Radeon RX 9060 XT vs Arc B580: 1080p Rasterization Tested

Two Budget GPUs, One Battleground

AMD’s Radeon RX 9060 XT and Intel’s Arc B580 are squaring off in the space that matters most to most PC gamers: 1080p rasterization, where raw frame rates and game compatibility define daily life at the desk. Both cards target the sub-$300 sweet spot, but they approach that price point from very different architectural directions – and the results are messier than either company would prefer.

Specs, Price, and What You’re Actually Buying

The RX 9060 XT arrives on AMD’s RDNA 4 architecture with 16GB of GDDR6 memory on the higher-tier SKU (an 8GB variant also exists), sitting at a launch MSRP of $299. Intel’s Arc B580 launched earlier at $249 with 12GB of GDDR6, making it the cheaper card by a meaningful margin at street price. That $50 difference is real money at this tier, and it matters when both cards are fishing in the same pond of cost-conscious buyers.

On paper, the 9060 XT carries more compute units and higher memory bandwidth than the B580, but Intel’s Xe2 architecture – the same foundation underpinning the B580 – has proven that raw specification comparisons don’t always map cleanly to gaming performance. Intel rebuilt its driver stack dramatically from Alchemist to Battlemage, and the B580 launched with noticeably better day-one game support than Arc cards had historically delivered. AMD, by contrast, carries RDNA 4’s strong rasterization improvements over RDNA 3, particularly in compute-heavy workloads.

Memory configuration is also worth flagging. The 9060 XT’s 16GB option gives it a significant buffer for texture-heavy games and future titles pushing higher VRAM loads, while the B580’s 12GB is still generous relative to competing NVIDIA options in the same price range. Neither card is memory-starved at 1080p today, but the gap becomes relevant when you’re also considering whether this GPU carries you through the next two or three years of releases.

Thermal and power profiles differ too. The B580 has a rated TDP of 190W, while AMD’s 9060 XT lands around 150-170W depending on the board partner configuration. That efficiency edge for AMD is real, particularly for smaller form-factor builds or systems running older power supplies. Both cards are single 8-pin connector territory, so neither requires a platform upgrade to install.

1080p Rasterization: Game by Game

Testing across a cross-section of titles at 1080p max settings, the RX 9060 XT holds an average performance lead over the Arc B580 of roughly 10 to 15 percent in most rasterization-heavy games. In titles like Cyberpunk 2077 (without ray tracing), Spider-Man: Miles Morales, and Assassin’s Creed Mirage, the 9060 XT consistently stays above the B580 by a visible margin – not a generational gap, but enough that frame pacing charts look meaningfully smoother on the AMD side.

Where Intel closes the gap is in DX12 and Vulkan titles that benefit from the B580’s strong async compute implementation. In games like Baldur’s Gate 3 and Forza Horizon 5, the gap between the two cards narrows to within 5 percent, and on a few specific in-game sequences the B580 briefly leads. This isn’t a consistent pattern across the board, but it’s enough to suggest that Intel’s Xe2 architecture has real strengths in specific workload types, and those strengths align with the API direction most modern games are moving toward.

The 1 percent low frametimes tell a more complicated story. The 9060 XT’s 1 percent lows are consistently better in DirectX 11 titles – games like older competitive shooters or titles that haven’t received API updates – because Intel’s Arc driver stack still carries legacy compatibility overhead in DX11. If your game library skews toward esports titles or older releases, this is a genuine consideration. Call of Duty: Warzone, Valorant at CPU-limited 1080p settings, and CS2 all run with less frame-time variance on the AMD card.

One area where the B580 genuinely stands out is XeSS upscaling. Intel’s proprietary upscaler has continued to improve and at 1080p output, XeSS Performance mode produces a cleaner image in supported titles than AMD’s FSR 3 at equivalent quality settings. If you’re planning to use upscaling regularly – and at this price tier, many players will – the B580 has an argument to make. AMD’s FSR works across more games by design since it’s open and doesn’t require developer integration, but image quality per setting tier currently favors XeSS where support exists.

Power consumption under sustained gaming load is one area where the 9060 XT pulls ahead cleanly. Running at roughly 30 to 40 watts less in comparable gaming scenarios, the AMD card produces noticeably less heat and runs quieter on most board partner cooler designs. For anyone gaming in a compact case or a warm room, that practical difference adds up over a long session.

The Value Argument and Where It Actually Lands

At MSRP, the Arc B580 at $249 makes a strong case for itself purely on price-to-performance ratio. You’re giving up that 10-15 percent average performance advantage, but you’re paying 20 percent less, and the B580’s 12GB framebuffer keeps it competitive in everything being played at 1080p today. The Radeon RX 9070 vs RTX 5070 comparison at 4K shows how AMD’s RDNA 4 architecture scales at higher resolutions, but the 9060 XT operates in a different competitive environment where Intel is actually fighting hard.

The RX 9060 XT earns its higher price if you’re buying at MSRP and value consistency across your whole library – especially if DX11 titles, legacy games, or competitive shooters make up a chunk of your hours. The 16GB VRAM option also extends the card’s relevance meaningfully. But if the B580 is available at or below its $249 MSRP at your local retailer, turning that price gap into a pure performance argument requires ignoring the fact that most buyers at this tier are weighing dollars directly against frame rates. At identical prices, the 9060 XT wins clearly. At the prices they actually launched at, it’s genuinely close.