Intel Arc Pro B60 Dual vs Intel Data Center GPU Max Subsystem Comparison

Head-to-Head Benchmarks

The recorded database contains no head-to-head benchmark entries for these two Intel accelerators. Both the Intel Arc Pro B60 Dual and the Intel Data Center GPU Max Subsystem have empty benchmark arrays, and the head-to-head benchmark field is also empty. Consequently, there are no measured performance scores, no delta percentages, and no win counts to compare directly. The average benchmark score for both products is recorded as zero, and each sits at the 50th percentile among all GPUs in the database, which indicates that neither has accumulated any standardized test results to date.

What the data does provide is a stark contrast in raw computational specifications, which can serve as a proxy for expected performance in different workloads. The Arc Pro B60 Dual delivers 12.29 TFLOPS of FP32 compute and 24.58 TFLOPS of FP16 compute with a 2:1 ratio. The Data Center GPU Max Subsystem, in comparison, delivers 52.43 TFLOPS of FP32 and 52.43 TFLOPS of FP16 with a 1:1 ratio. That is a 4.27x advantage in FP32 throughput and a 2.13x advantage in FP16 throughput for the data center part, based strictly on the figures in the database.

Memory bandwidth shows an even larger gap. The Arc Pro B60 Dual has a 456.0 GB/s memory bandwidth over a 192-bit bus with 24 GB of GDDR6. The Data Center GPU Max Subsystem has 3.21 TB/s of memory bandwidth over an 8192-bit bus with 128 GB of HBM2e. Converting the data center part to the same unit, 3.21 TB/s equals 3,210 GB/s, which is 7.04x the bandwidth of the Arc Pro B60 Dual. The bus width difference is 8192 bits versus 192 bits, a 42.67x difference in interface width, though the memory clock differs as well: 2375 MHz (19 Gbps effective) for the Arc Pro B60 Dual versus 1565 MHz (3.1 Gbps effective) for the Data Center GPU Max Subsystem.

Texture rate also favors the data center part substantially. The Arc Pro B60 Dual reaches 384.0 GTexel/s, while the Data Center GPU Max Subsystem reaches 1,638.4 GTexel/s, a 4.27x advantage that mirrors the FP32 ratio. Pixel rate, however, is a notable anomaly: the Arc Pro B60 Dual produces 192.0 GPixel/s, while the Data Center GPU Max Subsystem is recorded at 0 MPixel/s, meaning it has no raster output units (ROPs are listed as 0). This indicates the data center part is not designed for traditional rasterized graphics output, which aligns with its lack of display outputs.

Neither product has any nearest rivals listed in the database, so no relative performance positioning against other GPUs can be established. The empty benchmark data means the only quantitative comparison available comes from the specification-level differences outlined above.

Where Each One Wins

Based on the recorded specifications, the Intel Arc Pro B60 Dual wins in several categories that matter for graphics-oriented and professional visualization workloads. It has 4x mini-DisplayPort 2.1 outputs, while the Data Center GPU Max Subsystem has no outputs at all. The Arc Pro B60 Dual supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, whereas the Data Center GPU Max Subsystem only supports DirectX 12 (12_1) and has no Vulkan support listed. The Arc Pro B60 Dual also has a higher base clock at 2000 MHz versus 900 MHz, and a higher boost clock at 2400 MHz versus 1600 MHz. Its pixel rate of 192.0 GPixel/s confirms it can drive displays, while the data center part cannot render pixels at all.

The Arc Pro B60 Dual also wins on physical dimensions and power integration. It measures 300 mm in length, 110 mm in height, and 40 mm in width, with a dual-slot design and a single 16-pin power connector. The Data Center GPU Max Subsystem measures 267 mm in length with no height or width recorded, also dual-slot, also with a single 16-pin connector. The Arc Pro B60 Dual suggests an 800 W power supply, while the Data Center GPU Max Subsystem suggests a 2800 W power supply. The Arc Pro B60 Dual has a lower TDP at 400 W versus 2400 W for the data center part, making it far more feasible for desktop-class systems.

The Intel Data Center GPU Max Subsystem wins decisively in compute and memory capacity. It has 128 GB of HBM2e memory versus 24 GB of GDDR6, a 5.33x capacity advantage. It has 16384 shading units versus 2560, a 6.4x advantage. It has 1024 texture mapping units versus 160, a 6.4x advantage. It has 128 ray tracing cores versus 20, a 6.4x advantage. Its FP32 throughput of 52.43 TFLOPS is 4.27x the Arc Pro B60 Dual's 12.29 TFLOPS. Its FP16 throughput of 52.43 TFLOPS is 2.13x the Arc Pro B60 Dual's 24.58 TFLOPS, and the data center part achieves this without a 2:1 ratio, meaning it does not halve its rate for FP16. The data center part also has a 64x wider memory bus (8192 bit versus 192 bit) and 7.04x the memory bandwidth.

The Data Center GPU Max Subsystem uses a PCIe 5.0 x16 interface, while the Arc Pro B60 Dual uses PCIe 5.0 x8. The data center part has a larger die at 1280 mm² versus 272 mm², more transistors at 100,000 million versus 19,600 million, and a higher transistor density at 78.1M per mm² versus 72.1M per mm².

The Verdict

The data indicates two distinct products with no overlapping use cases. The Intel Arc Pro B60 Dual is a graphics-capable accelerator with display outputs, graphics APIs, and a pixel pipeline. It supports DirectX 12 Ultimate, Vulkan 1.4, and OpenGL 4.6, and it can output to four displays simultaneously. Its 12.29 TFLOPS FP32 and 24.58 TFLOPS FP16 performance, combined with 456.0 GB/s bandwidth and 24 GB of memory, positions it for professional graphics tasks such as rendering, visualization, and content creation on workstations. Its 400 W TDP and 800 W suggested power supply make it compatible with standard workstation power delivery.

The Intel Data Center GPU Max Subsystem is a compute-focused accelerator with no display outputs, no pixel rate, and no Vulkan support. Its 52.43 TFLOPS FP32 and FP16 performance (at a 1:1 ratio) and 3.21 TB/s bandwidth over 128 GB of HBM2e target large-scale data center compute workloads such as high-performance computing, AI training, and scientific simulation. Its 2400 W TDP and 2800 W suggested power supply require dedicated data center infrastructure. The 8192-bit memory bus and 128 GB capacity indicate a design for massive datasets that cannot fit in the 24 GB frame buffer of the Arc Pro B60 Dual.

The absence of benchmark results in the database means the verdict rests entirely on architectural intent. The Arc Pro B60 Dual wins for any workload that requires rasterization, display output, or graphics API compatibility. The Data Center GPU Max Subsystem wins for any workload that prioritizes raw compute throughput, memory capacity, and memory bandwidth, and that does not require graphics output. The data shows no benchmark overlap, so any performance ranking between the two cannot be derived from measured results. Instead, the specification differences point to complementary roles: one for the desktop and professional graphics market, one for the server and data center market.

The release dates also reflect this split. The Arc Pro B60 Dual has a release date of 2025-09-04, while the Data Center GPU Max Subsystem has a release date of 2023-01-09. The Arc Pro B60 Dual carries a launch MSRP of 1,199 USD, which can be stated once per the database record. The Data Center GPU Max Subsystem has no recorded launch MSRP. The Arc Pro B60 Dual is built on a 5 nm TSMC process, while the Data Center GPU Max Subsystem uses a 10 nm Intel process. The Arc Pro B60 Dual has a successor listed as none, while the Data Center GPU Max Subsystem has a successor listed as H3C Graphics.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: The Intel Data Center GPU Max Subsystem has 52.43 TFLOPS of FP32 performance, which is 4.27x the 12.29 TFLOPS of the Intel Arc Pro B60 Dual.

Q: Does the Intel Data Center GPU Max Subsystem support display outputs?

A: No. The database lists "No outputs" for the Data Center GPU Max Subsystem, while the Arc Pro B60 Dual has 4x mini-DisplayPort 2.1 outputs.

Q: What is the memory capacity difference between the two?

A: The Data Center GPU Max Subsystem has 128 GB of HBM2e memory, while the Arc Pro B60 Dual has 24 GB of GDDR6 memory. The data center part has 5.33x the capacity.

Q: Which GPU has a wider memory bus?

A: The Data Center GPU Max Subsystem has an 8192-bit memory bus, which is 42.67x wider than the 192-bit bus of the Arc Pro B60 Dual.

Q: What are the power requirements for each?

A: The Arc Pro B60 Dual has a 400 W TDP and suggests an 800 W power supply. The Data Center GPU Max Subsystem has a 2400 W TDP and suggests a 2800 W power supply.

Q: Which GPU supports Vulkan?

A: The Arc Pro B60 Dual supports Vulkan 1.4. The Data Center GPU Max Subsystem has no Vulkan version listed in the database.

Architecture Differences

The two GPUs come from different architectural generations and design philosophies. The Intel Arc Pro B60 Dual is based on the BMG-G21 chip using the Xe2-HPG architecture, part of the Battlemage (Pro Series) generation. It is manufactured on a 5 nm process at TSMC. The Intel Data Center GPU Max Subsystem is based on the Ponte Vecchio chip using the Generation 12.5 architecture, part of the Data Center GPU (Ponte Vecchio) generation. It is manufactured on a 10 nm process at Intel's own foundry.

Transistor counts differ dramatically. The Arc Pro B60 Dual has 19,600 million transistors on a 272 mm² die, yielding a density of 72.1M per mm². The Data Center GPU Max Subsystem has 100,000 million transistors on a 1280 mm² die, yielding a density of 78.1M per mm². The data center part has 5.1x the transistors and 4.7x the die area, but only a 1.08x higher transistor density.

The compute architectures also diverge in their FP16 handling. The Arc Pro B60 Dual achieves 24.58 TFLOPS of FP16 through a 2:1 ratio, meaning it processes two FP16 operations per FP32 operation. The Data Center GPU Max Subsystem achieves 52.43 TFLOPS of FP16 at a 1:1 ratio, meaning it does not gain any throughput advantage for FP16 over FP32. This suggests the data center part is built for FP32-heavy workloads or for FP16 workloads that require the same precision path.

The rendering pipelines are fundamentally different. The Arc Pro B60 Dual has 80 ROPs and a pixel rate of 192.0 GPixel/s, indicating a full rasterization pipeline. The Data Center GPU Max Subsystem has 0 ROPs and a pixel rate of 0 MPixel/s, indicating no rasterization capability. The Arc Pro B60 Dual has 160 TMUs and a texture rate of 384.0 GTexel/s, while the Data Center GPU Max Subsystem has 1024 TMUs and a texture rate of 1,638.4 GTexel/s. The data center part has 6.4x the TMUs and 4.27x the texture rate.

Ray tracing hardware also differs. The Arc Pro B60 Dual has 20 ray tracing cores, while the Data Center GPU Max Subsystem has 128 ray tracing cores, a 6.4x advantage. Neither product lists tensor cores in the database.

The memory subsystems use entirely different technologies. The Arc Pro B60 Dual uses GDDR6 at 2375 MHz (19 Gbps effective) over a 192-bit bus. The Data Center GPU Max Subsystem uses HBM2e at 1565 MHz (3.1 Gbps effective) over an 8192-bit bus. The data center part compensates for a lower memory clock with a vastly wider bus, achieving 3.21 TB/s versus 456.0 GB/s. The Arc Pro B60 Dual has a 24 GB frame buffer, while the data center part has a 128 GB frame buffer.

Specification Differences

The two GPUs differ in nearly every recorded specification field. The Arc Pro B60 Dual uses a 5 nm TSMC process, while the Data Center GPU Max Subsystem uses a 10 nm Intel process. The Arc Pro B60 Dual has 19,600 million transistors, while the data center part has 100,000 million. Die size is 272 mm² versus 1280 mm². Transistor density is 72.1M per mm² versus 78.1M per mm².

Clock speeds are higher on the Arc Pro B60 Dual: base clock of 2000 MHz versus 900 MHz, boost clock of 2400 MHz versus 1600 MHz. Memory clock is 2375 MHz (19 Gbps effective) versus 1565 MHz (3.1 Gbps effective). Memory size is 24 GB GDDR6 versus 128 GB HBM2e. Memory bus width is 192 bit versus 8192 bit. Memory bandwidth is 456.0 GB/s versus 3.21 TB/s.

The compute unit counts show the data center part's advantage: 2560 shading units versus 16384, 160 TMUs versus 1024, 80 ROPs versus 0, and 20 ray tracing cores versus 128. Pixel rate is 192.0 GPixel/s versus 0 MPixel/s. Texture rate is 384.0 GTexel/s versus 1,638.4 GTexel/s. FP32 is 12.29 TFLOPS versus 52.43 TFLOPS. FP16 is 24.58 TFLOPS (2:1) versus 52.43 TFLOPS (1:1).

TDP is 400 W versus 2400 W. Slot width is dual-slot for both. Power connectors are 1x 16-pin for both. Suggested PSU is 800 W versus 2800 W. Bus interface is PCIe 5.0 x8 versus PCIe 5.0 x16. Display outputs are 4x mini-DisplayPort 2.1 versus none. DirectX support is 12 Ultimate (12_2) versus 12 (12_1). OpenGL is 4.6 for both. Vulkan is 1.4 versus none.

Dimensions show the Arc Pro B60 Dual at 300 mm length, 110 mm height, and 40 mm width. The Data Center GPU Max Subsystem is 267 mm in length with no height or width recorded. The Arc Pro B60 Dual has a release date of 2025-09-04 and a launch MSRP of 1,199 USD. The Data Center GPU Max Subsystem has a release date of 2023-01-09 and no launch MSRP. The Arc Pro B60 Dual has no predecessor or successor listed. The Data Center GPU Max Subsystem has no predecessor but lists H3C Graphics as its successor. Production status is Active for both.