AMD Steam Machine GPU vs NVIDIA RTX PRO 5000 72 GB Blackwell Comparison

Head-to-Head Benchmarks

The database records no direct head-to-head benchmark comparisons between the AMD Steam Machine GPU and the NVIDIA RTX PRO 5000 72 GB Blackwell. The AMD Steam Machine GPU has no benchmark entries, an average benchmark score of 0, and a percentile rank of 50 among all GPUs. The NVIDIA RTX PRO 5000 72 GB Blackwell, by contrast, has eight recorded benchmark scores and an average benchmark score of 6407, placing it at the 37th percentile.

The NVIDIA card's individual results show a wide spread across test types. In 3DMark Steel Nomad DX12, it scores 9579.5, which is its strongest recorded result. PassMark G3D delivers a score of 24310, while PassMark GPU Compute reaches 15667. The DirectX-specific PassMark tests produce lower figures: DirectX 9 scores 311, DirectX 10 scores 175, DirectX 11 scores 219, and DirectX 12 scores 78. The PassMark G2D test, which measures 2D graphics performance, returns 914.

The nearest rivals in the database for the NVIDIA card are all older or lower-tier products. The GeForce GTX 460 SE and GeForce GTX 580M each have an average score of 6389, which is a delta of 0.3 percent relative to the RTX PRO 5000's 6407 average. The AMD Radeon Vega 10 Mobile scores 6476, a negative delta of 1.1 percent, and the Quadro M5000M scores 6481, also a negative delta of 1.1 percent. These comparisons indicate that the RTX PRO 5000's average score sits within a narrow band around these legacy parts, despite its much higher peak scores in specific tests. The large gap between its 3DMark result and its PassMark DirectX 12 score suggests the card's performance is highly workload-dependent.

Because the AMD part has no recorded benchmarks, no quantitative win/loss ratio can be established from the data. The recorded winsA and winsB fields are both zero, reflecting the absence of head-to-head tests. The analysis must therefore rely on architectural specifications and the NVIDIA card's standalone benchmark profile.

Architecture Differences

The two GPUs come from different vendors, use different process nodes, and target entirely different market segments. The AMD Steam Machine GPU uses the Navi 33 chip built on RDNA 3.0 architecture, with the codename Hotpink Bonefish. It belongs to the Console GPU (Valve) generation. The process node is 6 nm at TSMC, with 13,300 million transistors on a 204 mm² die, yielding a transistor density of 65.2 million per square millimeter. The NVIDIA RTX PRO 5000 72 GB Blackwell uses the GB202 chip on Blackwell 2.0 architecture, part of the Blackwell PRO W (x000) generation. It is built on a 5 nm process at TSMC, with 92,200 million transistors on a 750 mm² die, for a transistor density of 122.9 million per square millimeter. The NVIDIA chip has nearly 7 times the transistor count and a die more than 3.6 times larger.

Memory configurations differ dramatically. The AMD card has 8 GB of GDDR6 on a 128-bit bus, with a memory clock of 2250 MHz (18 Gbps effective) and bandwidth of 288.0 GB/s. The NVIDIA card has 72 GB of GDDR7 on a 384-bit bus, with a memory clock of 1750 MHz (28 Gbps effective) and bandwidth of 1.34 TB/s. The NVIDIA card offers 9 times the capacity and 4.65 times the bandwidth.

Compute resources follow the same pattern. The AMD GPU has 1792 shading units, 112 texture mapping units, 64 render output units, and 28 ray tracing cores. It has no tensor cores. The NVIDIA GPU has 14,080 shading units, 440 TMUs, 160 ROPs, 110 ray tracing cores, and 440 tensor cores. The NVIDIA card has roughly 7.9 times the shading units, 3.9 times the TMUs, 2.5 times the ROPs, and 3.9 times the ray tracing cores. Pixel rate for the AMD part is 156.8 GPixel/s, texture rate is 274.4 GTexel/s, and FP32 throughput is 17.56 TFLOPS. FP16 matches at 17.56 TFLOPS with a 1:1 ratio. The NVIDIA card delivers 380.3 GPixel/s, 1,045.9 GTexel/s, and 66.94 TFLOPS in both FP32 and FP16, also at a 1:1 ratio.

Clock speeds are similar despite the size disparity. The AMD base clock is 1720 MHz with a boost of 2450 MHz and a game clock of 2250 MHz. The NVIDIA base clock is 1740 MHz with a boost of 2377 MHz. Power consumption differs substantially: the AMD card has a 110 W TDP, while the NVIDIA card has a 300 W TDP. The NVIDIA card is Dual-slot, uses a 1x 16-pin power connector, and suggests a 700 W PSU. The AMD card lists no slot width, no power connectors, and no suggested PSU, indicating it draws power through its host system.

Both cards support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Display outputs differ: the AMD card has 1x HDMI 2.1a and 1x DisplayPort 2.1, while the NVIDIA card has 4x DisplayPort 2.1b. The NVIDIA card uses PCIe 5.0 x16, while the AMD card has no recorded bus interface. Physical dimensions also differ: the AMD card measures 156 mm in length, 152 mm in height, and 162 mm in width. The NVIDIA card measures 267 mm in length, 111 mm in height, and 40 mm in width.

Release dates are separated by roughly 8 months. The NVIDIA card released on 2025-10-20, with the predecessor listed as Workstation Ada. The AMD card released on 2026-06-28, with no predecessor or successor recorded. Both are marked as Active in production status.

Where Each One Wins

The data supports distinct use-case advantages for each card, though the AMD side has no benchmark scores to confirm its strengths empirically.

The NVIDIA RTX PRO 5000 72 GB Blackwell wins on raw compute throughput. Its FP32 and FP16 performance of 66.94 TFLOPS is 3.8 times the AMD card's 17.56 TFLOPS. Texture rate is 3.8 times higher, pixel rate is 2.4 times higher, and memory bandwidth is 4.65 times higher. The 72 GB GDDR7 memory capacity, combined with the 384-bit bus, makes this card suitable for workloads with very large datasets, such as rendering scenes or compute tasks that exceed the 8 GB capacity of the AMD part. The presence of 440 tensor cores gives the NVIDIA card capabilities in tensor-based workloads that the AMD card cannot perform at all, since it has no tensor cores.

The AMD Steam Machine GPU wins on power efficiency and physical footprint. Its 110 W TDP is less than half the NVIDIA card's 300 W TDP. It requires no power connectors, which simplifies installation in compact systems. Its dimensions are roughly 156 mm by 152 mm by 162 mm, compared to the NVIDIA card's 267 mm by 111 mm by 40 mm. The AMD card is shorter in length but taller and wider, making it a more compact overall package. Its 6 nm process node and 13,300 million transistors on a 204 mm² die result in a transistor density of 65.2M per square millimeter, which is lower than the NVIDIA card's 122.9M per square millimeter, but the AMD chip uses far fewer transistors overall.

The NVIDIA card's benchmark profile shows where it excels and where it does not. Its 3DMark Steel Nomad DX12 score of 9579.5 indicates strong DX12 gaming performance in that specific test. Its PassMark G3D score of 24310 and GPU Compute score of 15667 show solid general 3D and compute performance. However, its PassMark DirectX 12 score of 78 is notably low, as are its DirectX 10 and DirectX 11 scores of 175 and 219. These results suggest the card's performance varies significantly depending on the API and test methodology. The DirectX 9 score of 311 is the highest among the PassMark DirectX tests, which may reflect driver behavior or test-specific characteristics rather than raw capability.

The AMD card has no recorded benchmarks, so its actual performance in any test cannot be quantified from the database. Its architecture suggests it is designed for a specific console context, given the "Console GPU (Valve)" generation label. Its 8 GB memory and 128-bit bus are modest by modern standards, and its 17.56 TFLOPS FP32 throughput places it in a lower performance tier than the NVIDIA card.

The Verdict

From the recorded data, the NVIDIA RTX PRO 5000 72 GB Blackwell is the clear choice for workloads requiring large memory capacity, high compute throughput, and tensor core functionality. Its 72 GB GDDR7 memory, 1.34 TB/s bandwidth, 66.94 TFLOPS FP32, and 440 tensor cores provide capabilities that the AMD Steam Machine GPU cannot match, as the AMD card has 8 GB GDDR6, 288.0 GB/s bandwidth, 17.56 TFLOPS, and no tensor cores. The NVIDIA card's standalone benchmark scores, including 9579.5 in 3DMark Steel Nomad DX12 and 24310 in PassMark G3D, demonstrate measurable performance in those tests.

The AMD Steam Machine GPU is the appropriate choice for systems where power consumption and physical size are the primary constraints. Its 110 W TDP, lack of power connectors, and compact dimensions make it suitable for small form factor or console-style enclosures. Its 6 nm process and 13,300 million transistors indicate a more modest design intended for constrained environments. The absence of benchmark data means no performance claims can be made for this card from the database, but its specifications define its role clearly.

Users who need maximum memory capacity for large datasets, or who require tensor core acceleration, should select the NVIDIA card. Users who need a low-power, compact GPU with no external power requirement should select the AMD card. The NVIDIA card's 300 W TDP and 700 W suggested PSU requirement make it unsuitable for systems with limited power delivery. The AMD card's 8 GB memory and 128-bit bus will limit its performance in memory-intensive tasks, but its lower power draw and smaller footprint offer advantages in constrained builds.

The percentile ranks provide additional context. The NVIDIA card sits at the 37th percentile among all GPUs, with an average benchmark score of 6407. The AMD card sits at the 50th percentile, but with an average benchmark score of 0, which reflects the absence of recorded benchmarks rather than measured performance. The nearest rivals for the NVIDIA card, all within 1.1 percent of its average score, are older or lower-tier parts, which suggests its average score is pulled down by weak results in certain PassMark tests despite strong showings in others.

FAQ

Q: Which GPU has more memory?

A: The NVIDIA RTX PRO 5000 72 GB Blackwell has 72 GB of GDDR7 memory, while the AMD Steam Machine GPU has 8 GB of GDDR6 memory.

Q: What is the FP32 throughput difference between the two cards?

A: The NVIDIA card delivers 66.94 TFLOPS FP32, while the AMD card delivers 17.56 TFLOPS FP32. The NVIDIA card is 3.8 times higher.

Q: Does the AMD Steam Machine GPU have tensor cores?

A: No. The AMD card lists no tensor cores, while the NVIDIA RTX PRO 5000 has 440 tensor cores.

Q: What is the power consumption of each card?

A: The AMD Steam Machine GPU has a 110 W TDP, and the NVIDIA RTX PRO 5000 has a 300 W TDP with a suggested PSU of 700 W.

Q: What are the recorded benchmark scores for the NVIDIA RTX PRO 5000?

A: The NVIDIA card has eight recorded scores: 9579.5 in 3DMark Steel Nomad DX12, 175 in PassMark DirectX 10, 219 in PassMark DirectX 11, 78 in PassMark DirectX 12, 311 in PassMark DirectX 9, 914 in PassMark G2D, 24310 in PassMark G3D, and 15667 in PassMark GPU Compute.

Q: Why does the AMD Steam Machine GPU have no benchmark scores?

A: The database records no benchmark entries for the AMD card, resulting in an average benchmark score of 0 and no nearest rivals.