AMD Ryzen Z2 Extreme GPU vs NVIDIA B200 SXM6 Comparison

Where Each One Wins

The recorded data places these two processors at opposite ends of the GPU spectrum. The AMD Ryzen Z2 Extreme GPU, a 28 W console-class part built on TSMC 4 nm, delivers a 3DMark Steel Nomad DX12 score of 516. That score places it in the 1st percentile of all GPUs in the database, meaning it outperforms roughly 1% of recorded parts. Its nearest rivals in the database are older integrated and mobile graphics solutions: the Intel HD Graphics P4000 (534 average score, 3.4% higher), the AMD Radeon HD 6870 (536 average score, 3.7% higher), and the AMD Radeon HD 6770M (569 average score, 9.3% higher). The only rival it beats in that immediate cluster is the AMD Radeon HD 6750M, which scores 484 and trails by 6.6%. This indicates a part designed for low-power, portable gaming scenarios where absolute frame rates are secondary to power efficiency and thermal restraint.

The NVIDIA B200 SXM6, by contrast, has no recorded benchmark scores in the database and no nearest rivals listed. Its 50th percentile placement across all GPUs is a default midpoint value, not a measured result, so direct numerical comparison in standard gaming workloads is not possible from the data. The B200 SXM6 is a server Blackwell part, built on TSMC 5 nm, with a 1000 W TDP and a suggested PSU rating of 1400 W. Its memory subsystem is 180 GB of HBM3e on an 8192-bit bus, delivering 8.19 TB/s of bandwidth. That configuration targets compute and AI workloads, not rasterized gaming. The absence of display outputs (the database lists "No outputs") reinforces that this is not a client-facing graphics product.

The wins are therefore split by workload domain. The Ryzen Z2 Extreme GPU wins in the only measured gaming benchmark available, and it wins on portability: 28 W TDP versus 1000 W, no power connectors required versus a 1400 W suggested PSU, and a single USB Type-C display output versus none. The B200 SXM6 wins on raw compute throughput, memory capacity, memory bandwidth, and shading resources, all of which are invisible in the gaming benchmark because no gaming benchmark was recorded for it.

The Verdict

The data indicates that these two parts should not be cross-shopped. The AMD Ryzen Z2 Extreme GPU is a console GPU with a 28 W TDP, 16 GB of LPDDR5X memory on a 128-bit bus, and a 128.0 GB/s bandwidth. It produces 5.530 TFLOPS of FP32 performance and 129.6 GPixel/s of pixel throughput. Its 3DMark Steel Nomad score of 516 sits just below the Intel HD Graphics P4000 and AMD Radeon HD 6870, and just above the AMD Radeon HD 6750M. For a user or system builder targeting a low-power handheld or embedded gaming device, this is the only one of the two with any recorded gaming data at all.

The NVIDIA B200 SXM6 is a server accelerator. Its 69.34 TFLOPS of FP32 performance is roughly 12.5 times the Ryzen part's figure, and its 8.19 TB/s memory bandwidth dwarfs the 128.0 GB/s of the AMD chip. It is built on a 1628 mm² die with 208,000 million transistors, versus 233 mm² and 34,000 million for the AMD part. The B200 SXM6 has no display outputs and no gaming benchmark scores, so anyone seeking a part to render frames to a screen should select the AMD product. Anyone seeking a compute accelerator for server workloads should select the NVIDIA product. The recorded data does not support any other conclusion.

Head-to-Head Benchmarks

The database contains no head-to-head benchmark entries for these two items. The only measured score belongs to the AMD Ryzen Z2 Extreme GPU: 516 in 3dmark_3dmark_steel_nomad_dx12. The NVIDIA B200 SXM6 has an empty benchmark array, so no direct comparison in the same test is possible.

What can be compared is the surrounding data. In FP32 compute, the B200 SXM6 records 69.34 TFLOPS against 5.530 TFLOPS for the Ryzen Z2 Extreme GPU, a 12.5x advantage. Texture rate favors the NVIDIA part at 1,083.4 GTexel/s versus 172.8 GTexel/s, a 6.3x advantage. Shading units number 18,944 for NVIDIA versus 1,024 for AMD, and TMUs number 592 versus 64. The B200 SXM6 also carries 592 tensor cores, while the Ryzen Z2 Extreme GPU lists none.

The AMD part wins on pixel rate: 129.6 GPixel/s versus 43.92 GPixel/s for the B200 SXM6. That is a 2.95x advantage in favor of the console GPU, which makes sense given the B200 SXM6's 24 ROPs versus 48 for the AMD part. The NVIDIA accelerator prioritizes texture and compute throughput over raster output.

Memory comparison shows the scale difference clearly. The B200 SXM6 has 180 GB of HBM3e on an 8192-bit bus with 8.19 TB/s bandwidth. The Ryzen Z2 Extreme GPU has 16 GB of LPDDR5X on a 128-bit bus with 128.0 GB/s bandwidth. The NVIDIA part has 11.25 times the memory capacity and 64 times the memory bandwidth. Clock behavior also differs: the AMD part boosts to 2700 MHz from an 800 MHz base, while the NVIDIA part boosts to 1830 MHz from a 120 MHz base, reflecting the different thermal and power envelopes.

FAQ

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

A: The AMD Ryzen Z2 Extreme GPU is the only one of the two with a recorded score, at 516. The NVIDIA B200 SXM6 has no recorded benchmark scores in the database.

Q: How does the AMD Ryzen Z2 Extreme GPU compare to its nearest rivals?

A: It scores 3.4% lower than the Intel HD Graphics P4000 (534), 3.7% lower than the AMD Radeon HD 6870 (536), and 9.3% lower than the AMD Radeon HD 6770M (569). It scores 6.6% higher than the AMD Radeon HD 6750M (484).

Q: What memory configurations do the two GPUs use?

A: The AMD Ryzen Z2 Extreme GPU uses 16 GB of LPDDR5X on a 128-bit bus with 128.0 GB/s bandwidth. The NVIDIA B200 SXM6 uses 180 GB of HBM3e on an 8192-bit bus with 8.19 TB/s bandwidth.

Q: Which GPU has more shading units and texture mapping units?

A: The NVIDIA B200 SXM6 has 18,944 shading units and 592 TMUs. The AMD Ryzen Z2 Extreme GPU has 1,024 shading units and 64 TMUs.

Q: Does the NVIDIA B200 SXM6 support DirectX, OpenGL, or Vulkan?

A: The database lists all three APIs as N/A for the B200 SXM6. The AMD Ryzen Z2 Extreme GPU supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: What is the power requirement difference between the two?

A: The AMD Ryzen Z2 Extreme GPU has a 28 W TDP and lists no power connectors. The NVIDIA B200 SXM6 has a 1000 W TDP and a suggested PSU rating of 1400 W.

Architecture Differences

The AMD Ryzen Z2 Extreme GPU uses the Strix Point chip with RDNA 3.5 architecture, fabricated on a 4 nm TSMC node. The die measures 233 mm² and contains 34,000 million transistors, giving a transistor density of 145.9 million per mm². It belongs to the Console GPU (AMD) generation and supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Its graphics pipeline includes 1,024 shading units, 64 TMUs, 48 ROPs, and 16 ray tracing cores. Tensor cores are not listed.

The NVIDIA B200 SXM6 uses the GB100 chip with Blackwell architecture, fabricated on a 5 nm TSMC node. The die measures 1,628 mm² and contains 208,000 million transistors, giving a transistor density of 127.8 million per mm². It belongs to the Server Blackwell (Bxx) generation and lists no graphics API support. Its compute pipeline includes 18,944 shading units, 592 TMUs, 24 ROPs, and 592 tensor cores. Ray tracing cores are not listed. The predecessor is listed as Server Hopper and the successor as Server Rubin.

The node difference matters for transistor density: the AMD part packs more transistors per square millimeter (145.9M versus 127.8M) despite being on a nominally larger process node number. The NVIDIA part uses a much larger die to house over six times the transistor count. The AMD part has three times the ROP count (48 versus 24), which explains its higher pixel rate. The NVIDIA part has 592 tensor cores, absent from the AMD specification, marking it as a compute-oriented design.

Specification Differences

The two parts differ in nearly every recorded specification field. Process node: 4 nm for AMD, 5 nm for NVIDIA. Die size: 233 mm² versus 1,628 mm². Transistors: 34,000 million versus 208,000 million. Base clock: 800 MHz versus 120 MHz. Boost clock: 2700 MHz versus 1830 MHz. Memory clock: 1000 MHz (8 Gbps effective) versus 2000 MHz (8 Gbps effective). Memory size: 16 GB versus 180 GB. Memory type: LPDDR5X versus HBM3e. Bus width: 128 bit versus 8192 bit. Bandwidth: 128.0 GB/s versus 8.19 TB/s.

Shading units: 1,024 versus 18,944. TMUs: 64 versus 592. ROPs: 48 versus 24. Ray tracing cores: 16 versus not listed. Tensor cores: not listed versus 592. Pixel rate: 129.6 GPixel/s versus 43.92 GPixel/s. Texture rate: 172.8 GTexel/s versus 1,083.4 GTexel/s. FP32: 5.530 TFLOPS versus 69.34 TFLOPS. FP16: 5.530 TFLOPS (1:1) versus 69.34 TFLOPS (1:1). TDP: 28 W versus 1000 W. Slot width: not listed versus SXM Module. Power connectors: none versus not listed. Suggested PSU: not listed versus 1400 W. Bus interface: not listed versus PCIe 6.0 x16. Display outputs: 1x USB Type-C versus no outputs. API support: DirectX 12 Ultimate, OpenGL 4.6, Vulkan 1.4 versus N/A for all three. Release date: 2025-07-08 versus 2024-10-31. The NVIDIA part has a launch MSRP of 34,999 USD. The AMD part has no launch MSRP recorded. Production status for both is Active.