AMD Radeon 890M vs NVIDIA RTX PRO 6000D Blackwell Max-Q Comparison

Head-to-Head Benchmarks

The recorded data contains a single direct comparison between these two GPUs: the 3DMark Steel Nomad DX12 test. The NVIDIA RTX PRO 6000D Blackwell Max-Q scores 11088, while the AMD Radeon 890M scores 590. This represents a delta of 1779.3% in favor of the NVIDIA part. In practical terms, the NVIDIA GPU delivers a performance margin that is not merely incremental; it is an order of magnitude higher in this specific workload.

The NVIDIA part's score of 11088 places it at the 50th percentile among all GPUs in the database. Its nearest rivals are tightly clustered: the NVIDIA RTX PRO 6000 Blackwell Max-Q scores identically at 11088 (0% delta), the AMD Radeon RX 550 scores 11075 (0.1% slower), the NVIDIA GeForce GTX 1650 SUPER scores 11047 (0.4% slower), and the AMD FirePro W4300 scores 11225 (1.2% faster). This clustering suggests that the RTX PRO 6000D is positioned in a highly competitive performance tier, where single-digit percentage differences separate it from its immediate peers.

The AMD Radeon 890M, by contrast, scores 590 in the same Steel Nomad DX12 test, which is a fraction of the NVIDIA result. However, the 890M's broader benchmark suite tells a different story. Its average benchmark score across all recorded tests is 9210, which is heavily influenced by its strong showings in other workloads. The 890M's percentile rank is 45, only five points behind the NVIDIA GPU, despite the massive gap in the Steel Nomad test. This indicates that the 890M is a versatile part that excels in certain categories, even if it falls far behind in this particular DX12 workload.

Looking at the 890M's nearest rivals based on average score: the AMD Radeon Vega 8 scores 9221 (0.1% faster), the NVIDIA GeForce GTX 960 scores 9273 (0.7% faster), the NVIDIA GeForce GTX 465 scores 9294 (0.9% faster), and the NVIDIA GeForce GTX 850M scores 9302 (1% faster). These are all within a 1% band, meaning the 890M's average performance is comparable to a range of older discrete GPUs, despite being an integrated graphics processor.

The database records one win for the NVIDIA part and zero wins for the AMD part in head-to-head comparisons. That single win, however, is decisive in magnitude. A 1779.3% delta is not a marginal victory; it is a categorical one in that specific benchmark. The data does not include any other head-to-head tests, so the full picture of relative performance across different workloads remains incomplete. What is clear is that in the Steel Nomad DX12 scenario, the NVIDIA GPU is overwhelmingly faster.

The Verdict

The data presents a clear hierarchy for anyone choosing between these two products based on pure performance metrics. The NVIDIA RTX PRO 6000D Blackwell Max-Q is the dominant part in the recorded 3DMark Steel Nomad DX12 benchmark, outscoring the AMD Radeon 890M by a factor of nearly 19. If the primary use case is demanding DX12 gaming or professional 3D workloads that resemble this benchmark, the NVIDIA GPU is the obvious choice, and the data does not support any other conclusion.

However, the AMD Radeon 890M is not without merit. Its average benchmark score of 9210, derived from a broader set of tests including Geekbench OpenCL, Geekbench Vulkan, and multiple Passmark categories, shows that it holds its own against a specific set of rivals. The 890M's percentile rank of 45 versus the NVIDIA's 50 indicates that, across the entire database, the two parts are not as far apart as the Steel Nomad result might suggest. This is because the 890M's other scores, such as 40808 in Geekbench Vulkan and 37254 in Geekbench OpenCL, contribute significantly to its average.

For the user who needs a discrete workstation GPU with massive memory capacity and raw compute throughput, the NVIDIA part is the only rational pick from this data. For the user who is constrained to an integrated GPU in a mobile or compact system, the AMD 890M offers a level of performance that the database shows is competitive with several older discrete GPUs. The data does not support the 890M as a replacement for the NVIDIA part in any high-end scenario, but it does validate it as a capable integrated solution.

The verdict, strictly from the recorded numbers: the NVIDIA RTX PRO 6000D wins the only head-to-head test by an enormous margin, and any buyer seeking that level of DX12 performance must choose it. The AMD Radeon 890M is a different class of product, and its data shows it is best suited for systems where a discrete GPU is not an option.

Where Each One Wins

The NVIDIA RTX PRO 6000D Blackwell Max-Q wins in the category that matters most for high-end graphics: raw DX12 rendering performance. Its 11088 score in 3DMark Steel Nomad DX12 is the only head-to-head metric recorded, and it wins that contest outright. The GPU's specifications in the database support this dominance: 24064 shading units, 752 texture mapping units, 192 ROPs, 188 RT cores, and 752 tensor cores. The FP32 compute is listed at 110.1 TFLOPS, which is a massive figure compared to the AMD part's 5.939 TFLOPS. The memory subsystem is also vastly different: the NVIDIA GPU has 96 GB of GDDR7 on a 512-bit bus with 1.79 TB/s of bandwidth, whereas the AMD part uses system-shared memory with system-dependent bandwidth.

The AMD Radeon 890M wins in the context of efficiency and integration. Its TDP is 15 W, compared to the NVIDIA part's 300 W. The 890M is an IGP with no power connectors and no slot width, meaning it fits into portable devices and compact systems where a dual-slot, 267 mm long, 111 mm high, 40 mm wide card with a 16-pin power connector would be impossible. The 890M's process node is 4 nm, which is smaller than the NVIDIA part's 5 nm node, and its transistor density is higher at 145.9M per mm² versus 122.9M per mm². This confirms AMD's design efficiency, even if the absolute performance is far lower.

The 890M also shows strength in specific non-DX12 benchmarks within its own record. Its Geekbench Vulkan score of 40808 is notably higher than its OpenCL score of 37254, suggesting that Vulkan workloads are a relative strength. The Passmark scores are varied: 979 in G2D, 8076 in G3D, 4157 in GPU compute, and lower scores in the DX9, DX10, DX11, and DX12 Passmark tests (97, 33, 73, and 37 respectively). These numbers indicate that the 890M is more capable in compute and general 3D tasks than in legacy or modern DirectX rasterization, which aligns with its low Steel Nomad result.

For the NVIDIA part, the data shows only the single Steel Nomad benchmark, so its performance in other APIs is not recorded. The database does list its API support as DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, which matches the AMD part's API list. Both support the same API versions, but the actual performance in those APIs beyond Steel Nomad is not measured for the NVIDIA part.

FAQ

Q: How much faster is the NVIDIA RTX PRO 6000D Blackwell Max-Q than the AMD Radeon 890M in the recorded head-to-head test?

A: In the 3DMark Steel Nomad DX12 benchmark, the NVIDIA GPU scores 11088 while the AMD GPU scores 590. This is a delta of 1779.3% in favor of the NVIDIA part.

Q: What is the average benchmark score for each GPU?

A: The NVIDIA RTX PRO 6000D has an average benchmark score of 11088, based on its single recorded Steel Nomad result. The AMD Radeon 890M has an average score of 9210, based on ten recorded benchmarks including Geekbench and Passmark tests.

Q: Which GPU has a higher percentile rank among all GPUs in the database?

A: The NVIDIA RTX PRO 6000D sits at the 50th percentile, while the AMD Radeon 890M sits at the 45th percentile. The five-percentile gap is modest, despite the large difference in the Steel Nomad test.

Q: What are the memory configurations of these two GPUs?

A: The NVIDIA GPU has 96 GB of GDDR7 memory on a 512-bit bus, with a bandwidth of 1.79 TB/s. The AMD GPU uses system-shared memory, with a system-shared type and bus width, and its bandwidth is system dependent.

Q: What is the TDP difference between the two parts?

A: The NVIDIA RTX PRO 6000D has a TDP of 300 W and requires a 16-pin power connector, with a suggested PSU of 700 W. The AMD Radeon 890M has a TDP of 15 W and uses no power connectors, as it is an integrated graphics processor.

Q: Do both GPUs support the same DirectX and Vulkan versions?

A: Yes, both are listed with DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4 support.

Architecture Differences

The NVIDIA RTX PRO 6000D Blackwell Max-Q is built on the GB202 chip, using the Blackwell 2.0 architecture, and is part of the Blackwell PRO W (x000) generation. It is fabricated on a 5 nm process at TSMC, with a transistor count of 92,200 million and a die size of 750 mm². This yields a transistor density of 122.9 million per mm². The GPU operates with a base clock of 1590 MHz and a boost clock of 2288 MHz, with memory running at 1750 MHz, which is listed as 28 Gbps effective. The memory subsystem is substantial: 96 GB of GDDR7 on a 512-bit bus, providing 1.79 TB/s of bandwidth.

The compute configuration of the NVIDIA part is massive. It has 24064 shading units, 752 TMUs, 192 ROPs, 188 RT cores, and 752 tensor cores. The pixel rate is 439.3 GPixel/s, the texture rate is 1,720.6 GTexel/s, and FP32 compute is 110.1 TFLOPS, with FP16 also at 110.1 TFLOPS (1:1). The card is dual-slot, 267 mm long, 111 mm high, and 40 mm wide. It uses a PCIe 5.0 x16 bus interface and has four DisplayPort 2.1b outputs. It was released on 2025-03-17 and has a launch MSRP of 8,565 USD.

The AMD Radeon 890M is built on the Strix Point chip, using the RDNA 3.5 architecture, and belongs to the Navi III IGP (Strix Point Mobile) generation. It is fabricated on a 4 nm process at TSMC, with 34,000 million transistors and a die size of 233 mm², resulting in a higher transistor density of 145.9 million per mm². The base clock is 400 MHz and the boost clock is 2900 MHz, with system-shared memory that has no fixed bandwidth. The compute configuration is far smaller: 1024 shading units, 64 TMUs, 32 ROPs, and 16 RT cores. It has no tensor cores listed. The pixel rate is 92.80 GPixel/s, the texture rate is 185.6 GTexel/s, and FP32 compute is 5.939 TFLOPS, with FP16 at the same figure (1:1).

The AMD part is an IGP with no slot width, no power connectors, and no listed dimensions. It uses a PCIe 4.0 x8 bus interface. Its display outputs are listed as portable device dependent. It was released on 2024-07-14 and has no launch MSRP recorded. The predecessor for the NVIDIA part is Workstation Ada, while the AMD part's predecessor is Navi II IGP.

The architectural divide is stark. The NVIDIA GPU is a full-size workstation card with a 300 W TDP, 96 GB of dedicated GDDR7, and a PCIe 5.0 x16 interface. The AMD GPU is a 15 W integrated processor with shared system memory and a PCIe 4.0 x8 interface. The process node advantage goes to AMD at 4 nm, and the transistor density advantage also goes to AMD, but the sheer scale of the NVIDIA chip, with nearly three times the transistors, dominates in raw performance. The data shows that the NVIDIA part is designed for maximum compute and rendering throughput, while the AMD part is designed for efficiency and integration into portable systems.