Intel Arc A380E vs Intel Arc B370 Comparison

Where Each One Wins

The recorded benchmark data splits these two Intel parts into very different performance classes. The Intel Arc A380E, despite being an end-of-life discrete graphics card, occupies a substantially higher performance tier than the Intel Arc B370, which is an active integrated graphics processor. The A380E sits at the 50th percentile among all GPUs in the database, meaning half of all recorded graphics processors perform below it and half perform above it. The B370, by contrast, sits at only the 5th percentile, placing it in the lower performance range of the database.

The B370 has one recorded benchmark result: a 3DMark Steel Nomad DX12 score of 1184. This is a modern, demanding DirectX 12 workload that stresses overall graphics throughput. The A380E has no recorded benchmark scores in the database, so its wins must be inferred from its architectural specifications and its percentile standing. What the data shows is that the A380E is positioned to win in scenarios requiring sustained rendering throughput, higher memory bandwidth, and dedicated graphics memory. The B370 wins in scenarios requiring minimal power draw and system integration, as it is an integrated graphics processor with a 25 W TDP.

For a builder choosing between a dedicated low-profile card and an integrated solution, the A380E wins on raw graphics capability. Its 50th percentile ranking indicates it can handle mainstream gaming and content creation workloads at reasonable settings. The B370, ranked at the 5th percentile, is suited for lightweight tasks, basic desktop acceleration, and legacy or undemanding applications. The data indicates the A380E is the choice for anyone needing reliable discrete graphics performance, while the B370 is the choice for ultra-low-power portable systems where dedicated graphics is not an option.

Architecture Differences

The two processors come from fundamentally different Intel architectures and manufacturing processes. The A380E uses the DG2-128 chip built on the Xe-HPG architecture, which belongs to the Alchemist generation, branded as Arc 3. It is fabricated on a 6 nm process at TSMC. The chip contains 7,200 million transistors on a die size of 157 mm², resulting in a transistor density of 45.9 million transistors per square millimeter.

The B370 uses the Panther Lake chip built on the Xe3-LPG architecture, belonging to the Arc Graphics-M generation for Panther Lake. It is fabricated on a 3 nm process at Intel's own foundry. The database lists transistor count and die size as unknown for this part, so no density comparison is possible. What is clear is that the B370 is a much more modern design, using a smaller process node, but it operates as an integrated graphics processor rather than a discrete card.

The A380E uses 6 GB of dedicated GDDR6 memory on a 96-bit bus, delivering 186.0 GB/s of bandwidth. The B370 uses system shared memory, meaning its memory size, type, bus width, and bandwidth are all system dependent. This is a critical architectural difference: the A380E has guaranteed dedicated memory bandwidth, while the B370 must share system memory with the CPU and other components.

The compute resources differ significantly as well. The A380E has 1,024 shading units, 64 texture mapping units, and 32 raster output units. The B370 has 1,280 shading units, 40 texture mapping units, and 20 raster output units. Despite having fewer shading units, the A380E has more TMUs and ROPs, which often matters more for fill-rate-bound workloads. The A380E also has 8 ray tracing cores, while the B370 has 10. Neither part includes tensor cores in the database.

The clock behavior is notably different. The A380E runs at a flat 2,000 MHz base and boost clock, with memory at 1,937 MHz or 15.5 Gbps effective. The B370 has a 300 MHz base clock and a 2,400 MHz boost clock, a much wider dynamic range typical of integrated processors that scale down aggressively when idle. The B370's memory clock is listed as system shared, so no fixed memory frequency applies.

The A380E is a single-slot card measuring 254 mm in length, 127 mm in height, and 20 mm in width, with no power connectors required and a suggested power supply of 250 W. The B370 is an integrated graphics processor with no slot width, no dimensions listed, and no suggested PSU. Both parts support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API feature support is identical.

Head-to-Head Benchmarks

The database contains no direct head-to-head benchmark results between the A380E and the B370. However, the B370 has one standalone benchmark score, and the A380E's performance position can be inferred from its 50th percentile ranking. The B370 scored 1184 in 3DMark Steel Nomad DX12, which places it in a narrow band of comparable GPUs.

The nearest rivals to the B370 in the database are all older, lower-performance parts. The ATI Mobility Radeon HD 5570 scores 1186, which is 0.2 percent higher than the B370. The ATI Radeon HD 5770 scores 1190, 0.5 percent higher. The AMD Radeon HD 7650M scores 1192, 0.7 percent higher. The AMD FirePro M2000 scores 1168, which is 1.4 percent lower than the B370. This tight clustering shows the B370 performs at a level comparable to a decade-old midrange desktop card and a slightly older mobile part.

The A380E, by contrast, sits at the 50th percentile of all GPUs. This percentile is a database-wide measure, not a direct comparison to the B370, but it indicates the A380E outranks a significant majority of recorded graphics processors. Since the B370 sits at the 5th percentile, the A380E is positioned far above it in overall performance ranking.

In practical terms, the A380E's FP32 compute is rated at 4.096 TFLOPS, while the B370 is rated at 6.144 TFLOPS. The B370 has a higher raw floating-point throughput in theory. However, the A380E has a higher pixel rate at 64.00 GPixel/s versus 48.00 GPixel/s for the B370, and a higher texture rate at 128.0 GTexel/s versus 96.00 GTexel/s. These fill-rate advantages, combined with dedicated GDDR6 memory bandwidth of 186.0 GB/s versus system-dependent shared memory, explain why the A380E holds a higher percentile despite lower raw FP32 numbers.

The B370's FP16 throughput is 12.29 TFLOPS, compared to 8.192 TFLOPS for the A380E, both at a 2:1 ratio to FP32. This gives the B370 a theoretical advantage in workloads that use reduced precision, but the memory bandwidth limitation of shared system memory likely constrains real-world gains.

FAQ

Q: Which GPU has more shading units?

A: The Intel Arc B370 has 1,280 shading units, while the Intel Arc A380E has 1,024 shading units. Despite this, the A380E holds a higher percentile ranking at 50 versus the B370's 5th percentile.

Q: What is the memory difference between the two?

A: The A380E has 6 GB of dedicated GDDR6 memory on a 96-bit bus with 186.0 GB/s bandwidth. The B370 uses system shared memory, so its memory size, type, bus width, and bandwidth are all system dependent.

Q: How do their power requirements compare?

A: The A380E has a 75 W TDP and is a single-slot card with no power connectors and a suggested PSU of 250 W. The B370 has a 25 W TDP and is an integrated graphics processor with no power connectors and no suggested PSU listed.

Q: What process nodes do these chips use?

A: The A380E uses a 6 nm process at TSMC. The B370 uses a 3 nm process at Intel. The A380E's die size is 157 mm² with 7,200 million transistors, while the B370's die size and transistor count are listed as unknown.

Q: Which GPU has a better benchmark score?

A: The B370 has a recorded 3DMark Steel Nomad DX12 score of 1184. The A380E has no recorded benchmark scores in the database, but it sits at the 50th percentile of all GPUs, while the B370 sits at the 5th percentile.

Q: Are there any differences in API support?

A: No. Both the A380E and the B370 support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Neither part has tensor cores listed in the database.

The Verdict

The data clearly separates these two Intel graphics processors into different use cases. The Intel Arc A380E is a discrete, single-slot graphics card with dedicated GDDR6 memory, a 75 W TDP, and a 50th percentile ranking among all GPUs. It is end-of-life, meaning it is no longer in production, but its specifications indicate it remains a capable option for systems that need a low-power discrete card with guaranteed memory bandwidth. Its 2,000 MHz flat clock, 64 TMUs, 32 ROPs, and 186.0 GB/s memory bandwidth make it suitable for mainstream rendering tasks where fill rate and memory bandwidth matter more than raw FP32 compute.

The Intel Arc B370 is an active, integrated graphics processor built on a modern 3 nm process with a 25 W TDP. It scores 1184 in 3DMark Steel Nomad DX12, placing it in the same performance band as the ATI Mobility Radeon HD 5570, ATI Radeon HD 5770, AMD Radeon HD 7650M, and AMD FirePro M2000. Its 5th percentile ranking means it is best suited for light workloads, basic display output, and power-constrained portable devices. Its 2,400 MHz boost clock and 1,280 shading units give it theoretical compute advantages, but the system-dependent shared memory and lower fill rates hold it back.

For a builder selecting between these two, the A380E is the choice for a desktop or workstation that needs dependable discrete graphics without external power connectors. The B370 is the choice for an ultra-portable or low-power integrated system where dedicated graphics is not feasible. The A380E wins on memory bandwidth, fill rate, and overall percentile standing. The B370 wins on power efficiency, process node modernity, and raw FP32 and FP16 compute figures. Neither part has head-to-head benchmark data, so the verdict relies on the recorded B370 score, the A380E's percentile position, and the architectural specifications in the database.

Specification Differences

The following fields differ between the Intel Arc A380E and the Intel Arc B370 in the database:

• Chip: DG2-128 versus Panther Lake
• Architecture: Xe-HPG versus Xe3-LPG
• Generation: Alchemist (Arc 3) versus Arc Graphics-M (Panther Lake)
• Process Node: 6 nm versus 3 nm
• Foundry: TSMC versus Intel
• Transistors: 7,200 million versus unknown
• Die Size: 157 mm² versus unknown
• Transistor Density: 45.9M / mm² versus null
• Base Clock: 2000 MHz versus 300 MHz
• Boost Clock: 2000 MHz versus 2400 MHz
• Memory Clock: 1937 MHz 15.5 Gbps effective versus system shared
• Memory Size: 6 GB versus system shared
• Memory Type: GDDR6 versus system shared
• Memory Bus Width: 96 bit versus system shared
• Memory Bandwidth: 186.0 GB/s versus system dependent
• Shading Units: 1024 versus 1280
• TMUs: 64 versus 40
• ROPs: 32 versus 20
• RT Cores: 8 versus 10
• Pixel Rate: 64.00 GPixel/s versus 48.00 GPixel/s
• Texture Rate: 128.0 GTexel/s versus 96.00 GTexel/s
• FP32: 4.096 TFLOPS versus 6.144 TFLOPS
• FP16: 8.192 TFLOPS (2:1) versus 12.29 TFLOPS (2:1)
• TDP: 75 W versus 25 W
• Slot Width: Single-slot versus IGP
• Suggested PSU: 250 W versus null
• Bus Interface: PCIe 4.0 x8 versus IGP
• Display Outputs: 4x DisplayPort 2.0 versus portable device dependent
• Dimensions: 254 mm 10 inches, 127 mm 5 inches, 20 mm 0.8 inches versus null
• Production Status: End-of-life versus active
• Release Date: 2024-03-31 versus 2026-01-26
• Predecessor: Xe Graphics versus null
• Successor: Battlemage versus null
• Percentile vs All GPUs: 50 versus 5
• Average Benchmark Score: 0 versus 1184
• Nearest Rivals: none versus ATI Mobility Radeon HD 5570, ATI Radeon HD 5770, AMD Radeon HD 7650M, AMD FirePro M2000

Fields that are identical include manufacturer, DirectX, OpenGL, and Vulkan support, power connectors (none), and tensor cores (null for both). The launch MSRP is null for both parts, so no pricing information is available in the database.