When Pearl Abyss launched Crimson Desert on PC in late May 2025, the action-RPG arrived to strong reviews and brisk sales. It also arrived broken on Intel Arc graphics cards. Not just a little broken. Unplayable.
Players with Intel’s Arc B-series GPUs — the company’s latest and most competitive discrete graphics hardware — reported crashes so severe that the game was essentially inaccessible. For a company spending billions to establish itself as a credible third option in the PC graphics market, this was more than a technical hiccup. It was a public embarrassment that Intel says was entirely avoidable.
Intel’s response has been unusually blunt. In a statement provided to multiple outlets, the company said it had repeatedly offered Pearl Abyss engineering support to ensure compatibility with Arc GPUs before and after the game’s launch. Those offers, Intel claims, were ignored. “We have various different different ways of reaching out and offering support,” Intel told Engadget. “Different teams from different different different angles who have been reaching out and offering support to Pearl Abyss.” The company added: “They have been ignoring our offers of help.”
That’s a striking accusation from a major chipmaker directed at a game studio. It’s also a window into one of the most underappreciated friction points in PC gaming: the behind-the-scenes work required to make modern games run properly across multiple GPU architectures.
Pearl Abyss, the South Korean studio best known for Black Desert Online, has not publicly responded to Intel’s claims. The studio has, however, acknowledged the Arc compatibility issues and indicated it is working on fixes. As of early June 2025, some patches have improved stability, but the experience on Arc hardware still lags well behind what Nvidia and AMD users enjoy.
The situation matters because Intel’s discrete GPU business is at an inflection point. The Arc B-series, led by the Arc B580, launched in late 2024 to genuinely positive reception. Reviewers praised its price-to-performance ratio. It sold out repeatedly. For the first time, Intel had a GPU that people actually wanted to buy — not out of curiosity, but because it offered real value against Nvidia’s and AMD’s midrange options.
But hardware is only half the equation. Software compatibility — specifically, game developers writing code that works well on Intel’s architecture — remains Intel’s Achilles’ heel. Nvidia has spent decades cultivating developer relationships. Its GameWorks and now RTX programs embed Nvidia engineers directly into game studios during development. AMD has a similar, if less extensive, program. Intel is trying to build equivalent relationships from scratch, and the Crimson Desert situation shows just how difficult that is.
The technical root of the problem appears to involve how Crimson Desert handles certain rendering calls. Intel’s Arc GPUs use a different driver architecture than Nvidia’s or AMD’s, and games that rely heavily on vendor-specific optimizations or that don’t adhere strictly to graphics API standards can break in unexpected ways. This isn’t unique to Intel — AMD users have dealt with similar issues for years when developers optimized primarily for Nvidia. But Intel, as the newest entrant, faces the steepest climb.
According to reporting by Tom’s Hardware, Intel’s outreach to Pearl Abyss included offers of direct engineering assistance — the kind of hands-on support where Intel engineers would examine the game’s code and help implement fixes specific to Arc hardware. This is standard practice in the GPU industry. Nvidia and AMD do it routinely with major titles. The fact that Pearl Abyss apparently declined or simply didn’t respond raises questions about the studio’s priorities and resources.
There are possible explanations that don’t paint Pearl Abyss as negligent. Small and midsize studios — even ones with hit games — often operate with lean engineering teams. Adding support for a third GPU vendor means additional testing, additional bug fixing, and additional ongoing maintenance. Intel’s market share in discrete desktop GPUs remains in the low single digits. A studio under pressure to ship might reasonably, if regrettably, deprioritize the platform with the fewest users.
Reasonable or not, it’s a problem Intel can’t afford to let fester.
The company’s graphics division has been a money pit. Intel has invested billions in developing its Xe graphics architecture, building out driver teams, and marketing Arc products. The first-generation Arc A-series, launched in 2022, was plagued by driver issues and poor game compatibility. Intel took the criticism, kept investing, and delivered a markedly better product with the B-series. But if major game launches continue to be broken on Arc hardware, consumer trust — painstakingly built over the past year — will erode fast.
And trust, once lost in the GPU market, is extraordinarily hard to regain. AMD spent the better part of a decade fighting the perception that its drivers were unreliable, a reputation that lingered long after the underlying issues were resolved. Intel doesn’t have a decade. The company is under intense financial pressure, with CEO Lip-Bu Tan executing a sweeping restructuring that has already resulted in thousands of layoffs. Every product line needs to justify its existence.
The broader context makes this even more fraught. Crimson Desert isn’t some niche indie title. It’s one of the most anticipated PC games of 2025, built on Pearl Abyss’s proprietary engine and featuring the kind of graphically demanding open-world environments that stress GPUs in exactly the ways that expose compatibility gaps. When a game this prominent doesn’t work on your hardware at launch, it sends a signal to every potential buyer: maybe wait before switching to Intel.
Intel has been proactive about building its game compatibility infrastructure. The company maintains a “Game On” driver program that releases optimized drivers timed to major game launches — similar to Nvidia’s Game Ready drivers and AMD’s day-one driver updates. For Crimson Desert, Intel issued driver updates aimed at improving stability. But there are limits to what driver-side fixes can accomplish when the game itself contains code paths that are fundamentally incompatible with a given GPU architecture. Sometimes the fix has to come from the developer’s side.
This is where Intel’s accusation gains its sharpest edge. The company isn’t just saying the game doesn’t work on Arc. It’s saying it tried to help fix it and was rebuffed. That framing shifts responsibility squarely onto Pearl Abyss and, by extension, onto any developer who chooses not to engage with Intel’s compatibility programs.
It’s a risky move. Publicly calling out a developer can alienate other studios. Game developers talk to each other. No one wants to be the next studio named in a chipmaker’s press statement. But Intel may have calculated that the greater risk lies in silence — in letting the narrative be “Intel GPUs can’t run major games” rather than “this developer didn’t do the work.”
The incident also highlights a structural disadvantage Intel faces. Nvidia’s dominance in discrete GPUs — the company holds roughly 80% market share — means developers naturally test on Nvidia hardware first, often exclusively during early development. AMD’s roughly 15% share earns it second-priority status. Intel’s sub-5% share puts it last. This creates a self-reinforcing cycle: developers don’t prioritize Intel because few gamers use Intel GPUs, and gamers don’t buy Intel GPUs because games don’t reliably work on them.
Breaking that cycle requires either a dramatic increase in Intel’s market share (unlikely in the short term), financial incentives for developers to prioritize Intel compatibility (expensive and unsustainable), or exactly the kind of proactive engineering support Intel says it offered Pearl Abyss. When that support is refused, Intel’s options narrow considerably.
Some industry observers have pointed to the console market as an instructive parallel. When Microsoft launched the original Xbox, it faced a similar chicken-and-egg problem with game support. Microsoft’s solution was aggressive: it bought studios, funded exclusive titles, and spent heavily on developer relations. Intel doesn’t have the option of buying game studios, but it could increase the financial incentives it offers for Arc optimization. Whether the economics of its GPU business support that level of spending is another question entirely.
For now, the practical impact on Crimson Desert players with Arc GPUs is straightforward frustration. Community forums and Reddit threads are filled with reports of crashes, graphical artifacts, and performance well below what the hardware should deliver. Some users have found partial workarounds — disabling certain graphical features, rolling back to older drivers — but none of these are real solutions.
Pearl Abyss has released several patches since launch, and the situation has improved incrementally. But “incrementally” isn’t good enough when your competitors’ hardware runs the game fine out of the box. Every day that Crimson Desert remains broken on Arc is a day Intel’s investment in discrete graphics yields a negative return — not in dollars, but in credibility.
The question going forward is whether this incident represents an anomaly or a pattern. If Crimson Desert is a one-off — a single stubborn developer that refused help — Intel can absorb the hit and move on. But if other major titles launch with similar Arc incompatibilities in the months ahead, the market will draw its own conclusions, regardless of who’s at fault.
Intel’s willingness to go public with its frustration suggests the company believes transparency is its best weapon here. It may be right. PC gamers are a technically sophisticated audience. They understand that compatibility is a two-way street. Showing that Intel did its part — and that the developer didn’t — could preserve goodwill in ways that quiet diplomacy wouldn’t.
But it also sets a precedent. Every future game that breaks on Arc will invite the question: did Intel offer help, and was it accepted? If the answer isn’t consistently “yes and yes,” the transparency strategy becomes a liability.
The stakes extend beyond any single game. Intel’s ability to compete in discrete graphics — and to justify the billions it has spent getting here — depends on games working. Not most games. All major games. That’s the standard Nvidia has set, and it’s the standard Intel will be measured against, fairly or not. Crimson Desert is a reminder that getting the hardware right was the easy part. The hard part is everything else.
WebProNews is an iEntry Publication