The best tactic we’ve come up with is the Cybrannosaurus Bubblebath. A hovering triangle wafts into your base and drops a dinosaur on you. The dinosaur is large, about the size of a dinosaur, and has a robotic head that breathes fire. A moment later, 25 translucent spheres pop up around it. The shields these Adaptors generate aren’t impenetrable, but they overlap and regenerate. They cling to the dinosaur as missiles spill from his back like he’s moulting. He stomps toward your commander as if to eat him, but of course he won’t. He’s an herbivore; that would be ridiculous.
Supreme Commander is like the fever dream of a Robot Wars contestant: you control hundreds of killing machines as they clash with hundreds more over land, air and sea. In the first game that got complicated, and if you weren’t an actual Robot Wars contestant you could be forgiven for giving up. The slightest error in establishing your economy could cause it to crash, leaving you crippled while your opponent’s forces spread like a metal virus. If that was you: good news, come in. You’re going to love this.
If that wasn’t you, if you spent your Sundays in eight-hour matches battling seven other commanders for control of a battlefield the size of the Isle of Wight, it’s best not to think of this as a sequel. It’s more like a side-project, as if Giant Robots and Vast Armies left the group to form their own band without Epic Battlefields and Advanced Economics – who was always kind of a square. We don’t know whether the old band will ever get back together, but don’t go into this expecting the nerdy glory of those four in concert. You could easily miss why Supreme Commander 2 is great.
There is a counter to the Cybrannosaurus Bubblebath, we’ve discovered. It’s called the ‘Screw you, I have a Magnetron.’ You drop your dino, the shields go up, and said shields suddenly jerk across the base and are minced in a maw of spinning metal teeth. Rex, wagging his flame-gouting head from side to side to try to see what’s happening, is dragged slowly backwards into the sparky deathcogs by a megaelectromagnet, where they chew through his flesh and metal with indifference. A Magnetron would eat your commander. A Magnetron would eat God if He had any metal on Him.
This is what’s great about SupCom 2: very different units, eating each other. In SupCom 1, if we got a Tech 3 factory up before you, we could produce a unit to deal with anything you threw at us. In SupCom 2, if you come at us with nanoshielded gunships and we’ve spent all our Research Points unlocking the Fatboy experimental mega-tank, we’re boned. Your decision about what to spend resources on is now more important than how efficiently you produce them.
Rather than hiding these options away in icons that only pop up once you’ve built a certain factory, everything you can construct and upgrade is laid out on a Research screen. Land units, air units, naval units, your structures and your commander each have a separate tech-tree of upgrades and unlockable units, including huge Experimentals.
The Research Points you spend on this stuff accumulate over time – faster when you kill things, and faster still once you’ve built some labs. Because they’re scarce early on, your strategy almost always revolves around the shortest possible route to unlocking something major.
By the time you’ve got it, of course, Research Points are coming thick and fast, so good strategies tend to have a phase two. That’s how the Cybrannosaurus Bubblebath came about: the dino takes a long time to earn and is very slow. By the time we’ve built him, we’re earning Research Points fast enough to quickly unlock the Experimental Air Transport and give him a ride to the enemy base.
Another strategy we like is Fire a Nuclear Warhead at Them, Killing Them. It doesn’t always work. If the enemy knows what you’re up to, they can build counter-nukes to intercept. That’s why you should back it up with the Illuminate Space Temple: it lets you teleport a strike force inside your enemy’s shields, destroy their nuclear defence silo, and teleport back just in time to see your warhead launch. Be careful not to play in the same room as someone you’re going to do this to, though, as it could easily trigger a counter-attack of Being Punched in the Face.
So the biggest change in SupCom 2 is a positive one: a clear and fun tech system that gets you coming up with two-phase plans and counter-strategies involving robots, magnets, nukes and bubblebath.
It also scores points for running faster than its predecessor: we get 40fps when zoomed-in during a six-player match of the first game, 80 when zoomed out. In the same sized conflict, SupCom 2 glides along at 90fps zoomed in, 70 zoomed out.
The rest of the changes are bad, but not disastrous. The worst is that the AI this time is just limp. It puts up a decent fight in the campaign, where missions stack the odds hugely in its favour, but in Skirmish it can’t match even an average player like us. We could only take on the smartest AIs in SupCom by making them fight each other: here, we can reliably beat a team of four. The only tougher setting is Cheating, which nukes you in a few minutes. We’re not looking to get obliterated; we’d just like an AI that can navigate the tech tree well enough to build an Experimental when we do. Currently, the only way to see a clash of the titans is to make a cup of tea after building yours.
It’s also smaller game. There’s only one eight-player map here, and it’s a drab, shrunk rehash of one from the first game: Seton’s Clutch. It’s still great, but that makes it all the more maddening: there should be dozens like this, and bigger still, because the tech system works even better for large-scale war. It’s only on a huge battleground that strategies of position, timing and logistics come into play: SupCom 1 was a deliberate demonstration of that.
Despite the AI and map size, Skirmish is still the best way to play. The other two modes, Multiplayer and Campaign, each have their own trouble with the new Research Trees. In multiplayer, some strategies just seem uncounterable. Most of your time-investment in a strategy – spending Research Points – happens in secret.
So by the time we see you start to build a nuke silo, it’s too late for us to earn enough Research Points to unlock nuke defence before you obliterate us. Similarly, if we research a Cybran Soul Ripper gunship, by the time you see us building one it’s too late to get up enough anti-air to destroy it.
The campaign is surprisingly worthwhile. You have to play as each of the three factions in turn, but the UEF commander you start as is an unusually likeable chap. When an early villain laughs, “You really think those Fatboys will stop me?” Maddox says exactly what we would have: “Honestly? Yeah.”
There are still way too many missions where your objective is "Defend this shitty base I shittily made for you against an unknown number of unknown enemies from an unknown direction until - too late! they got past the turrets I pointlessly placed miles from the base by coming from an angle I forgot to say they might come from."
The bigger problem is that each faction’s tech-trees are restricted until the final mission, so you get one chance per race to try all of the most interesting stuff in the game. SupCom 2 is all about the tech trees: locking bits of them off reduces the campaign to a 12-hour tutorial.
So SupCom 2 is a great game struggling to find the right format. Skirmish is as close as it gets: it’s enormous fun despite the toothless AI, particularly if you play with a friend and stack the bots against you. We play every lunchtime now, and talk constantly about what strategy we’re going to try next.
What makes it so infectious is partly the diverse and ridiculous units, partly the way their specialised weaponry tesselates, and partly the system for unlocking them. It keeps you thinking about the roads not taken, the pairings not combined, and the hyperdeathbots unbuilt.
