AMD Instinct MI355X vs Intel Arc Pro B60 Comparison

Head-to-Head Benchmarks

The recorded data contains no direct head-to-head benchmark comparisons between the AMD Instinct MI355X and the Intel Arc Pro B60. The AMD Instinct MI355X has no benchmark entries in the database, while the Intel Arc Pro B60 has eight recorded test scores. This absence of overlapping measurements means the comparison must rely on the Intel card’s absolute scores and its position relative to its own nearest rivals, alongside the AMD card’s raw specification-derived capabilities.

The Intel Arc Pro B60’s strongest recorded result is in PassMark G3D, where it scores 14,580. This places it in the 20th percentile among all GPUs in the database. Its average benchmark score across all tests is 3,182. For context, the Intel card’s nearest rivals in the database are the NVIDIA Quadro P1000 with an average score of 3,163 (0.6% higher), the NVIDIA GeForce GT 640 at 3,210 (0.9% lower), the NVIDIA GeForce 920M at 3,287 (3.2% lower), and the NVIDIA GeForce RTX 5080 SUPER at 3,075 (3.5% higher). The Intel Arc Pro B60 essentially trades places with these older or lower-tier cards, sitting within a 3.5% band of all four rivals.

In compute workloads, the Intel Arc Pro B60 records a PassMark GPU Compute score of 7,029, which is more than double its G3D score and indicates a stronger compute-to-graphics ratio. The DirectX-specific PassMark tests show scores of 179 for DirectX 9, 122 for DirectX 11, 76 for DirectX 12, and 61 for DirectX 10. The G2D score of 763 reflects 2D desktop performance. In the 3DMark Steel Nomad DX12 test, the Intel card scores 2,646.

The AMD Instinct MI355X, by contrast, has no recorded benchmark scores, no average score, and no nearest rivals. Its percentile rank of 50 is a positional placeholder rather than a measured result. Therefore, direct numerical comparisons between the two cards cannot be made from the database. What the data does show is that the Intel Arc Pro B60 occupies a low-to-mid tier position, while the AMD card’s compute specifications suggest a fundamentally different performance class, though the database does not quantify that difference through benchmarks.

Architecture Differences

The AMD Instinct MI355X uses the CDNA 4.0 architecture on a 3 nm TSMC process, built around the MI350 256CU chip. The Intel Arc Pro B60 uses the Xe2-HPG architecture on a 5 nm TSMC process, built around the BMG-G21 chip. The manufacturing node gap is significant: 3 nm versus 5 nm, which directly impacts transistor density and power efficiency characteristics.

The AMD chip contains 185,000 million transistors on a die size of 2,380 mm², resulting in a transistor density of 77.7 million per mm². The Intel chip contains 19,600 million transistors on a die size of 272 mm², yielding 72.1 million per mm². While the AMD die is nearly nine times larger in area and holds roughly 9.4 times more transistors, the density difference is modest, only about 7.8% higher on the AMD side.

Memory architecture diverges sharply. The AMD Instinct MI355X uses 288 GB of HBM3e memory on an 8,192-bit bus, delivering 8.19 TB/s of bandwidth. The Intel Arc Pro B60 uses 24 GB of GDDR6 memory on a 192-bit bus, delivering 456.0 GB/s. The AMD card has 12 times the memory capacity and roughly 18 times the memory bandwidth. The AMD memory clock is listed at 2000 MHz (8 Gbps effective), while the Intel memory clock is 2375 MHz (19 Gbps effective), but the Intel card’s narrow bus limits its total throughput.

Compute unit counts differ by an order of magnitude. The AMD card has 16,384 shading units, 1,024 texture mapping units, and zero ROPs, reflecting its compute-focused design. The Intel card has 2,560 shading units, 160 TMUs, and 80 ROPs, and also includes 20 ray tracing cores. The AMD card has no RT cores listed. Pixel rate for the AMD card is recorded as 0 MPixel/s, while the Intel card achieves 192.0 GPixel/s. Texture rates are 2,457.6 GTexel/s for AMD and 384.0 GTexel/s for Intel.

FP32 throughput is 78.64 TFLOPS for the AMD card, while the Intel card delivers 12.29 TFLOPS. FP16 performance is 78.64 TFLOPS (1:1 ratio) for AMD, versus 24.58 TFLOPS (2:1 ratio) for Intel. This means the AMD card has 6.4 times the FP32 throughput and 3.2 times the FP16 throughput. Clock speeds are identical at boost: both cards run at 2400 MHz boost. The AMD base clock is 1000 MHz, while the Intel base clock is 2000 MHz.

Power and physical design reflect the different use cases. The AMD card has a TDP of 1400 W, requires a suggested 1800 W PSU, uses an OAM module form factor, and has no power connectors because it is designed for server sleds. The Intel card has a TDP of 200 W, requires a 550 W PSU, uses a dual-slot form factor with one 8-pin connector, and includes four mini-DisplayPort 2.1 outputs. The AMD card has no display outputs at all. Bus interfaces also differ: PCIe 5.0 x16 for AMD, PCIe 5.0 x8 for Intel.

The Verdict

The database shows two cards with fundamentally different design intents. The AMD Instinct MI355X is a compute accelerator with no display outputs, no graphics API support (DirectX, OpenGL, and Vulkan are all listed as N/A), and a 1400 W TDP. The Intel Arc Pro B60 is a workstation graphics card with full API support including DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, along with four display outputs and a 200 W TDP.

Based strictly on the recorded data, the AMD card’s compute specifications are far beyond the Intel card: 6.4 times the FP32 throughput, 3.2 times the FP16 throughput, 12 times the memory capacity, and roughly 18 times the memory bandwidth. However, the AMD card has zero benchmark scores in the database, so measured performance cannot be confirmed. The Intel card has eight recorded benchmarks, placing it in the 20th percentile overall, with an average score of 3,182 and nearest rivals within a 3.5% band.

The AMD card’s 50th percentile ranking is a default value rather than a measured result, so it should not be interpreted as a performance estimate. The data indicates the AMD card targets server compute roles where graphics output and consumer APIs are irrelevant, while the Intel card targets professional graphics workstations. Users needing display output and graphics API support must choose the Intel card, as the AMD card offers none. Users needing maximum compute throughput for server workloads, based on specification data, would select the AMD card, but the database contains no benchmark evidence to validate its real-world performance.

Specification Differences

The two cards differ in nearly every measurable specification. Process node: 3 nm versus 5 nm. Transistor count: 185,000 million versus 19,600 million. Die size: 2,380 mm² versus 272 mm². Transistor density: 77.7M per mm² versus 72.1M per mm².

Base clock: 1000 MHz versus 2000 MHz. Boost clock is identical at 2400 MHz for both. Memory clock: 2000 MHz (8 Gbps effective) versus 2375 MHz (19 Gbps effective). Memory size: 288 GB versus 24 GB. Memory type: HBM3e versus GDDR6. Memory bus: 8,192 bit versus 192 bit. Memory bandwidth: 8.19 TB/s versus 456.0 GB/s.

Shading units: 16,384 versus 2,560. TMUs: 1,024 versus 160. ROPs: 0 versus 80. RT cores: none versus 20. Pixel rate: 0 MPixel/s versus 192.0 GPixel/s. Texture rate: 2,457.6 GTexel/s versus 384.0 GTexel/s. FP32: 78.64 TFLOPS versus 12.29 TFLOPS. FP16: 78.64 TFLOPS (1:1) versus 24.58 TFLOPS (2:1).

TDP: 1400 W versus 200 W. Slot width: OAM Module versus Dual-slot. Power connectors: None versus 1x 8-pin. Suggested PSU: 1800 W versus 550 W. Bus interface: PCIe 5.0 x16 versus PCIe 5.0 x8. Display outputs: No outputs versus 4x mini-DisplayPort 2.1. DirectX support: N/A versus 12 Ultimate. OpenGL: N/A versus 4.6. Vulkan: N/A versus 1.4.

Dimensions: the AMD card is 102 mm long and 165 mm wide, while the Intel card is 167 mm long, 69 mm high, and 40 mm wide. Release dates: the AMD card launched on June 11, 2025, while the Intel card launched on September 4, 2025. The Intel card has an active production status; the AMD card’s status is not recorded. The Intel card has a launch MSRP of 499 USD; the AMD card has no listed MSRP.

FAQ

Q: Which card has higher FP32 compute throughput?

A: The AMD Instinct MI355X delivers 78.64 TFLOPS FP32, which is 6.4 times the Intel Arc Pro B60’s 12.29 TFLOPS.

Q: Does the AMD Instinct MI355X support display outputs?

A: No. The AMD card has no display outputs, while the Intel Arc Pro B60 has four mini-DisplayPort 2.1 outputs.

Q: What is the memory bandwidth difference?

A: The AMD card provides 8.19 TB/s over an 8,192-bit HBM3e interface, while the Intel card provides 456.0 GB/s over a 192-bit GDDR6 interface, a roughly 18-fold difference.

Q: How does the Intel Arc Pro B60 compare to its nearest rivals in the database?

A: The Intel card’s average benchmark score is 3,182. Its nearest rival, the NVIDIA Quadro P1000, scores 3,163 (0.6% higher), and the NVIDIA GeForce GT 640 scores 3,210 (0.9% lower). The NVIDIA GeForce RTX 5080 SUPER scores 3,075 (3.5% higher).

Q: Which card has ray tracing cores?

A: The Intel Arc Pro B60 has 20 ray tracing cores. The AMD Instinct MI355X has no RT cores listed.

Q: What are the power requirements for each card?

A: The AMD card has a 1400 W TDP with a suggested 1800 W PSU and uses an OAM module with no power connectors. The Intel card has a 200 W TDP with a suggested 550 W PSU and uses one 8-pin connector.

Where Each One Wins

The AMD Instinct MI355X wins decisively in raw compute specifications. Its FP32 throughput of 78.64 TFLOPS dwarfs the Intel card’s 12.29 TFLOPS, and its FP16 output matches FP32 at 78.64 TFLOPS, indicating a design that prioritizes consistent compute across precision formats. The 288 GB HBM3e memory pool with 8.19 TB/s bandwidth is suited for large model residency and high-throughput data movement. The 3 nm process node and 185,000 million transistor count indicate a chip built for maximum density and compute scale.

The Intel Arc Pro B60 wins in every graphics-oriented category. It has 80 ROPs versus none, a pixel rate of 192.0 GPixel/s versus 0, and full DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4 support. The 20 ray tracing cores enable hardware-accelerated ray tracing workloads. Four mini-DisplayPort 2.1 outputs support multi-monitor professional setups. The 200 W TDP and 550 W suggested PSU make it deployable in standard workstation towers, unlike the AMD card’s 1400 W TDP and OAM module form factor, which require server infrastructure.

The Intel card also wins on measured benchmark data. It has eight recorded scores, including a PassMark G3D score of 14,580 and a 3DMark Steel Nomad DX12 score of 2,646. The AMD card has zero benchmark entries, so the database provides no evidence of its real-world performance. The Intel card’s 20th percentile ranking, while modest, is at least a measured position. The AMD card’s 50th percentile is a default placeholder, not a test result.

For use-case selection: the AMD Instinct MI355X is the choice for server compute environments where graphics output is unnecessary and the supporting power and cooling infrastructure exists. The Intel Arc Pro B60 is the choice for professional graphics workstations requiring display output, API compatibility, and ray tracing capability. The data does not support using the AMD card for any graphics workload, and the data does not support using the Intel card for large-scale compute workloads given its 24 GB memory ceiling and 12.29 TFLOPS FP32 limit.