Intel Arc Pro A60M vs NVIDIA B200 SXM6 Comparison

# Head-to-Head Benchmarks

The recorded data shows no direct head-to-head benchmark results between the Intel Arc Pro A60M and the NVIDIA B200 SXM6. Both entries hold a 50th percentile position among all GPUs in the database, with an average benchmark score of zero for each. This means the two cards do not share any measured performance tests, and their relative standing in the database is based solely on their specifications and architectural characteristics rather than empirical racing.

Where the numbers do exist, the gap is dramatic. The B200 SXM6 delivers 69.34 TFLOPS of FP32 compute, which is 13 times the 5.325 TFLOPS of the Arc Pro A60M. In FP16, the B200 maintains 69.34 TFLOPS at a 1:1 ratio, while the Arc Pro A60M reaches 10.65 TFLOPS through a 2:1 rate. The texture rate on the B200 sits at 1,083.4 GTexel/s versus 166.4 GTexel/s on the Intel part, a 6.5 times advantage. Memory bandwidth tells a similar story: 8.19 TB/s on the B200 compared to 256.0 GB/s on the A60M, a 32 times difference.

One area where the Intel card shows a per-clock advantage is pixel throughput. The Arc Pro A60M manages 83.20 GPixel/s from its 64 ROPs, while the B200 SXM6 produces 43.92 GPixel/s from only 24 ROPs. This reflects the different design goals: the Intel part is a mobile graphics solution with conventional rasterization focus, whereas the NVIDIA part is a compute-oriented server accelerator with minimal ROP allocation.

Because the head-to-head benchmark set is empty, the database assigns zero wins to each side. Any practical comparison must therefore draw from the specification sheet and architectural analysis, which is what the following sections do.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: The NVIDIA B200 SXM6 records 69.34 TFLOPS of FP32, compared to 5.325 TFLOPS on the Intel Arc Pro A60M. The B200 outperforms the A60M by roughly 13 times in single-precision floating-point work.

Q: What is the memory capacity difference?

A: The B200 SXM6 carries 180 GB of HBM3e memory across an 8192-bit bus, while the Arc Pro A60M has 8 GB of GDDR6 on a 128-bit bus. The B200 also leads in bandwidth at 8.19 TB/s versus 256.0 GB/s.

Q: Do both GPUs support DirectX 12 Ultimate?

A: No. The Intel Arc Pro A60M supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The NVIDIA B200 SXM6 lists N/A for DirectX, OpenGL, and Vulkan, indicating it is not designed for conventional graphics API workloads.

Q: Which card has more shading units?

A: The B200 SXM6 has 18,944 shading units, whereas the Arc Pro A60M has 2,048. The B200 also carries 592 tensor cores and 592 texture mapping units, while the A60M has no tensor cores and 128 TMUs.

Q: What is the power draw difference?

A: The B200 SXM6 has a 1000 W TDP and a suggested PSU of 1400 W. The Arc Pro A60M has a 95 W TDP. The B200 consumes over 10 times the power of the Intel mobile chip.

Q: Which GPU is newer?

A: The NVIDIA B200 SXM6 was released on 2024-10-31, while the Intel Arc Pro A60M launched on 2023-06-05. The B200 is the newer product by about 17 months.

Architecture Differences

The two GPUs come from entirely different architectural lineages. The Intel Arc Pro A60M uses the Xe-HPG architecture on the DG2-256 chip, part of the Alchemist generation for Pro-Series Mobile. It is built on a 6 nm process at TSMC, with 11,500 million transistors on a 269 mm² die. The transistor density works out to 42.8 million per square millimeter.

The NVIDIA B200 SXM6 uses the Blackwell architecture on the GB100 chip, belonging to the Server Blackwell (Bxx) generation. It is fabricated on a 5 nm process, also at TSMC, with 208,000 million transistors on a 1628 mm² die. That gives a density of 127.8 million transistors per square millimeter, which is nearly three times the Intel figure.

The B200 features 592 tensor cores, while the A60M has none listed. The Intel part instead includes 16 ray tracing cores, whereas the B200 does not list any RT cores. This shows a clear split: Intel targets graphics workloads with ray tracing support and conventional API compatibility, while NVIDIA focuses on tensor-heavy compute tasks without any display outputs or graphics API support.

The B200 has 24 ROPs, which is unusually low for its transistor count, confirming that rasterization is not its purpose. The A60M has 64 ROPs, a more balanced allocation for a graphics-oriented mobile GPU. The B200 also uses HBM3e memory, a stacked high-bandwidth design, while the A60M uses GDDR6, a more traditional memory type for mobile graphics.

The B200's predecessor is listed as Server Hopper, and its successor as Server Rubin, placing it in a clear server product family. The A60M has no predecessor or successor listed in the database.

Specification Differences

The following fields differ between the two parts:

• Process node: 6 nm (Intel) versus 5 nm (NVIDIA)
• Transistors: 11,500 million versus 208,000 million
• Die size: 269 mm² versus 1628 mm²
• Transistor density: 42.8M / mm² versus 127.8M / mm²
• Base clock: 900 MHz versus 120 MHz
• Boost clock: 1300 MHz versus 1830 MHz
• Memory clock: 2000 MHz, 16 Gbps effective versus 2000 MHz, 8 Gbps effective
• Memory size: 8 GB versus 180 GB
• Memory type: GDDR6 versus HBM3e
• Memory bus width: 128 bit versus 8192 bit
• Memory bandwidth: 256.0 GB/s versus 8.19 TB/s
• Shading units: 2048 versus 18,944
• TMUs: 128 versus 592
• ROPs: 64 versus 24
• RT cores: 16 versus none listed
• Tensor cores: none listed versus 592
• Pixel rate: 83.20 GPixel/s versus 43.92 GPixel/s
• Texture rate: 166.4 GTexel/s versus 1,083.4 GTexel/s
• FP32 compute: 5.325 TFLOPS versus 69.34 TFLOPS
• FP16 compute: 10.65 TFLOPS (2:1) versus 69.34 TFLOPS (1:1)
• TDP: 95 W versus 1000 W
• Slot width: IGP versus SXM Module
• Suggested PSU: not listed versus 1400 W
• Bus interface: PCIe 4.0 x16 versus PCIe 6.0 x16
• Display outputs: Portable Device Dependent versus No outputs
• DirectX support: 12 Ultimate (12_2) versus N/A
• OpenGL support: 4.6 versus N/A
• Vulkan support: 1.4 versus N/A
• Release date: 2023-06-05 versus 2024-10-31
• Predecessor: none versus Server Hopper
• Successor: none versus Server Rubin
• Launch MSRP: none listed versus 34,999 USD

Both share the same foundry (TSMC) and the same memory clock base frequency of 2000 MHz, though the effective data rates diverge.

Where Each One Wins

The Intel Arc Pro A60M wins in any scenario that requires conventional graphics output. It has display outputs (Portable Device Dependent), supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, and includes 16 ray tracing cores. Its 64 ROPs deliver a higher pixel rate than the B200, 83.20 GPixel/s versus 43.92 GPixel/s. For mobile workstations running CAD, media playback, or light content creation with standard graphics APIs, the A60M is the only one of the two that can function as a display adapter. Its 95 W TDP also makes it suitable for laptops or compact systems where power constraints are severe. The 8 GB of GDDR6 memory is enough for typical mobile graphics workloads.

The NVIDIA B200 SXM6 wins in raw compute throughput. Its 69.34 TFLOPS FP32 and 69.34 TFLOPS FP16 (1:1) performance dwarfs the Intel part. The 592 tensor cores are designed for matrix operations common in AI training and inference. The 180 GB of HBM3e memory with 8.19 TB/s bandwidth supports massive model weights and large datasets that would not fit in the A60M's 8 GB. The B200 also holds a density advantage, packing 127.8M transistors per mm² versus 42.8M on the Intel chip, which reflects a more advanced process and denser compute logic. The PCIe 6.0 x16 interface, while not benchmarked, indicates a much newer interconnect standard than the A60M's PCIe 4.0 x16.

In terms of shading throughput, the B200 has 18,944 shading units and 592 TMUs, giving it a texture rate of 1,083.4 GTexel/s. The A60M's 2,048 shading units and 128 TMUs produce 166.4 GTexel/s. Every compute-oriented metric in the database favors the B200.

The Verdict

The data separates these two GPUs into distinct product categories with almost no overlap. The Intel Arc Pro A60M is a mobile graphics processor. It provides display output, supports the full range of graphics APIs, includes ray tracing hardware, and operates at 95 W. Its performance metrics, 5.325 TFLOPS FP32 and 256.0 GB/s bandwidth, are sufficient for professional mobile graphics tasks but are not in the same class as server accelerators.

The NVIDIA B200 SXM6 is a server compute module. It has no display outputs, no graphics API support, and a 1000 W TDP with a 1400 W suggested PSU. It is designed for data center workloads such as AI training, scientific simulation, and high-performance computing. Its 69.34 TFLOPS FP32 and FP16 performance, 592 tensor cores, and 180 GB HBM3e memory make it a high-capacity compute engine. The launch MSRP of 34,999 USD reflects its enterprise positioning.

A buyer selecting between these two would not be choosing between competitors in the same market. The A60M serves laptops and mobile workstations that need graphics acceleration. The B200 serves server racks that need dense compute throughput. Benchmark results in the database confirm no direct competition: zero head-to-head wins for either side, shared 50th percentile ranking, and no overlapping benchmark entries.

The choice is dictated by the workload. If the task requires rendering, display output, or standard graphics APIs, the Arc Pro A60M is the only viable option. If the task requires maximum FP32 or FP16 throughput, massive memory capacity, or tensor core acceleration, the B200 SXM6 is the answer. The 13 times gap in FP32 compute, the 32 times gap in memory bandwidth, and the 10.5 times gap in TDP all point to fundamentally different purposes. The data does not support a direct comparison, because the two parts occupy separate tiers of the hardware landscape.