason a bit of the 'ol European PC weirdness has snuck into an outwardly unassuming family-friendly game. Baddies and obstacles range from wily rats to industrial machinery, plus all sorts of toys now animated to attack poor Saint Nick. The antagonists themselves look real goofy; I love "Mr. Weather" for how much he resembles Fingore, but also The Timekeeper for looking like a lost Clockwork Knight boss. Shout outs to the floating hearts in the Wood Factory level, you saucy fellas.

Moment-to-moment play in Daze Before Christmas follows a neat formula: hop and skip from A to B, collecting and opening presents to rack up score and lives/health, and occasionally pick up a hot cuppa to transform into Anti-Santa. It's very standard stuff, and I wish your alter ego had meaningful new mechanics beyond temporarily hurting a roadblock enemy, but the controls feel just weighty and responsive enough that I can't complain much. Many level designs feel like the best bits of Earthworm Jim and other Western platformers mimicking the more ambitious Eastern examples, for better or worse. Maybe they could have cut down on the number of pace-killing hazards and auto-scrolling sections; those bumpers and magic carpets got old fast. I think kids would have had a challenge in navigating the ice caves and sewers, but the homes and snowy hills are straightforward to navigate, albeit with a smattering of secret goodies to find off the main path(s). There's also the odd flying stage, where you float your sleigh all over to chuck gifts down chimneys like it's a bombing raid. Lack of variety, repetitive props and sequences, and plenty more déjà vu hold this game way back from greatness, but it's the kind of mediocrity I can understand and let flow right by me.

| Oh what Fun it is to Com |

If I'm making this out to be some flavorless adventure, then you'd be partly correct; this Norwegian oddity has some strange vibes I didn't expect to encounter, though. One stage essentially reworks the Boo House concept as a more setpiece-driven gauntlet of boxing cacti, hurtful ghost rats, and an end boss which requires players to repeatedly stun Louse the Mouse with an Eggman-like spike ball. Another couple runs have Santa dropping into gradually flooding sewers to muck around for elves and their stolen handiwork, with bits of platforming that require actual engagement (comparing well to the factory stages with their many bottomless pits and elevation spikes). Coupled with how little of the game occurs during the day, there's a subtle gothic flair to everything one can experience in Daze Before Christmas, reinforced by a groovy set of rearranged tunes and carols within a smooth FM-synthesis palette.

Summing up a game this brief and frankly shallow feels just a bit difficult when looking at its surprisingly rich presentation. Right away, the options menu gives more control to players than so many of its peers, from audio levels to how long the Anti-Santa morph lasts. Difficulty modes don't vary much in practice; I suppose they affect how frequently a red box gives out bombs and other traps, rather than freeing Santa's helpers or health hats. Then there's the aforementioned soundtrack, which properly conforms to strengths of both the SNES and Mega Drive sound chips, using pleasant instrumentation and sometimes taking unexpected turns (ex. the ripping guitar solo during boss fights!). And for all its missed potential, whether due to inexperience or a short turnover time (which I won't know until the History of Sunsoft Vol. 2 book releases), Daze Before Christmas is certainly a competent romp that anyone can bounce in and out of. Playthrough take around an hour or two at most, password saves are available, and the button layout's totally standard for 1994. The package here is well tested to a fault, other than some momentary collision glitching.

Hope everyone's enjoying their Hannukah, Kwanzaa, Festivus, Saturnalia, and whatever festivities we seem to obsess over every year at this time. I'll do my part and speak respect for what Funcom did here with Daze Before Christmas, regardless of time, budget, or scope the studio had back then. Its bizzaro Australia-only release on Mega Drive, only expanded to Europe for SNES, meant it had little impact during the 16-to-32-bit transitional period, while later Xmas-themed works like Christmas NiGHTS have rightly shown how much more ambitious game developers can get with holiday theming. Still, I think it's impressive how effortlessly Funcom's title plays compared to many derided Euro-platformers it shares heritage with, to the point that I rarely see any comments making that connection. That's a valuable accomplishment for the period! I'm just sad I can't make a Midsummer pun for this Midplatformer…whoops.

[AVGN voice] More like faxing in a doodie! [Cue skit]

Simon's Quest never stood a chance. Zelda II? Just another competitor for the best NES action-RPG of 1987. Xanadu who? That ain't affordable! Why buy a home computer when you can get the [approximate] thrills of computer role-playing action on your TV?! Faxanadu, baby. It's only the lovechild of two of Japan's most lauded developers, seizing the chance to entertain untold numbers of schoolboys with fantasy swashbuckling and dungeoneering like nothing else. Well, maybe that last part's an exaggeration. Maybe this was the best you could get on a cartridge, but any kid lucky enough to own a Famicom Disk System had similar titles well within reach, all iterating on the likes of Xanadu, Zelda, Metroid, and Castlevania.

Poisoned roots flow from the base of World Tree, our hero's old home now corrupted by a cosmic evil. Faxanadu toys with a kind of dark fantasy that had so far evaded the Famicom's library, bewitching its owners with sojourns into caverns, forests, castles, and places that should not be. The people once called Dwarfs now scour this Yggdrasil in mutated forms, and the wise men of the realm wait patiently by desecrated fountains and shrines, hoping for a hero patient and skilled enough to heed their call. While I'm happy to report the game comports itself admirably versus the other ARPGs in the room, its niggling flaws and inability to transcend its inspirations pose a problem for the game today. Nonetheless, it's a testament to how quickly its developers improved at making Famicom exclusives, and one of the breeziest but filling genre exercises that Hudson Soft made for the system.

| "Daggers and wingboots, mantras and monsters await you." |

Famicom Xanadu, as the portmanteau suggests, had a mission to fulfill: bring Nihon Falcom's seminal 1985 PC RPG to the console-bound masses, no matter what it costed Hudson. The two companies knew each other distantly via the Japanese PC games market already, and this collaboration marked the start of a several years' long partnership. Hudson later ported key Falcom titles like Ys I & II and Dragon Slayer: The Legend of Heroes to the former's new console of choice, the PC Engine (CD). Falcom, meanwhile, reaped the crop, collecting hefty residuals from sales of their games on platforms they had no experience with, expertly handled by their partners from Hokkaido. From '87 onward, Falcom made inroads on the far larger console market without having to divert their staff away from the tighter, more competitive J-PC realm, at least until they made their own PC-to-console push in the early '90s.

What we now deem the side-scrolling ARPG had blossomed beyond its niche by '87, thanks to early hits like Namco's Dragon Buster and the diversification shown with cult classics like Layla and Mugen Senshi Valis. Developers tackling the genre sought to meld dungeon crawling, cinematic pacing, and skillful play into something you could play on the simplest of systems. In that sense, '87 was maybe the most important year in ARPG history: Falcom came out swinging with three of the genre's all-time classics (Ys, Sorcerian, and Drasle Family), followed by the likes of Konami (Maze of Gailous, Arumana no Kiseki), Capcom (Black Tiger), Westone (Wonder Boy in Monster Land), Game Arts (Zeliard), and even the NES remake of Rygar. I bring all these up to explain why I've rated Faxanadu roughly in the middle of the pack. There was a lot of competition in the ARPG space at this point in time, and Hudson arguably came in on the back foot.

You'd be right to doubt Hudson's record with console-oriented adventures after the janky results of Challenger, Milon's Secret Castle, Mickey Mousecapade, and arguably the first Adventure Island. The company had triumphed early on with solid renditions of Lode Runner and former J-PC games like Binary Land, but their competitors had leapfrogged them in the art of hours-long Famicom blockbusters. 1987 saw a lot of ups and downs for Hudson's various Famicom teams, with Faxanadu thankfully indicating an end to all but the occasional abject failure. Such a turning point helped double-fold for their upcoming PC Engine projects. Neutopia and Dungeon Explorer, derivative as they are, benefited greatly from the lessons in relatively consistent design that the developer learned this fateful year. I wish we knew who precisely worked on Faxanadu other than musical maestros Jun Chikuma and Toshiaki Takimoto, but good job to everyone else!

The adventure starts off in the main castle town, Eolis, beneath our once verdant World Tree. Despondent residents mill around streets and dim rooms, waiting for relief. The "dwarfs", now resembling creatures from the likes of Alien or Metroid, creep around the town walls, forcing new players to learn the jumping mechanics within the first few minutes. A desperate passer-by hands over their ring, the symbol of Elf kind and one's ticket to an audience with Eolis' king. He tells of a meteorite, crash landed high up within the World Tree, which has poisoned most of the waterways and especially the Dwarf fortress, leading to decay and depression all throughout. After receiving 1500 "golds" and the opportunity to train strength or magic (which just fills those bars, no stat increases here…), players must set off to reactivate the fountains, subdue the Dwarf incursions, and vanquish the evil brought here from outer space.

| "I am free from injury because of the ointment!" |

An immediate strength of Faxanadu is its pacing, with most of the game following a linear progression from below to atop the Tree, foisting small and large challenges upon you with plenty of leniency. The protagonist mainly grows via obtaining new equipment: swords, magic spells, armor, and shields, all buyable or present in corners of the world. Stat progression happens invisibly via changing equipment, while the visible EXP state simply correlates to each title a guru/priest bestows upon you at each temple. This all ties into a robust password save feature, allowing players to recover their progress (ex. equipment, key items, story flags, etc.) and rarely have to retread familiar ground. It took me roughly 7-to-8 hours to clear my first playthrough (not counting fiddling with emulator settings), and maybe an hour of that time involved any notable backtracking.

So, like with Ys and other item-focused ARPGs, there's never much worry about character building, just a well-incentivized goal of getting the best loot (which appears on your guy!) and reaching victory via guile and quick reactions. Controls are mostly well adjusted, both for '87 and compared to the larger Famicom library. Up +/down + A to either use magic or your active item takes some getting used to, but there's little in the way of obscure combat/movement mechanics to complicate things. What I will criticize, however, is the stiff, inconsistent jumping and collision physics. It's admirable of Hudson to test players' ability to time jumps and approach enemies with caution, yet I had way too many close calls where I should have just landed on platforms, only to fall and subsequently retry a section. Regarding enemy hitboxes, these are fine for the most part, but vexing when their attacks bump you back as far as they do. Since Faxanadu is a flip-screen adventure, not a fully scrolling one, this means baddies can and will force you off-screen, prompting extra loading times and enemy respawns which hurt an already precarious balance between affordances for players and the AI alike.

There's way to mitigate getting cornered or too low on health to continue, thankfully, as Faxanadu offers Red Potions, Hourglasses, and other power-ups that let you tank damage, stop time, so on and so forth. Most of the time one should stock up on these at shops in-town, but it's possible to find them out in the wild, either reappearing indefinitely or spawned after defeating a specific set of monsters on the right screen. Better yet, just use magic to avoid risking upfront damage entirely! The spells here evoke their counterparts from Falcom's Xanadu, but are generally simpler, being mostly front-ejecting projectiles and a late-game option which curves and flies upward. Judicious alternation between melee and magic makes the biggest difference in clearing dungeons and traversing the overworld. I would have appreciated more variety in types of weapons, spells, and restoratives on offer, but altogether there's plenty to work with here, especially compared to most Famicom contemporaries.

| "The power of the Hourglass is gone." |

It's by the halfway point—a daredevil trip into the skies, preceding a long and arduous trip through the World Tree's foggy, forlorn trunk full of abandoned sites—that Faxanadu shows its hand, warts and all. This game excels most when rewarding the player's tenacity with a sense of discovery, the feeling of a larger universe than can feasibly be shown. Lengthy hallways inside roots and branches give way to dusty castles possessed by the deranged, where warlocks, monstrosities, and confused soldiers patrol for Elves to slaughter. Each "tower" offers its own set of action, platforming, and item challenges, striving to fully exploit the game's possibility space. I always had a hoot strapping on some Wing Boots, levitating high above foes and hazards (bottomless death pits never appear, thankfully), only to slam right into a miniboss or something cool like that.

See, there's just one (or two, or three) problem(s) with what Faxanadu's, uh, doing here: it rarely if ever captures the tension and complexity of its source material. Xanadu on PC-88 had many flaws of its own, but one couldn't deny the sheer ambition, variety of traps and monsters, and involved leveling + customization Yoshio Kiya and co. managed to squeeze onto those floppies. Here, with limited storage (no Disk System upgrade here!) and a need to accommodate much younger players, Hudson's take on the concept finds itself at an awkward crossroads. The line between regular enemy, miniboss, and boss blurs without confidence. Chikuma's soundtrack, though somewhat catchy and appropriate, struggles to rise above the aural noise reserving sound channels she really needed for her arrangements. A mix of engaging but merely competent platforming and simplistic hack 'n' cast combat makes locations like the Tower of Suffer feel like, erm, suffering. Some truly ill-thought level design comes to mind, too, like mooks patrolling right in front of doors you need to exit and enter (meaning certain death in some scenarios), let alone the lack of unlockable shortcuts or fast travel by late-game.

To further illustrate my reservations, let's look at the keys system, maybe the most belabored aspect of Faxanadu. Many doors throughout the adventure require you to manually equip specific keys (jack, joker, king, queen, and ace) from the inventory, then use them to proceed. Since you can't drop any expendable items at any point, this can lead to accidentally stocking up the wrong type of key, and there's only 8 item slots to work with! So let's say I try selling them off in town, the logical thing to try. It turns out you can only sell specific types to key vendors who already stock them, meaning I'm stuck with early-game items by endgame unless I do way more backtracking than should be necessary. Going in, I honestly thought I'd get more frustrated with the Pendant bug than something like this, but here we are. In terms of which problems are intentional design or just coincidental, I'm unsure where the Pendant's infamous (and seemingly contradictory) damage debuff falls, but Hudson knew what they were doing with keys here. Inventory management, coupled with slow-ass text speeds, can quickly put a damper on one's fun if you aren't careful.

Though Faxanadu is still more intuitive and less guide-dependent than its spiritual predecessor, I find Falcom's game so much more consistent, dedicated to stumping the player while making their accomplishments feel that much more meaningful (and avoiding many aforementioned pitfalls like with keys!). Obtaining better gear and then demolishing them through hitbox cheese is fun, but I wish Hudson could have either added more features to enrich the experience, or taken a bit more time to sand off the edges which remind you this game's trying to evoke Japanese PC RPG jank. For example, where the hell's crouching in this ARPG full of monsters that attack from above and below? Why can't I attack at all on ladders except after getting hit, thus bugging out my hero's sprite and letting me act? With all these omissions and issues working against the player, it's hard not to feel like the latter half of the game becomes a slog. It all starts to feel rushed by the end, especially when fighting the final boss in all its Giger-esque glory. Perhaps the droning, repetitive nature of OG Xanadu is hard to stomach now, yet I find more conviction in that release's design direction than here. Close but no cigar, I feel.

| "Don't have negative thoughts. Remember your mantras." |

I don't hold any production difficulties or inexperience against Hudson staff, despite my hangups, and Faxanadu accomplishes more than it's often given credit. One's moxie in battle and exploration often leads to hidden rewards, or the satisfaction of clobbering bad guys before they ever get the chance to react. Minibosses show up frequently as the story progresses, with each posing a unique problem to solve (will I jump over fireballs or block them with my shield? or how about pinning these jumping eldritch things against the wall to not dodge all the time?). While it can be tedious to backtrack all the way to the beginning or grind for a big wallet early on, doing so grants access to a couple of the best items in the game, rewarding anyone who takes their time to learn the game ASAP and preserve one's life long enough.

Hudson seemed to have a knack for immediately following poor design choices with smart ones; one screen filled with hard-to-hit ranged attackers can just as quickly lead to another with health-dropping fodder! I find Faxanadu pleasantly accessible for an ARPG of its vintage despite the pendulum swinging from quite (sometimes) very good sequences to weak ones. Hit feedback's always satisfying, as is the crunchy but legible audiovisual design. Having just eight items means, in an ideal run, you can stock a bevy of potions and powerful items to seize the advantage, especially when fighting those damn flying bees or disabling bigger guys' means of claiming space. At its best, the game makes you feel empowered through both well-paced encounters and some of the more satisfyingly cheesy strategies one can pull off in an ARPG. Figuring out simply where to go and what to do is also a highlight, thanks to nicely written NPCs throughout the realm giving you hints and clearly delineating questlines.

However, Faxanadu's most convincing strengths lie in the atmospheres it evokes, defying the Famicom's limitations to create something rarely matched in 8-bit xRPGs. Simply put, it looks incredible for an '87 console game. Tile upon tile of ornate details, greebles, and textures found both in pixels and CRT blurs combine to illustrate an otherwise fairly generic world in greater depth. Florid passages transition into harsh fortresses, then heading into the Tree's ethereal insides, without ever feeling thrown together. This game does much better at capturing the interminable, oppressive sadness and claustrophobia of Falcom's Xanadu while providing real variety in environments. (I wish each town looked more unique after Eolis, but I doubt they had enough memory to achieve that.) By the game's finale, I felt that creeping dread on my back which the best gothic adventures provide, somehow captured within the confines of Hudson's simple premise and handiwork.

In a year of superlative action role-playing titans, Faxanadu takes a pratfall or two, desperate to keep up, and I can't help but admire it.

Inclement weather on the horizon: bullet rain, blood runoff in the streets, industrial parks desecrated by their own designs. No glorious anthems play in the background, just stings of death, the musique concrete of battle. A Wellsian steam-plane leads the charge, swerving and shooting past all that approach it. The pilot smirks, then grins in morbid delight. "Let 'em wail, like any of them can stop me!", he cries. Like a flying shark, the well-dressed dilletante in his top hat and goggles then gores the seas, cities, and alien motherships of his time. It's just the flight man's burden.

"Earth, 1999. But the Earth [is] of a different timeline. There are no microwave ovens and TV is still in black and white." Banshee's manual tells a simple premise: as Sven Svardensvart, steampunk viking he is, you must defeat the imposing Styx Empire, first by liberating a thinly-disguised United States from the aliens before hijacking their technology to escape Earth and face the Styxians on their own planet. It's a humans vs. invaders plot in a vertical arcade shooter, what more does one need? In the peanut gallery, I hear five words: "Anything but an Amiga shmup!" And make no mistake, I did a lot of research before settling on my first Western STG from the microcomputer glory days. There's a wide range in quality between, say, Tyrian and Xenon 2: Megablast (it's okay, Bitmap Brothers, you made up with The Chaos Engine later). American and European PC shooters tend to suffer in key ways which reflect either (a) developers' crippling isolation from Japanese arcades and fan communities around which major innovations could foment, and/or (b) a focus on presentation and pushing technology at the expense of fun, skillful play, and optimized controls. Extra negative points if the game in question targets a genuinely obscure or difficult-to-obtain platform like the 3DO or, in this case, the ill-fated Amiga CD32. With so many CD32-first games bombing commercially and critically, how well could this one hope to do?!

I'm happy to say that Banshee mostly avoids those pitfalls. This ended up less like a poor Scandinavian take on Toaplan, Capcom, and Konami STGs than I feared; at times I felt like I was instead playing the best parts of Flying Shark or 1943, now without the fuss or bumps in difficulty (at least until the end). Though hardly without flaws, or any ambition beyond a polished take on WWII-adjacent STGs, I was impressed by the first impression this title gave me, and those new expectations held up for the most part.

| Fly with me and the daring Swedes |

Banshee wouldn't exist without the rakish vision of Søren "Hannibal" Hansen and Jacob Anderson, who brought their game to Core Design and got the publishing greenlight for a cross-Europe release. To their credit, Core also provided Martin Iveson, their veteran in-house musician, to help score the small but groovy EDM soundtrack introducing and ending each playthrough. But beyond manufacturing, distributing, and paying royalty checks, Core left this program to its contract creators. There aren't any differences between the Amiga 1200 (floppy) and CD32 (uh, CD) versions, apart from loading times for the latter. Emulating the latter was actually more consistent on my end when playing in WinUAE, but I'll chalk that up to misconfiguration.

Right away the game launches into stage 1, a vast expanse of ocean spanning multiple screens' worth of playfield. Enemies only respawn within the player's view frustum, as is typical, and astute players will quickly notice the health-bar on left. Banshee only commits one of the supposed cardinal sins of Euro-shooters by letting your plane take multiple hits before meeting its violent end. Here, though, the developers have very good reason to give players some leeway: this game has no continues. Yep, you just read that. It's 1994 and some devs somewhere still had the courage to prevent quarter-feeding entirely. Just learning the first stage, then, becomes a priority simply to gain ground and attune to Hansen & Jacobsen's rhythm of design.

Of the more classic influences here, 1942 and Twinbee stick out the most, with loop-de-loop attacks and shootable power-ups in play. The former allows you to temporarily escape fire, damaging anything under your path in the process, while the latter gives players the chance to switch power-up types so long as they can keep the item on-screen. That 'ol juggling act's back, in steampunk form, with a need to both defend against enemy waves, evade tricky patterns, and keep the ship in tip-top shape for what's to come. I found myself using loops as infrequently as expected, while I took my time on each run flipping pickups until I got just what I needed. Need to wreck some large but ponderous aircraft? Cycle to the missiles and get ready to dodge more fire than normal. Scared of incessant UFOs and boulders on the horizon? Like most shooters, there's a perfectly usable spread-shot ability worth nabbing. Best of all is the "build-up" system. Upon collecting three of the eponymous upgrades, your best weapon status gets saved for your next life. This reduces the frustration of accidentally dying to a boss and only having a pea-shooter upon revival, or when forgetting to pick up shield upgrades partway through a level. Overall, there's solid meat-and-potatoes STG design happening here—nothing revelatory, but more considerate than its Amiga peers by and large.

| "Blardax Maldrear, uglier than a Phreenian Foot-toad" |

Making one's way to Planet Veenix, the throne-world of this story's antagonist, won't be easy even with the aforementioned abilities and lifelines. Banshee's true sin is not pacing itself better, featuring four very long levels which, while diverse in environs, tend to drag like in weaker Compile STGs. Couple this format with no way to continue and we've got ourselves an often frustrating adventure. Hitting 1-up thresholds and keeping the build-up meter at max becomes its own game, one you'd be foolish not to master. Hannibal & Jacobsen are perfectly happy to take their time, slowly iterating on enemy groups and shot patterns with breathing room to spare. This feels great when the run's going according to plan (or, in the learning stages, at least feels manageable), but one bad turn and the urge to restart grows. This urge comes upon me something fierce in stricter genre examples, like Raiden or Hishousame, so to sense it here saps some of my enjoyment. Many patterns and mini-bosses felt a bit underbaked, too, though not enough to ruin the experience. While this program could stand to run at a faster framerate, it's decently smooth as is, and the keyboards controls are as intuitive as the glossy brass UI.

Something this game never falters with is its B-movie sense of humor, both in visuals and sound design. I've seen a couple reviewers compare the art-style in Banshee to a mix of Cannon Fodder and Metal Slug (maybe some Boogie Wings as well), with deep earthy colors and attention to detail in the unlikeliest places. Pseudo-World War vehicles and soldiers all scramble to swat you down from the skies, yet their own deaths range from grisly to gallows humor. Burnt lackeys wither to bone and dust; city-goers mind their own business while a 1 vs. 1000 war rages across avenues; lighthouses, factories, and hulking ships all go up in brilliant flame and debris once the titular craft has its way. I'm not sure what's more sad, the fate of these brainwashed Gis or the thought of having one less unexpectedly hilarious STG to play. Canned PCM-audio cries of agony and constructs shredding into chaos make great use of the Amiga's Paula sound chip. And let's not sleep on Iveson's moody electronica during loading screens, a helpful reminder of just how '90s this game is.

The latter half of Banshee takes off the kid gloves, moving us from seaside bases and an Edwardian metropolis to icy, arid wastes before taking a climactic journey into the stars that'll make Metal Slug 3 blush. It's here where the no-continues, no-checkpoints structure can either heighten tension for the better or spoil a good run. The key factor which Hansen & Jacobsen nailed was balance; rarely did I ever feel unfairly outgunned or juked by the game. It's honestly designed from start to finish, even in the final half-hour's worth of intertwining projectiles and kamikaze attacks. Don't go in expecting a swerve towards danmaku leanings, or any overabundance of stage hazards (though some sequences dabble in them to great effect). Think Jamestown but with a more conventional arcade-y template and a comparatively refined minute-to-minute game loop. For lack of a true Sky Captain and the World of Tomorrow shoot-em-up, this can fill in the blank for the foreseeable future.

| Stir in an Alternative Dimension |

Core Design had a talent in the early '90s for featuring the best works of just anyone they could find off the proverbial Amiga street. Sure, most of their releases were self-developed projects, but they had a frequent collaborator in The 8th Day (Heimdall [2], Premiere) and adopted other outside teams' works a few times, always to great results. I think it's telling that something like Banshee arrived in time to herald the CD32's push for games that could directly compete with better offerings on NEC, Nintendo, and SEGA platforms. Don't get me wrong—these efforts weren't successful, and the CD32 was stillborn despite the occasional bright spot like this game. But there's a fascinating "what if?" to consider where Core, Team17, Psygnosis, Rainbow Arts, Ubi-Soft, Cryo, etc. put their all into accessible yet unique games that could have buoyed the ailing Amiga marketplace.

Banshee in particular feels like the kind of throwback arcade shooter which rarely happens today. Its level of sincerity, genre awareness, and muted presentation is a hard sell to those browsing digital stores now, expecting more "bang for your buck" features and runtime. But back in 1994, this could still pass for a full-price release; some groans about just four levels aside, there's plenty to engage with here when playing for score. I haven't yet tried Hard difficulty yet, but Normal feels, well, the way a normal and engaging STG of the time should. Press outlets all had loads of praise for the title, with many proclaiming it the new standard in Amiga shooters against a slate of competing games. Most of those accolades are hard to take for granted, given the dearth of strong original STGs on any Western PC by then, but reviewers and fan letters were on to something, however quaint their opinions have become.

The transition to 3D and multimedia-driven consoles, along with dwindling outlooks on the arcade industry and its game design trends, meant there'd be little room for something like Banshee to take root and thrive. If Japanese shooters were starting to have trouble staying relevant to all but a niche audience, then what more could those dubious Euroshooters hope for? Banshee could handily showcase some of the finest AGA-accelerated Amiga graphics around, but the pursuit of 2D excellence largely became the task of versus fighters and xRPGs. Hansen & Jacobsen must have seen these changes coming, soon joining Scavenger Inc. to work on their technically-outstanding games A.M.O.K. and Scorcher for PC and Sega Saturn. The coder eventually worked his way up into Shiny Entertainment and then Microsoft, while the artist found footing later at IO and now Unity. I hope this game, among other formative Amiga-era works, hasn't just become a footnote to overlook, but a case study in how to nail design fundamentals and make something more than the sum of its parts.

All my waffling aside, Banshee is worth one's time when looking for a solid mid-'90s PC shooter. Worst case, you could even play through this for the "I beat a Euroshmup!" credo without feeling robbed of that hypothetical better game. It's a bit janky in spots, but only noticeably so due to the game otherwise working as well as it does. There's enemy variety for days, a lonely but fitting soundscape, and four rounds of generally well-paced action to dig into. I doubt it sold any CD32 units back in the day, but that's no longer a concern when WinUAE makes this easy to boot up and enjoy with a controller of one's choice. Now, if only *I* could get my hands on an impossibly well-equipped biplane for my own battles…

Xtreme sports gaming and the British fascination with motorsports—name a more iconic combo. Then rizz it up with Fyre Festival's acceptable cousin and you get this game, predating the Forza Horizon series and the paradigm it's established. While modern Forza trades on its easy open-world structure and room for customization, MotorStorm: Pacific Rift hails from a time when impressive audiovisuals and tight arcade-y campaigns were more than enough. Ex-Psygnosis staffers at Evolution Studios not only had a competitor in Bizarre Creations to keep them honest, but enough cachet with Sony at large to try something this extravagant. The original MotorStorm played a key role in selling the PS3 to wider audiences, so all its sequel had to do was iterate louder, longer, and harder more confident than ever. Challenge accepted, I guess. In the end, Evolution made something reaching well beyond expectations, for better or worse, with much more stuff and challenge to offer.

| Bodies and bikes beyond repair |

Pacific Rift knows what it wants and how to deliver it. We're no longer on the continent, sliding through muddy canyons or atop arid cliffs, but having a once-in-a-lifetime demolition derby across a suspiciously tidy tropical isle. Like every amalgam of Hawaii, Tahiti, and New Zealand ever featured in games, this resort island has it all: pristine beaches, fiery volcanoes, dense jungle, and treacherous peaks to race upon. Let's put aside the cavalier desecration of Earth's last refuges, all for the amusement of the bourgeosie, and accept that Evolution just wants players to perform awe-inspiring feats of racing and stunt-craft. It's still the '90s in these developers' teary eyes, and that means no end of leagues, minigames, and gleefully impolite road raging for hours on end. We get nothing less than the decline of Western civilization dressed up as pop punk, off-roaders wearing brand sponsorships, and air pollution thick enough to send the crowds into a fugue. As a hypothetical funeral ceremony for the xtreme sports age, it ticks many boxes.

For as much sheer exhilaration as Evolution's down-and-dirty racer gives me, there's plenty more frustration than I had hoped for. This mainly boils down to excessive rubber-banding—lightweight early on, but quite noticeable heading into the later stages and leagues. If you ever end up on a teeny lil' bike or have to race big rigs in an ATV, then godspeed! The AI loves to punt, shove, shunt, and wreck the player as much it can after the opening hour or so of racing, which makes driving anything but the tankiest vehicles a chore when having to restart. Rather than give more leeway via a qualifying lap to learn each course, or a way to reduce/disable rubber-banding entirely, Pacific Rift enforces its "our way or bust" progression and difficulty balance to a fault. Not a fan.

| The smell of oil and gas in the air |

Thankfully the game offers 16 race-tracks and many variations on the iconic MotorStorm ride types to keep things varied. I especially love how the buggies, bikes, and rally cars handle throughout each level, with different track surfaces having a tangible effect on each vehicle. Controls are weighty, almost simcade-like compared to the golden age of Burnout, but refined from the prequel and nary hard enough to use when navigating traffic. Part of this ease comes from the level designs themselves, with ample space to corner around opponents and room enough to go several racers wide in spots. Evolution balks at the claustrophobia of other racers, instead asking "what if all our courses were as expansive as battlefields?" And they made the right choice. Watching and participating in the sheer carnage that is high-level Pacific Rift racing, from Knievel-ish leaps of faith to hardcore brawls in the dirt and sand with neither contestant coming out on top.

Customization, though present via choosing multiple driver skins and liveries, never becomes the centerpiece it ought to. I'd have loved to tinker around with custom skins and other ways of tuning up vehicles beyond a few selectable adjustments. Had I been around for the game's online scene, this would have brought some much-needed longevity into the game loop. With no way to really make any ride your own or set up custom race series, Evolution must have figured their average player was already overwhelmed and needed a simpler structure to keep the chaos parse-able. I've got no better way to explain why, despite building off the original MotorStorm's framework, the sequel lacks that ambition in meaningful playtime which the Horizon games at least try to attain. Regardless, I'd bet this ended up more fun with humans than when going up against an AI mainly interested in forcing you to wipeout, let alone race competitively.

| And the glint of a solitary shaft of chromium steel |

What I can't ruthlessly criticize is the presentation in Pacific Rift. No amount of aging textures, lower-poly environments, or carefully hacked VFX can take away from how succulent these graphics are. It's a great combo of arcade realism and baffling technical wizardry, something the PS3 could have used more of. Water ripples and shines convincingly, the earth molds and deforms according to tectonic and artificial pressures, and particles leap up from all directions onto the screen in so much detail. At most, I'll concede that this game and its prequel are some of the more visually busy racers one can play, but never enough to make following the racing line and other drivers difficult or annoying. I really wish I had proper minimap, however!

Sony just couldn't deny itself the chance to add a record store's worth of its own labels' licensed music, either, following the tradition they started with Gran Turismo. I struggle to remember a lot of these tracks, even ones from bands like Queens of the Stone Age, but damn if the overall atmosphere isn't fitting. Part of this pleasant amnesia comes from the game's sound design, which outdoes the predecessor's raucous mix of engine roars, crunching metal, and sudden silence when watching yourself careen in slow motion. Combine all this with plenty of post-processing (visual and aural) and the whole thing becomes phantasmagoric, a whirlwind of athleticism and sensory overload that would make Mad Max proud.

Despite only making small improvements over the formula and falling into a bit of an irritating, repetitive slump partway through, I had a damn fun time with MotorStorm: Pacific Rift, just as I hoped. Without having yet played Drive Club, it's easy for me to grasp why people mourn the loss of Evolution Studios, Bizarre, Studio Liverpool, and other arcade racer studios of old. This series shows almost precisely how you can translate the goofy, unrealistic blockbuster delights of '90s racing classics into the HD+ era, what with MotorStorm: Apocalypse going all in on surviving one set-piece after another. Here, though, the action's more organic, shaped by capricious crowds swerving and overtaking against each other to dodge peril and take the podium. It's tempting to compare this to kart racers, so I'll compromise and deem this more of a modern take on what SEGA's Power Drift evoked back in the '80s: a semblance of real motorsports corrupted by ballooning budgets, loud personalities, and proudly throwing caution to the wind. The eternal weekend of motor mayhem lives on.

More like Bromance of the Twink Kingdoms, amirite Backlo—shot

While I was away from Backloggd (long story), I got so far removed from my Japanese PC game schedule that I somehow drifted back around to it via Koei. The company's extensive library of sim/strategy classics has always taunted me from the gallery, begging me to try and parse its magnanimity. So here I am, marching across all of classical China with a growing retinue of generals, politicians, and all-around badasses trying to undo the mess the eunuchs and warlords have made. It ain't easy, but it's an honest living, and Romance of the Three Kingdoms VII was probably the best way I could have started with this franchise now that I've tried more of the PS2-era releases.

In short, this RoTK entry plays mostly like the others: choose a starting scenario from some point in the Three Kingdoms era (end of the Han to rise of the Jin), then try and reunite the land under one power through whatever means necessary. Before VII, this meant taking the role of liege (ruler, aka ye olde Cao Cao and the like) and campaigning against other lords for hegemony, first through obtaining officers and then by conquering each province. RoTK VII was a paradigm shift for the series, then 15 years old, by introducing "officer play". Now you start off as just one individual out of hundreds spanning the huaxia realm, meaning you can play your own custom officer or a vaunted icon of the period in whatever position they initially had. By doing this, RoTK expanded its definition of grand strategy to include more role-playing opportunities, where famous officials can resign to wander the land and other vagabonds could instead rise to significance.

The average game flow in RoTK VII, discounting leaders, has players working their way up from wanderers ("ronin" in Koei's parlage) to vassals (regular officers), prefects (city leaders), and lieges, always with the choice to forgo those positions and/or switch alliances. Contrasting the "dream" system of Romance of the Three Kingdoms VI, which set a lifelong aspiration for each character, this installment now gives you the Fame and Deeds statistics for measuring one's influence across China, arguably the endgame goal for everyone in the game. With both Northern and Southern Mystics periodically evaluating who has the greatest Fame in the land (and rewarding the best with money and a Bless skill), there's many reasons to prioritize hoarding Fame over just reuniting the land in record time to end a playthrough. I had a lot of fun helping out the common people by investigating corruption and banditry, or personally capturing the toughest officers in battle, just to hit the top 10 list of famous officers during my runs.

Every officer has an Action Points total determining how many actions one can take during a turn (one month). The game's very much a matter of managing your action economy, delegating minor tasks to others when possible while reserving important stuff for your own guy. Since many actions cost 50 AP for you and not others, this especially matters when working to develop a city's economy, comprised of agriculture, commerce, safety, and defense stats which mostly falter during warfare and natural disasters. Methodical development of cities is a great way to build fame and one's own stats, albeit at the cost of not visiting other regions, networking with officers and leaders, etc. Every RoTK installment has a challenging balancing act of priorities to manage, whether just for leaders or for all the potential characters to play, and RoTK VII ultimately favors new players a lot more than most of the games. It takes just 10 AP, out of likely above a hundred, to mail an officer and build bond points with them. And since most historical personas have hidden bonuses with others (reflecting their connections from the source novel), playing the likes of Guan Yu, Lu Xun, or the Xiahou brothers can be extra lenient for anyone wanting a smooth introduction to the series.

Winning the game technically only matters insofar as characters will die of old age, with the game moving you to the officer most bonded to your initial one. The open-ended progression and objectives of this RoTK design era makes it engaging just to pursue one's own excellence as much as leading your force to dominance over the country, but only the latter ends the game for good. And believe me, you'll spend a lot of time accomplishing that. There are more devious, "peaceful" means of swaying cities and leaders to your control, whether as liege or an underling, but combat's inevitable and the game's sketchy AI makes it risky to even engage, let alone defend against invasion. Because all movement across China happens rigidly through roads of different quality (ex. plains and waterways vs. mountain and gallery passes), it's smart to turtle up in a promising city, build up its economy and defenses, while developing an army and set of officers ready to march onward. The overworld's design makes it possible and entertaining to start a whole new force at a neglected corner of the map, traveling through other provinces to meet and acquire officers, before finally striking out and holding your own against greater powers via choke points and subterfuge.

Should any mix of induced riots, defections, coups, or annexation fail, there's always the battlefield. Battles in RoTK VII deviate from the hex-grids of old, instead using a Fire Emblem-esque square grid and turn-based systems to dictate combat flow. Battle prep starts with March commands or reacting to another force marching on you (or your allies requesting your reinforcements); this involves selecting officers/troop numbers, choosing weapon types (ex. crossbows, Rattan armor, cavalry et al.), and who leads as commander and tactician. Whether as offense or defense, you always get a briefing screen prior to matches where one can define strategy and ploys/traps. Combat units all move according to their WAR stats and troop counts—no one side moves all at once here—and winning fights sometimes boils down to just capturing the enemy commander or lowering a force's morale to 0. Many of my battles ended via abusing the latter condition to no end, often by squatting on the invaders' captured outposts to cut off their supply lines and thus their morale across turns.

Compared to what I've played of RoTK VIII (which in many ways exceeds this game), combat in RoTK VII plays much faster and distinguishes itself nicely vs. the slower, deliberate pace of the sequel. I had a grand 'ol time surrounding enemy troops with my own to gain massive damage boosts during attacks, or using weak but cunning officers' tactics and ranged attacks to disable and whittle down the game's most powerful units (think Lu Bu or Ma Chao). Battles themselves transition into castle sieges should some enemy units escape capture or retreat early, which results in a fun minigame where you ram down walls and use ploys like Feint or Riot to hurt the enemy without getting touched. Speaking of tactics and ploys, there's a bunch of them, some way better than others and restricted to field or siege modes. Pro tip: Rush, Rumor, and Rally ploys are supreme for getting faster reinforcements, scaring enemies off the map, and keeping your own morale/officer ailments in check. It's a great feeling when you finally get that Doctor skill item and can finally heal all those injured troops without having to camp at outposts for valuable turns on end (of which you only get 30 per battle).

Of all the myriad systems to master in RoTK VII, character skills are maybe the easiest and most personally satisfying. Making custom officers is engaging partly due to the chance to roll for some great starting skills such as Spy (better scouting), Trade (cheaper market prices), Reversal (turn an enemy's failed tactics against them), and it just goes on. RoTK VI had a similar skills framework before, but its successor fills out the roster and makes it easier to learn new skills via random when training (itself augmented by earning your base city's trust). The right combos of skills and stat-buff items can make any officer potentially a total winner! And that's why I think RoTK VII is maybe the easiest way to get into the series today, both because of its relative simplicity and the ease with which one can snowball. Worst case, a game ends quickly and you can move on to the next run, which might go far better. I can certainly see how the series earned its legacy in East Asia thanks to this addictive one-more-turn pace.

When comparing the PS2 and PC versions, I'm happy to say you aren't missing much with the former's lack of Power-Up Kit expansion. Later installments lean much heavier on their PUK to add critical features like child-raising for custom officers, but RoTK VII just predates Koei starting to punish players who don't/can't opt in. On the other hand, this entry's arguably the last which prizes sheer accessibility over all else, so it makes sense that a PUK wouldn't add much beyond a couple hundred extra officers (most of which are minor even by hardcore RoTK standards). Menus are easy to navigate in either version, though I bet the extra HUD elements on PC can help when figuring out strategies from the city screen. And while I ended up playing this much more for its mechanics and game loop, RoTK VII has solid presentation for its genre, with detailed 2D/pseudo-3D graphics (sober vs. Dynasty Warriors, but luscious) and a small but memorable soundtrack ranging from modern Chinese-styled orchestral to wild prog rock during combat. RoTK VII on PC now has an English fan patch for anyone unable to get a rare-ish PS2 copy, thankfully.

It's obviously hard to recommend most Koei strategy games to just anyone, given the lack of Western familiarity with stories like the Romance or Sengoku-era Japan for Nobunaga's Ambition, plus the complex systems at play. But I'm surprised these games aren't more widely played and discussed now in an age of successful grand strategy series from Paradox (ex. Crusader Kings) and Creative Assembly (Total War). Koei pioneered the genre and succeeded with it in ways that Western developers/publishers wouldn't for decades, and I'm unsurprised at how strong this earliest PS2 entry holds up. And it's insulting how Koei has run their oldest series into the ground via exploitative DLC and pricing schemes, making it easier to ignore them than ever. These games deserve much better treatment, that's for certain. While I can't wait to dig into more of their older PC output, RoTK VII might end up as my comfort Romance simulator of choice, just for how fast I can play through a what-if scenario and its more SRPG-esque battle dynamics. These qualities should make it a good starting place for others, too, who balk at the more complicated sequels and wonder how anyone could get onboard with RoTK wargaming.

tl;dr Do, in fact, pursue Lu Bu in this one. He's easy to confuse, can get misled into attacking his own guys, and is fun after the fight because he can tutor your own officer to become China's greatest warrior. Then proceed to curbstomp the countryside with him if one desires, but never forget about your cities' safety because it sucks when the mayors betray you for another ruler. Let the power of Rumor ploys compel you!