AMD Radeon 890M vs NVIDIA GeForce RTX 5070 Ti SUPER Comparison

AMD Radeon 890M vs NVIDIA GeForce RTX 5070 Ti SUPER

Where Each One Wins

The recorded data splits these two GPUs into entirely different performance classes. The NVIDIA GeForce RTX 5070 Ti SUPER wins the single shared benchmark, 3DMark Steel Nomad DX12, by a decisive margin. The AMD Radeon 890M does not win any of the recorded head-to-head comparisons. The database records the NVIDIA part at a 90.6% higher score in that specific test, which is the only directly comparable measurement between the two products.

That said, the database contains a broader set of benchmarks for the AMD Radeon 890M, covering older DirectX versions, compute workloads, and 2D tests. The NVIDIA GeForce RTX 5070 Ti SUPER has only one recorded benchmark score, so the comparison is limited. The AMD part shows a Passmark G3D score of 8076, a Passmark GPU Compute score of 4157, and a Geekbench Vulkan score of 40808. The NVIDIA part has no equivalent entries in these tests, so no direct comparison is possible outside of the Steel Nomad result.

The use-case split is therefore clear from the available data. The NVIDIA GeForce RTX 5070 Ti SUPER is the choice for modern DirectX 12 workloads, particularly the demanding Steel Nomad scene. The AMD Radeon 890M, with its Passmark DirectX 9 score of 97 and DirectX 11 score of 73, retains relative strength in legacy API scenarios, though its absolute scores are far below the NVIDIA part. The AMD GPU also shows compute capability through its Geekbench OpenCL score of 37254, but again, no comparable NVIDIA measurement exists in the database.

The AMD part sits at the 45th percentile of all GPUs, while the NVIDIA part sits at the 36th percentile. This is a counterintuitive result given the large benchmark gap, but it reflects the broader distribution of scores in the database. The AMD Radeon 890M's average benchmark score of 9210 places it within 0.1% of the AMD Radeon Vega 8, while the NVIDIA GeForce RTX 5070 Ti SUPER's average score of 6270 exactly matches the NVIDIA GeForce RTX 4070 Ti SUPER AD102.

Architecture Differences

The two GPUs come from different manufacturers and different architectural generations. The AMD Radeon 890M uses the Strix Point chip built on RDNA 3.5 architecture, fabricated on a 4 nm process at TSMC with 34,000 million transistors on a 233 mm² die. The NVIDIA GeForce RTX 5070 Ti SUPER uses the GB203 chip built on Blackwell 2.0 architecture, fabricated on a 5 nm process at TSMC with 45,600 million transistors on a 378 mm² die. The AMD chip has a higher transistor density at 145.9 million per mm², while the NVIDIA chip has 120.6 million per mm².

The execution resources differ substantially. The AMD Radeon 890M has 1024 shading units, 64 texture mapping units, and 32 raster output units. The NVIDIA GeForce RTX 5070 Ti SUPER has 8960 shading units, 280 texture mapping units, and 96 raster output units. The NVIDIA part also carries 70 ray tracing cores and 280 tensor cores, while the AMD part has 16 ray tracing cores and no recorded tensor cores. These resource counts explain the raw throughput differences: the NVIDIA part achieves 43.94 TFLOPS FP32 and 686.6 GTexel/s, while the AMD part achieves 5.939 TFLOPS FP32 and 185.6 GTexel/s.

Memory architecture is another major divider. The AMD Radeon 890M uses system shared memory, with bandwidth described as system dependent and no dedicated VRAM. The NVIDIA GeForce RTX 5070 Ti SUPER has 16 GB of GDDR7 memory on a 256 bit bus, delivering 896.0 GB/s of bandwidth. This means the NVIDIA part has dedicated, high-bandwidth memory for rendering, while the AMD part relies on the host system's memory controllers.

Clock behavior also differs. The AMD part has a base clock of 400 MHz and a boost clock of 2900 MHz, while the NVIDIA part has a base clock of 2295 MHz and a boost clock of 2452 MHz. The AMD part's wide clock range reflects its integrated design, which can scale down aggressively when idle. The NVIDIA part runs closer to a fixed high frequency. The NVIDIA memory clock is listed at 1750 MHz with 28 Gbps effective, while the AMD memory clock is listed as system shared.

Power and physical design separate the two products further. The AMD Radeon 890M is an integrated GPU with a 15 W TDP, no power connectors, and an IGP slot width. The NVIDIA GeForce RTX 5070 Ti SUPER is a dual-slot discrete card with a 350 W TDP, a single 16-pin power connector, and dimensions of 304 mm in length, 137 mm in height, and 48 mm in width. The AMD part uses a PCIe 4.0 x8 interface, while the NVIDIA part uses PCIe 5.0 x16. Display outputs also differ: the AMD part is portable device dependent, while the NVIDIA part offers 1x HDMI 2.1b and 3x DisplayPort 2.1b.

Both parts support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The AMD part was released in July 2024, while the NVIDIA part was released in December 2025. The AMD part's predecessor is the Navi II IGP, while the NVIDIA part has no recorded predecessor. The AMD part remains active in production, as does the NVIDIA part.

The Verdict

The data indicates that these two products target different segments entirely. The AMD Radeon 890M is an integrated GPU with a 15 W TDP, designed for portable devices where power draw and physical space are constrained. The NVIDIA GeForce RTX 5070 Ti SUPER is a discrete, dual-slot card with a 350 W TDP and a 16-pin power connector, designed for desktop systems with dedicated power delivery. The benchmark gap reflects this: the NVIDIA part scores 6269.5 in 3DMark Steel Nomad DX12, while the AMD part scores 590, a difference of 90.6%.

The AMD Radeon 890M should be selected for systems that need a GPU integrated into the processor, with no separate power connector and minimal board space. Its 4 nm process and 34,000 million transistors on a 233 mm² die make it a compact solution. The NVIDIA GeForce RTX 5070 Ti SUPER should be selected for systems that can accommodate a 304 mm long, 137 mm tall, 48 mm wide card with a 16-pin power connector and 350 W power draw. Its 16 GB of GDDR7 memory and 896.0 GB/s bandwidth provide dedicated high-speed memory for demanding workloads.

The database shows the NVIDIA part matching its nearest rival, the NVIDIA GeForce RTX 4070 Ti SUPER AD102, with a 0% delta in average score. The AMD part sits 0.1% below the AMD Radeon Vega 8 and 0.7% below the NVIDIA GeForce GTX 960. These rival comparisons place the AMD part in the range of older discrete GPUs, while the NVIDIA part aligns with a recent high-end card. The percentile rankings reflect this: the AMD part at the 45th percentile, the NVIDIA part at the 36th percentile, though the latter is based on a single benchmark score.

Users with a portable device that relies on shared system memory will find the AMD Radeon 890M sufficient for the recorded legacy DirectX and compute workloads. Users with a desktop system that can supply 350 W through a 16-pin connector will find the NVIDIA GeForce RTX 5070 Ti SUPER vastly ahead in the only directly comparable modern DirectX 12 test. The choice hinges on the physical and power constraints of the host system, not on raw performance parity, since no such parity exists in the data.

FAQ

Q: Which GPU has the higher 3DMark Steel Nomad DX12 score?

A: The NVIDIA GeForce RTX 5070 Ti SUPER scores 6269.5, while the AMD Radeon 890M scores 590. The NVIDIA part wins by 90.6%.

Q: How much memory does each GPU have?

A: The NVIDIA GeForce RTX 5070 Ti SUPER has 16 GB of GDDR7 memory on a 256 bit bus. The AMD Radeon 890M uses system shared memory, with no dedicated VRAM.

Q: What are the power requirements for each GPU?

A: The AMD Radeon 890M has a 15 W TDP and no power connectors. The NVIDIA GeForce RTX 5070 Ti SUPER has a 350 W TDP and uses a single 16-pin power connector.

Q: Which GPU has more shading units?

A: The NVIDIA GeForce RTX 5070 Ti SUPER has 8960 shading units. The AMD Radeon 890M has 1024 shading units.

Q: How do the average benchmark scores compare?

A: The AMD Radeon 890M has an average benchmark score of 9210, while the NVIDIA GeForce RTX 5070 Ti SUPER has an average benchmark score of 6270. The AMD part is 0.1% below the AMD Radeon Vega 8, while the NVIDIA part is exactly even with the NVIDIA GeForce RTX 4070 Ti SUPER AD102.

Q: What is the release date difference?

A: The AMD Radeon 890M was released on July 14, 2024. The NVIDIA GeForce RTX 5070 Ti SUPER was released on December 31, 2025.

Head-to-Head Benchmarks

The only directly comparable benchmark in the database is 3DMark Steel Nomad DX12. The NVIDIA GeForce RTX 5070 Ti SUPER records a score of 6269.5, while the AMD Radeon 890M records a score of 590. This results in a delta of 90.6% in favor of the NVIDIA part. The NVIDIA part wins the only head-to-head benchmark, giving it 1 win out of 1 recorded comparison.

Beyond this single test, the database contains additional AMD-only benchmarks. The AMD Radeon 890M scores 37254 in Geekbench OpenCL and 40808 in Geekbench Vulkan. In Passmark tests, it scores 979 in G2D, 8076 in G3D, and 4157 in GPU Compute. For legacy DirectX, it scores 97 in DirectX 9, 73 in DirectX 11, 37 in DirectX 12, and 33 in DirectX 10. These scores provide context for the AMD part's capabilities in older APIs and compute tasks, but without equivalent NVIDIA measurements, no cross-product comparison is possible.

The NVIDIA GeForce RTX 5070 Ti SUPER has no recorded Passmark or Geekbench results in the database. Its only benchmark is the Steel Nomad result, which limits the depth of analysis. The average benchmark score for the NVIDIA part is 6270, identical to its Steel Nomad score, while the AMD part's average of 9210 is higher than its Steel Nomad score due to the additional Passmark and Geekbench entries.

The delta percentages in the nearest rivals list provide further context. The AMD Radeon 890M is 0.1% below the AMD Radeon Vega 8, 0.7% below the NVIDIA GeForce GTX 960, 0.9% below the NVIDIA GeForce GTX 465, and 1% below the NVIDIA GeForce GTX 850M. The NVIDIA GeForce RTX 5070 Ti SUPER is 0% different from the NVIDIA GeForce RTX 4070 Ti SUPER AD102, 0.2% above the NVIDIA Quadro K620, 0.8% below the AMD FirePro W600, and 0.9% above the AMD Radeon R7 M350. These figures place the AMD part in the range of mid-2010s discrete GPUs, while the NVIDIA part matches a recent high-end card exactly.

The texture and pixel rates reinforce the gap. The NVIDIA part delivers 686.6 GTexel/s and 235.4 GPixel/s, while the AMD part delivers 185.6 GTexel/s and 92.80 GPixel/s. The FP32 throughput is 43.94 TFLOPS for the NVIDIA part versus 5.939 TFLOPS for the AMD part. These raw throughput numbers align with the Steel Nomad result: the NVIDIA part has roughly 7.4 times the FP32 throughput and 3.7 times the texture rate, consistent with its 90.6% lead in the modern DirectX 12 test.