BELLEVUE, Washington—I was about to sit down to play Valve’s first new PC game series in five years. But this being Valve, the world’s first press demo for Artifact was preceded by something almost equally rare: a speech from company cofounder Gabe Newell. The speech wasn’t just a how-to of gameplay mechanics, nor was it focused on the day’s major surprise reveal—that Valve had hired Magic: The Gathering (MtG) creator Richard Garfield four years ago to start working on this game.

Instead, Newell gave a high-level overview of the game, with unexpected comments about the company’s history, corporate structure, and economics. Valve devotees tend to scrutinize every word uttered by Newell; he’s a five-years-ahead type of thinker in the games industry and generally plays his cards very close to the vest. And looking back, his Thursday evening speech feels like a compelling examination, and possibly a criticism, of where Valve has gone over the past few years.

“From a high-level perspective, we want to stay away from pay-to-win,” Newell said. Within nearly the same breath, he announced that Artifact will not function as a free-to-play game—you’ll have to buy something akin to a “starter” pack of gameplay cards, another staffer later clarified, before you can load into a game. Further cards can be purchased in packs and singles, either directly from Valve or from fellow players. And this, Newell insisted, will do something important for how the psychology of its paid cards plays out. “When you’re in a free-to-play environment, you get into a tendency that rarity equals power,” Newell said. He called this an “artificial relationship” and insisted that in Artifact, “that’s not the case at all. Lots of common cards will be super powerful.”

Flops and gongs

Newell wasted no time plunging into his grand vision for what could be the most ambitious digital trading card game (TCG) since Hearthstone: “Artifact is to trading card games what Half-Life 2 was to single-player action games.” Whoa, big words there, Gabe.

After Newell’s speech, attending press were seated at individual PC stations to play two Artifact matches against the game’s AI. This was our introduction to how the game works, as we started without any apparent tutorial baked into the game just yet (the current version is listed as 59.7456, if you’re keeping track). We each had an Artifact development team member hanging out as a guide, mentor, and (at least in my case) someone to tell me “just do this thing if you actually want to win.” That attendant made my goal explicit: destroy two of your opponent’s defense towers before it destroys two of yours.

You may have heard reports that Artifact would be a Dota-like TCG, and indeed, that’s what you get. In traditional TCG fashion, you build a deck of cards with various attack, defense, buff, and status-changing effects and then play them on a board against an opponent doing the same. You’ll also deal with the usual mana limitations for how much can be done in a single turn, with the supply growing automatically each round (though not all cards require mana to get them on the board).

But unlike most TCGs, Artifact leans into its Dota inspirations by making every game play out on three separate gameplay boards, akin to Dota 2’s three lanes of combat. This forces difficult decisions about where to place your five “hero” cards included in every Artifact deck, the lifeblood of how players get anywhere in this game.

The deck you bring into battle must include at least 40 “spell” cards to trigger those attacks, buffs, and so on. But to play those cards, in addition to mana, you need to have a matching hero of that card’s color (blue, green, red, or black) on a given gameplay board.

At all times, a variety of heroes and “creeps”—as in, wimpy filler cards that appear and accumulate on a regular basis—slash away at each other on all three game boards. These scrub cards don’t just kill the opposing heroes and creeps; they can also do damage to one of those three defense towers. Much like in Dota 2, a dead hero always returns to the action, though in Artifact‘s case, that happens after a rigid one-turn wait and not a variable pause based on how long a match has gone on.

Artifact incorporates some familiar Magic: The Gathering-like rules and systems with a few huge new strategic wrinkles. Sacrificing a hero can be costly in terms of that one-turn wait, but since the heroes always come back, such sacrifices seem to be a legitimate option in a way that doesn’t have a parallel in most other card games. With multiple battle lanes, you have decisions to make about which cards to use on which lanes—and furthermore, which color of heroes you want to send to which lanes. Swapping a living hero to another lane isn’t an automatic process, either; it requires a particular card in your deck.

And while cards from your deck can be played on any lane, each lane gets its own pool of mana. This can become interesting very quickly—especially when certain maneuvers can be triggered from one lane and applied to another one. Maybe you’d benefit from placing a green-card turret on your left lane, which will cause 2 points of defense tower damage every turn, but you only have a green hero on your center  lane, which itself is in dire straits. Do you blow your center-lane mana on the left lane turret? Do you spend that mana on a center-lane need? Or do you try to have it all by burning one of your valuable item cards (which cost money, an additional in-match currency beyond mana)?

Available spell cards include “reactive abilities,” which can pile on bonuses based on events that play out through an entire match. Some of these feel very Magic-like. One card applied to my hero Bristleback stacked two more armor points after every enemy card he personally killed, and that stacking bonus remained even if Bristleback died and came back. But others really get whimsical with Artifact‘s hero-driven system, like an attack card that does a certain amount of damage based on how many turns a particular hero had lived through up to that point. An attentive opponent, knowing that card could be coming, might go stomping after that blue Hero a little more quickly just to cut off that potential power.

Of the 40-card minimum for your deck, 15 slots are determined by the five hero cards you slap into it via three-spell sets that are attached to those heroes. Players must also include at least nine “item” cards in their decks, which are brought into a match by killing heroes and grunts to accrue gold that can be spent on items between rounds of combat.

Garfiiiiiield!

Notice that use of “minimum.” Artifact is really keen on not setting maximums. Do you want to have as many as 12 attacking cards laid down in a single lane? Go ahead. Want to build a deck with 200 spell cards? Valve doesn’t recommend doing so, but you can if you want.

When digital TCGs began to explode, Artifact team lead Richard Garfield told Ars that he was almost immediately frustrated with ones that simplified the genre’s mechanics. That didn’t bother him in terms of bringing in newcomers but rather in making the resulting gameplay feel “narrow.” He wanted to inject Magic-like open-endedness back into the genre, even as he admitted that Magic was never very good at translating to digital properties (he struggled with the conundrum since the first MtG video game port project began between Wizards of the Coast and Microprose in 1995.)

“There’s no reason not to get that [feeling] onto a computer!” Garfield told Ars. “A game where board state didn’t constantly clear itself to fit onto a telephone. We said, how many cards can you have? As many as you like! Creatures? Mana? I wanted those as big and open as possible.” Of course, a single day’s test of two decks got us nowhere near appreciating the impact of that openness on how the game may unfold among its harder-core players.

Garfield admitted that Artifact‘s basic concepts, of hero cards impacting what can be played in which lane, had existed in a digital game concept he’d been toying with before allying with Valve. It began as a “trading object game” prototype concept that he pitched to Valve roughly four years ago, he said, though his desire to make a robust, “open” TCG for computers and game systems had picked up in earnest roughly 10 years ago. When asked about comparisons to more modern digital TCGs, the game’s development team doubled down by claiming Artifact development began in earnest “before Hearthstone existed.”

When asked how his prototype and the Dota 2 universe came together, Garfield says the process was similar to his work on King of Tokyo, a board game that began as a “generic fantasy game” before evolving with the theme of kaiju destruction. “The basic concepts we were working with [on Artifact] were very flexible. There’s a lot of art and science in matching up an IP to a game mechanic and having it feel correct. If it wasn’t related to Dota, maybe it’d be six heroes per side. It’s just a few constraints.”

https://arstechnica.com/?p=1272431