Crimson Desert Review: The Game That Scored a 6 and Finished the Year in the Top Five

Crimson Desert arrived in March carrying six years of expectation and immediately split the room. Review scores landed around a six. A lot of best-of lists dropped it entirely. And then, three months later, when players were asked what they had actually loved most this year, Crimson Desert finished in the top five.

That gap is not a mystery and it is not people being wrong. Having put a very large number of hours into this game, we think both reactions are honest descriptions of it. They are just descriptions of different parts of it, and the part you get depends almost entirely on how long you stay.

Here is the blunt version. The first eight hours of Crimson Desert are bad. Not disappointing — bad. Boring quests, no reason to care, systems dribbled out one at a time, an inventory that fights you, and a story that gives you nothing to hold on to. If you were reviewing this on a deadline against a ninety-hour game, those eight hours were a meaningful chunk of what you saw. And then somewhere around hour ten it opens up, and it keeps opening up for the next eighty, and it becomes one of the most generous games anybody has made in years.

You play as Cliff, a warrior of the Grey Mains, who is killed by a rival faction in the opening and brought back by something that does not explain itself. The early game asks you to find your people, build a camp, and get your bearings. On paper that is a fine setup. In practice it is hours of errands with almost no narrative pull behind them, in a game that is simultaneously trying to teach you two dozen systems by trickling them out one at a time.

It also glosses over things. Mechanics appear without explanation. Some of the most useful abilities in the game are locked behind side content you have no reason to seek out yet. And the inventory at launch was so restrictive that reviewers reported spending a quarter of their playing time in menus. Pearl Abyss raised the starting slots from twenty to fifty during the review period after feedback, which helped enormously, and which also tells you how bad it was.

We are being this direct about it because it is the single most useful thing we can tell you. Everything that people disliked about Crimson Desert is concentrated in its opening, and everything that people love about it is spread across the eighty hours after. If you bounce off this game at hour five, your opinion of it is accurate and incomplete at the same time.

It starts almost too simple. Light attack, heavy attack, block, parry. Then you get a force push. Then a grappling hand that works a lot like Tears of the Kingdom's Ultrahand, letting you seize enemies and rearrange the environment. Then a skill tree where individual nodes have multiple upgrade paths, each unlocking new moves. Eighty hours in we were still finding combinations we had not tried.

What makes it feel different from most action games is how physical it is. You can throw people. You can clothesline them. You can grab someone, drag them somewhere else, and drop them off a thing. Swords land with real weight and a small punctuation of gore. It is expressive in a way that invites messing about rather than optimising, which is the correct instinct for a game this size.

Character swapping deepens it further. You switch between playable characters in the style of GTA 5, and each one has their own weapons, their own skill tree and their own gear. These are not reskins. The downside is administrative — you have to carry everyone's equipment and swap it manually, skills do not appear to transfer between characters despite earlier indications they would, and you end up permanently short of skill points across several trees at once.

The one place combat overreaches is crowd size. Enemy camps ask you to grind a percentage counter down to zero, and that can mean thirty unbroken minutes of cutting through dozens of bodies. With that many enemies you find one reliable tactic and repeat it, because giving them room to react gets you killed. Smaller, sharper encounters would have shown the combat system off far better than a mob does.

The puzzles are the worst of it. Some are optional and fine to walk away from. Others sit directly in the main story and will stop you dead, because they introduce mechanics the game has not explained and offer no hints of any kind. During the review period Pearl Abyss set up a Discord so that reviewers could share puzzle solutions with each other. That is an extraordinary thing to need to do, and it means you should expect to look up a guide at some point.

Bosses are wildly inconsistent. The good ones are excellent — long, tactical, properly staged, and a real test of whether your gear investment paid off. The bad ones are simply too aggressive to react to, and dying to one is expensive: fast travel points are sparse, you often respawn a long way back, and you may have to redo a puzzle you already solved to get another attempt. Between-phase cutscenes can be sped up but not skipped. Some of the early bosses are harder than later ones, so the difficulty curve wanders.

Then there are the bugs. Individually most are small. Collectively they are constant, and they get noticeably worse in the back half — including quest bugs that can block progression. The day one patch cleared out a lot and Pearl Abyss has been unusually quick to respond, but the honest read is that this game needed longer in development.

And the story never really arrives. There are strong moments, particularly the quieter ones, and the sense of scale between warring factions is genuinely Game of Thrones-ish. But Cliff is not a character you will think about afterwards, the flashback-helmet device used to deliver backstory is a poor substitute for actual cutscenes, and we finished the game feeling attached to the world and to nobody in it. The world is the main character. That is a real limitation, and it is also, for this particular game, survivable.