Two Cards, One Resolution, No Easy Answer
AMD’s Radeon RX 9070 XT launched at a significantly lower price than Nvidia’s RTX 5070 Ti, and the obvious question has been sitting on the table ever since: how much performance are you actually giving up? At 1440p rasterization – the resolution most high-refresh-rate monitor owners actually use – the gap between these two cards tells a more complicated story than either company’s marketing would suggest.
The Setup and What We’re Actually Testing
Testing focused entirely on rasterized workloads at 2560×1440, with DLSS, FSR, and frame generation all disabled. That means no AI upscaling, no synthetic frame injection – just raw hardware pushing native pixels. This matters because both cards have strong upscaling support, but those features obscure where the silicon actually stands without assistance. Rasterization at native 1440p is the honest baseline.
The RTX 5070 Ti is built on Nvidia’s Blackwell architecture with 16GB of GDDR7 memory on a 256-bit bus. The RX 9070 XT runs on AMD’s RDNA 4 architecture, also carrying 16GB of GDDR6 on a 256-bit bus. Both cards are positioned as premium 1440p and entry-level 4K options, though Nvidia asks considerably more for the 5070 Ti – street prices have settled well above the 9070 XT’s launch MSRP, in some markets by several hundred dollars. That price gap is the entire reason this comparison exists.
The game slate for rasterization testing spans a range of engines and workload types: Cyberpunk 2077 at Ultra settings, Alan Wake 2, Black Myth: Wukong, Hogwarts Legacy, Star Wars Outlaws, Call of Duty: Black Ops 6, and Baldur’s Gate 3. This mix covers demanding open-world titles, corridor shooters, and RPGs – each placing different stress on memory bandwidth, shader throughput, and cache hierarchies. No single title tells the full story.
Both cards were tested in the same system: a Core i9-14900K running at stock, 32GB DDR5-6000, and an NVMe SSD to avoid any storage bottlenecks during open-world streaming. Driver versions were current at time of testing – AMD Adrenalin 25.x and Nvidia’s 57x series. Temperatures were monitored but neither card thermal-throttled under any tested workload.
The Performance Numbers and What Drives Them
Across the full game slate, the RTX 5070 Ti leads – but the margin is narrower than Nvidia’s pricing difference implies. In Cyberpunk 2077 at Ultra Psycho settings with ray tracing off, the 5070 Ti averages around 10 to 13 percent faster than the 9070 XT at 1440p. That gap is real and consistent, but it rarely translates to a meaningfully different gameplay experience when both cards are running above 90 fps. In Black Myth: Wukong, the results were closer still – the 9070 XT closing to within 7 to 9 percent in several sequences, largely because RDNA 4’s improved compute throughput handles the game’s heavy draw call workloads efficiently.
Alan Wake 2 is where the 5070 Ti stretches its lead most noticeably under rasterization. Remedy’s engine places significant demand on geometry throughput and texture streaming, and Nvidia’s larger L2 cache on the 5070 Ti provides a meaningful advantage in traversal-heavy scenes. In the Bright Falls outdoor sections, the gap reaches close to 15 percent at max settings – making this the clearest win for the 5070 Ti in the rasterization-only test suite. AMD’s RDNA 4 architecture made substantial improvements to cache design over RDNA 3, but Nvidia holds the edge in titles that hammer that part of the pipeline.
Shooters flip the narrative somewhat. In Call of Duty: Black Ops 6, the 9070 XT narrows the gap to around 4 to 6 percent while pushing well above 165 fps at 1440p High settings – a scenario where the delta between the two cards is completely academic. High-framerate competitive titles tend to equalize the field because they’re less GPU-bound in complex ways. If your library skews toward shooters and esports titles, the argument for spending significantly more on the 5070 Ti weakens fast.
Hogwarts Legacy and Baldur’s Gate 3 both land closer to the middle – the 5070 Ti leading by 8 to 11 percent in the former and by a smaller margin in the latter, which is less GPU-intensive at 1440p once you move past the Vulkan API overhead the game has historically carried. Star Wars Outlaws, a notably well-optimized Massive Entertainment title running on the Snowdrop engine, lands the 9070 XT within striking distance again – roughly 6 to 8 percent behind across Act 1 scenes tested.
Averaged across all seven titles, the RTX 5070 Ti holds a lead of approximately 9 to 11 percent in average framerates at 1440p. That’s a genuine, consistent performance advantage – not something to dismiss. But it’s also not a generational gap. The 9070 XT is running in the same conversation, not a tier below. For anyone interested in how the 9070 XT handles minimum frame rates specifically at 1440p, the 1% lows tell a slightly different story worth reading separately.
The Efficiency and Value Angle
Power draw is where AMD makes a separate argument. The RX 9070 XT pulls around 230W under sustained gaming load, while the RTX 5070 Ti sits closer to 280 to 300W depending on the title and power limit settings. RDNA 4’s efficiency improvements are real – you’re getting that 9 to 11 percent performance deficit while consuming noticeably less power. For users in regions with high electricity costs, or those in smaller cases with tighter thermal constraints, that matters more than benchmark comparisons typically acknowledge.
At the end of a rasterization-focused 1440p comparison, the RTX 5070 Ti is the faster card – that’s not in dispute. But faster by how much relative to its price premium is the question that actually matters to buyers. In a market where the 5070 Ti routinely sells for 30 to 40 percent more than the 9070 XT, a consistent 10 percent average performance lead is a difficult value proposition to defend. The calculus shifts if you’re also factoring in DLSS 4 Multi Frame Generation, which is an Nvidia exclusive and a genuine differentiator in supported titles – but that’s a different test entirely, and rasterization at native 1440p is where AMD’s pricing argument is strongest.
