AMD Radeon RX 9070 XT vs NVIDIA RTX PRO 6000D Blackwell Max-Q Comparison

The Verdict

The data in this comparison is unusually one-sided, but not in the way the raw benchmark score might suggest. The AMD Radeon RX 9070 XT and the NVIDIA RTX PRO 6000D Blackwell Max-Q are aimed at completely different buyers, and the recorded measurements reflect that split.

For gaming and general rasterization workloads, the AMD Radeon RX 9070 XT is the practical pick. It posts a 3DMark Steel Nomad DX12 score of 7260, which is 34.5% behind the NVIDIA card, but it does so with a 16 GB GDDR6 memory configuration and a 644.6 GB/s bandwidth. That is a capable setup for high-refresh gaming at typical resolutions, and the card's active production status and recent release date (2025-03-05) indicate it is a current-generation consumer product.

For professional visualization, AI-adjacent compute, and massive dataset handling, the NVIDIA RTX PRO 6000D Blackwell Max-Q is the obvious choice. Its 96 GB GDDR7 memory, 1.79 TB/s bandwidth, and 110.1 TFLOPS FP32 throughput are in a different class entirely. The 3DMark Steel Nomad score of 11088 confirms it is the faster card in this head-to-head, but that is almost secondary to the memory capacity advantage, which is what actually matters for workstation tasks.

The verdict is simple: if you need gaming performance per watt and a modern consumer feature set, the RX 9070 XT wins. If you need to load enormous models or render complex scenes that exceed 16 GB, the RTX PRO 6000D is the only choice between these two. The RTX PRO 6000D's 50th percentile ranking among all GPUs, versus the RX 9070 XT's 55th percentile, suggests the AMD card is better positioned for the broader consumer market, but that is a reflection of the NVIDIA card's niche workstation role, not a knock on its capability.

Architecture Differences

The architectural gap between these two GPUs is substantial, and it explains most of the performance divergence. The AMD Radeon RX 9070 XT uses the Navi 48 chip built on RDNA 4.0 architecture, manufactured on TSMC's 4 nm process. It packs 53,900 million transistors into a 357 mm² die, giving it a transistor density of 151.0M per mm². The NVIDIA RTX PRO 6000D Blackwell Max-Q uses the GB202 chip on Blackwell 2.0 architecture, manufactured on TSMC's 5 nm process. That chip contains 92,200 million transistors on a much larger 750 mm² die, with a lower density of 122.9M per mm².

The core configurations could not be more different. The RX 9070 XT has 4096 shading units, 256 texture mapping units, 128 ROPs, and 64 ray tracing cores. The RTX PRO 6000D has 24064 shading units, 752 TMUs, 192 ROPs, 188 ray tracing cores, and 752 tensor cores. The NVIDIA card has nearly six times the shading units and nearly three times the ray tracing cores. The RX 9070 XT has no tensor cores listed, which means it lacks the dedicated AI acceleration hardware that the RTX PRO 6000D brings to the table.

Clock speeds tell a different story. The RX 9070 XT runs at a 1660 MHz base clock and boosts to 2970 MHz, with a game clock of 2400 MHz. The RTX PRO 6000D has a 1590 MHz base and 2288 MHz boost, which is lower but expected given the massive core count and 300 W TDP. The AMD card's higher clocks help it stay competitive in latency-sensitive workloads, but they cannot overcome the NVIDIA card's raw compute throughput.

Memory architecture is where the NVIDIA card pulls decisively ahead. The RX 9070 XT uses 16 GB of GDDR6 on a 256 bit bus, delivering 644.6 GB/s. The RTX PRO 6000D uses 96 GB of GDDR7 on a 512 bit bus, delivering 1.79 TB/s. That is six times the capacity and nearly three times the bandwidth. For any workload that is memory-bound, the NVIDIA card wins by a landslide.

Head-to-Head Benchmarks

The only direct benchmark shared between these two cards in the database is 3DMark Steel Nomad DX12. The results are clear: the NVIDIA RTX PRO 6000D Blackwell Max-Q scores 11088, while the AMD Radeon RX 9070 XT scores 7260. That is a delta of 34.5% in favor of NVIDIA.

This is a significant margin, but it needs context. The RX 9070 XT's 7260 score places it in the 55th percentile of all GPUs, with an average benchmark score of 13543 across all its recorded tests. The RTX PRO 6000D's 11088 score places it in the 50th percentile, with an average of 11088 since that is its only recorded benchmark. The percentile difference is interesting: the AMD card is ranked slightly higher overall despite losing the head-to-head, which suggests its broader benchmark suite shows strength in other areas like OpenCL, Vulkan, and PassMark tests.

Looking at the RX 9070 XT's other benchmarks, it scores 17428 in Geekbench OpenCL, 65879 in Geekbench Vulkan, and 26795 in PassMark G3D. Its PassMark GPU Compute score is 15864. These numbers indicate the AMD card is well-rounded across different API workloads. The RTX PRO 6000D has no such data, so the database only reflects its DX12 performance.

The nearest rivals for the RX 9070 XT are all much older, lower-end cards: the AMD Radeon HD 7770M (0.1% delta), NVIDIA GeForce GTX 570 (0.2% delta), NVIDIA P106-090 (0.5% delta), and AMD Radeon Pro 555 (1% delta). This clustering is odd for a modern 4 nm card, but it reflects the fact that the RX 9070 XT's average score of 13543 is dragged down by its low PassMark DX9, DX10, and DX12 scores (353, 154, and 78 respectively). For the RTX PRO 6000D, its nearest rival is itself (the non-Max-Q version, 0% delta), followed by the AMD Radeon RX 550 (0.1% delta) and NVIDIA GeForce GTX 1650 SUPER (0.4% delta). The AMD FirePro W4300 is 1.2% ahead.

Specification Differences

The table below lists only the fields where the two cards differ, based on the recorded data:

| Specification | AMD Radeon RX 9070 XT | NVIDIA RTX PRO 6000D Blackwell Max-Q |

|---|---|---|

| Chip | Navi 48 | GB202 |

| Architecture | RDNA 4.0 | Blackwell 2.0 |

| Process Node | 4 nm | 5 nm |

| Transistors | 53,900 million | 92,200 million |

| Die Size | 357 mm² | 750 mm² |

| Transistor Density | 151.0M / mm² | 122.9M / mm² |

| Base Clock | 1660 MHz | 1590 MHz |

| Boost Clock | 2970 MHz | 2288 MHz |

| Game Clock | 2400 MHz | Not specified |

| Memory Clock | 2518 MHz (20.1 Gbps effective) | 1750 MHz (28 Gbps effective) |

| Memory Size | 16 GB | 96 GB |

| Memory Type | GDDR6 | GDDR7 |

| Memory Bus Width | 256 bit | 512 bit |

| Memory Bandwidth | 644.6 GB/s | 1.79 TB/s |

| Shading Units | 4096 | 24064 |

| TMUs | 256 | 752 |

| ROPs | 128 | 192 |

| RT Cores | 64 | 188 |

| Tensor Cores | Not specified | 752 |

| Pixel Rate | 380.2 GPixel/s | 439.3 GPixel/s |

| Texture Rate | 760.3 GTexel/s | 1,720.6 GTexel/s |

| FP32 Performance | 48.66 TFLOPS | 110.1 TFLOPS |

| FP16 Performance | 48.66 TFLOPS (1:1) | 110.1 TFLOPS (1:1) |

| TDP | 304 W | 300 W |

| Power Connectors | 2x 8-pin | 1x 16-pin |

| Display Outputs | 1x HDMI 2.1b, 3x DisplayPort 2.1a | 4x DisplayPort 2.1b |

| Dimensions | Not specified | 267 mm x 111 mm x 40 mm |

| Release Date | 2025-03-05 | 2025-03-17 |

| Predecessor | Navi III | Workstation Ada |

Both cards share the same PCIe 5.0 x16 interface, 700 W suggested PSU, dual-slot design, DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4 support. The TDP difference is negligible (304 W vs 300 W), which is notable given the NVIDIA card's far higher core count and memory capacity.

FAQ

Q: Which card has more memory bandwidth?

A: The NVIDIA RTX PRO 6000D Blackwell Max-Q with 1.79 TB/s versus the AMD Radeon RX 9070 XT's 644.6 GB/s. The NVIDIA card also has a wider 512 bit bus compared to 256 bit.

Q: What is the FP32 performance difference?

A: The RTX PRO 6000D delivers 110.1 TFLOPS of FP32 compute, while the RX 9070 XT delivers 48.66 TFLOPS. That is more than double the single-precision throughput for the NVIDIA card.

Q: Which card has a higher boost clock?

A: The AMD Radeon RX 9070 XT boosts to 2970 MHz, while the NVIDIA RTX PRO 6000D boosts to 2288 MHz. The AMD card also has a higher base clock at 1660 MHz versus 1590 MHz.

Q: How do they compare in 3DMark Steel Nomad DX12?

A: The NVIDIA card scores 11088, which is 34.5% higher than the AMD card's 7260. This is the only shared benchmark in the database.

Q: Which card has more ray tracing cores?

A: The NVIDIA RTX PRO 6000D has 188 ray tracing cores, compared to 64 on the AMD Radeon RX 9070 XT. The NVIDIA card also has 752 tensor cores, which the AMD card lacks entirely.

Q: What are the memory capacities?

A: The NVIDIA card has 96 GB of GDDR7, while the AMD card has 16 GB of GDDR6. That is a six-fold difference in capacity, which matters for large datasets and multi-model workloads.

Where Each One Wins

The AMD Radeon RX 9070 XT wins in scenarios that favor high clock speeds and a balanced consumer feature set. Its 2970 MHz boost clock and 2400 MHz game clock are substantially higher than the NVIDIA card's 2288 MHz boost, which helps in latency-sensitive gaming workloads. The 16 GB GDDR6 memory is sufficient for current-generation games at high settings, and the 644.6 GB/s bandwidth is adequate for 1440p and 4K rasterization. The RX 9070 XT also has a higher average benchmark score (13543) across its full test suite, which includes strong Geekbench Vulkan (65879) and OpenCL (17428) results. Its 55th percentile ranking among all GPUs suggests it sits well within the consumer sweet spot. The dual 8-pin power connectors and 304 W TDP make it easier to slot into a standard gaming build with a 700 W PSU.

The NVIDIA RTX PRO 6000D Blackwell Max-Q wins in compute-heavy, memory-hungry professional environments. Its 96 GB GDDR7 memory is the standout feature, enabling workloads that simply cannot run on 16 GB. The 1.79 TB/s bandwidth and 512 bit bus make it ideal for large-scale data processing, high-resolution texture streaming, and multi-GPU rendering pipelines. The 110.1 TFLOPS FP32 and FP16 performance, combined with 752 tensor cores and 188 ray tracing cores, position it as a workstation tool for AI inference, scientific simulation, and photorealistic ray tracing. The 3DMark Steel Nomad score of 11088 confirms it is the faster card in raw DX12 performance, and its 439.3 GPixel/s pixel rate and 1,720.6 GTexel/s texture rate outpace the AMD card's 380.2 GPixel/s and 760.3 GTexel/s. The single 16-pin connector and 300 W TDP are a practical choice for a professional workstation, and the specified dimensions (267 mm x 111 mm x 40 mm) indicate a manageable dual-slot footprint.

In short, the RX 9070 XT is the better gaming and consumer card, while the RTX PRO 6000D is the better professional compute and memory-capacity card. The 34.5% benchmark gap in favor of NVIDIA is real, but it comes with a massive price difference in the recorded launch MSRP (599 USD for AMD versus 8,565 USD for NVIDIA). The choice hinges entirely on whether you need the NVIDIA card's professional capabilities or the AMD card's consumer accessibility.