Intel Arc A310E vs Intel Arc B370 Comparison

Head-to-Head Benchmarks

The recorded database contains a single benchmark result for the Intel Arc B370, a 3DMark Steel Nomad DX12 score of 1184. The Intel Arc A310E has no benchmark entries in the database, so direct numerical comparison between the two is limited to architectural throughput metrics rather than application tests. The Arc B370's score places it in the 5th percentile of all GPUs, indicating a low overall performance tier. Its nearest rivals in the database are the ATI Mobility Radeon HD 5570 at 1186 (0.2% faster), the ATI Radeon HD 5770 at 1190 (0.5% faster), and the AMD Radeon HD 7650M at 1192 (0.7% faster). The Arc B370 trails each of these legacy parts by a fraction of a percent, while it leads the AMD FirePro M2000 by 1.4%, which scores 1168.

Because the Arc A310E lacks recorded benchmark scores, the head-to-head comparison must rely on computed pixel, texture, and floating-point rates. The Arc B370 delivers 48.00 GPixel/s versus 32.00 GPixel/s for the Arc A310E, a 50% advantage in pixel throughput. Texture rate shows a similar gap: 96.00 GTexel/s versus 64.00 GTexel/s, again a 50% lead for the Arc B370. FP32 compute is 6.144 TFLOPS for the Arc B370 versus 3.072 TFLOPS for the Arc A310E, exactly double. FP16 compute follows the same 2:1 ratio, with 12.29 TFLOPS for the Arc B370 and 6.144 TFLOPS for the Arc A310E. These figures indicate that in raw shading and rasterization throughput, the Arc B370 is consistently ahead by a factor of 1.5 to 2.0.

The Arc A310E does hold an advantage in memory configuration. It has 4 GB of dedicated GDDR6 memory on a 64-bit bus, delivering 124.0 GB/s of bandwidth. The Arc B370 uses system shared memory, with bandwidth listed as system dependent. This means the Arc A310E has a fixed, predictable memory resource, while the Arc B370's performance will scale with the host platform's memory speed and allocation. For workloads that are sensitive to memory latency or sustained bandwidth, the dedicated memory of the Arc A310E may narrow the gap in real-world tests, despite the Arc B370's higher compute rates.

Clock behavior also differs sharply. The Arc A310E runs at a flat 2000 MHz for both base and boost, while the Arc B370 has a 300 MHz base clock and a 2400 MHz boost clock. The Arc B370's boost clock is 20% higher than the Arc A310E's fixed clock, but its idle and low-load base clock is dramatically lower, reflecting its integrated design and power management strategy. The Arc B370's higher boost clock contributes to its throughput advantage, while the Arc A310E's constant clock suggests more consistent operation under sustained load.

FAQ

Q: Which GPU has a higher benchmark score in the database?

A: Only the Intel Arc B370 has a recorded benchmark score: 1184 in 3DMark Steel Nomad DX12. The Intel Arc A310E has no benchmark entries in the database.

Q: How does the Arc B370 compare to its nearest rivals?

A: The Arc B370 scores 1184, which is 0.2% below the ATI Mobility Radeon HD 5570 (1186), 0.5% below the ATI Radeon HD 5770 (1190), 0.7% below the AMD Radeon HD 7650M (1192), and 1.4% above the AMD FirePro M2000 (1168).

Q: What is the FP32 compute difference between the two cards?

A: The Arc B370 delivers 6.144 TFLOPS FP32, exactly double the Arc A310E's 3.072 TFLOPS.

Q: Does the Arc A310E have dedicated memory?

A: Yes, the Arc A310E has 4 GB of GDDR6 on a 64-bit bus with 124.0 GB/s bandwidth. The Arc B370 uses system shared memory with system-dependent bandwidth.

Q: What are the power requirements of each GPU?

A: The Arc A310E has a 75 W TDP and suggests a 250 W power supply. The Arc B370 has a 25 W TDP and lists no suggested PSU, consistent with its integrated design.

Q: Which GPU has a higher pixel fill rate?

A: The Arc B370 has a pixel rate of 48.00 GPixel/s, which is 50% higher than the Arc A310E's 32.00 GPixel/s.

Architecture Differences

The two GPUs come from different architectural generations within Intel's Xe family. The Arc A310E uses the DG2-128 chip based on Xe-HPG architecture, belonging to the Alchemist (Arc 3) generation. The Arc B370 uses the Panther Lake chip based on Xe3-LPG architecture, belonging to the Arc Graphics-M (Panther Lake) generation. This represents a generational leap from the first discrete Arc architecture to the newer integrated Xe3 design.

Manufacturing processes differ substantially. The Arc A310E is fabricated on TSMC's 6 nm process with 7,200 million transistors on a 157 mm² die, yielding a transistor density of 45.9M per mm². The Arc B370 is fabricated on Intel's 3 nm process, with transistor count and die size listed as unknown in the database. The process node advancement from 6 nm to 3 nm is significant, though without transistor data for the Arc B370, density comparisons cannot be made.

Execution resources scale differently across the two chips. The Arc B370 has 1280 shading units, 40 texture mapping units, 20 ROPs, and 10 ray tracing cores. The Arc A310E has 768 shading units, 32 TMUs, 16 ROPs, and 6 ray tracing cores. The Arc B370 therefore has 66.7% more shading units, 25% more TMUs, 25% more ROPs, and 66.7% more ray tracing cores. Neither GPU lists tensor cores in the database, so AI acceleration hardware cannot be compared.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, indicating identical API feature levels despite the architectural generation gap. The Arc A310E is a discrete PCIe card using a PCIe 4.0 x8 interface, while the Arc B370 is an integrated graphics processor with an IGP bus interface. This fundamental difference in integration affects memory architecture, power delivery, and physical form factor.

Production status also differs. The Arc A310E is end-of-life, released on 2024-03-31, with a predecessor listed as Xe Graphics and a successor of Battlemage. The Arc B370 is active, released on 2026-01-26, with no predecessor or successor listed. The Arc A310E is a mature product at the end of its lifecycle, while the Arc B370 is a current, actively produced part.

Specification Differences

The two cards differ across nearly every recorded specification. Process node: 6 nm for the Arc A310E versus 3 nm for the Arc B370. Foundry: TSMC for the Arc A310E versus Intel for the Arc B370. Transistor count: 7,200 million for the Arc A310E versus unknown for the Arc B370. Die size: 157 mm² for the Arc A310E versus unknown for the Arc B370.

Clock speeds: the Arc A310E has a 2000 MHz base and 2000 MHz boost, while the Arc B370 has a 300 MHz base and 2400 MHz boost. Memory: the Arc A310E uses 4 GB GDDR6 at 1937 MHz (15.5 Gbps effective) on a 64-bit bus with 124.0 GB/s bandwidth; the Arc B370 uses system shared memory with system-dependent bandwidth. Shading units: 768 versus 1280. TMUs: 32 versus 40. ROPs: 16 versus 20. Ray tracing cores: 6 versus 10.

Pixel rate: 32.00 GPixel/s versus 48.00 GPixel/s. Texture rate: 64.00 GTexel/s versus 96.00 GTexel/s. FP32: 3.072 TFLOPS versus 6.144 TFLOPS. FP16: 6.144 TFLOPS versus 12.29 TFLOPS, both at a 2:1 ratio. TDP: 75 W versus 25 W. Slot width: single-slot versus IGP. Power connectors: none for both. Suggested PSU: 250 W for the Arc A310E versus none listed for the Arc B370.

Bus interface: PCIe 4.0 x8 for the Arc A310E versus IGP for the Arc B370. Display outputs: 4x mini-DisplayPort 2.0 for the Arc A310E versus portable device dependent for the Arc B370. Dimensions: the Arc A310E measures 168 mm by 69 mm by 20 mm; the Arc B370 has no listed dimensions. Both GPUs share the same API support: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

The Verdict

The data separates these two GPUs into distinct usage classes. The Arc B370 is the stronger compute part in every measured throughput metric: it has double the FP32 performance, 50% higher pixel fill, 50% higher texture fill, more shading units, more ROPs, more ray tracing cores, and a higher boost clock. Its benchmark score of 1184, while low in the overall GPU percentile ranking at 5%, provides concrete evidence of its real-world performance tier. The Arc A310E, with no benchmark scores, cannot be placed in the same direct ranking, but its compute rates are consistently half to two-thirds of the Arc B370's.

The Arc A310E's advantages are structural rather than performance-based. It has dedicated 4 GB GDDR6 memory with fixed 124.0 GB/s bandwidth, a discrete card form factor with PCIe 4.0 x8 connectivity, and four mini-DisplayPort 2.0 outputs. These features make it a self-contained graphics solution for systems that require a dedicated card. The Arc B370, as an integrated processor, depends on the host system for memory and display output, which makes its performance variable based on platform configuration.

From the database perspective, the Arc B370 is the faster GPU in raw throughput and the only one with a recorded benchmark result. The Arc A310E remains a viable option for workloads that need dedicated memory and a fixed physical interface, but it is end-of-life and has no benchmark data to validate its standing against the Arc B370's measured score.

Where Each One Wins

The Arc B370 wins in all compute and rasterization throughput categories. FP32 compute is 6.144 TFLOPS versus 3.072 TFLOPS, a 100% advantage. Pixel rate is 48.00 GPixel/s versus 32.00 GPixel/s, a 50% advantage. Texture rate is 96.00 GTexel/s versus 64.00 GTexel/s, a 50% advantage. FP16 compute is 12.29 TFLOPS versus 6.144 TFLOPS, again a 100% advantage. It also carries more execution resources: 1280 shading units, 40 TMUs, 20 ROPs, and 10 ray tracing cores versus 768, 32, 16, and 6 respectively. Its 2400 MHz boost clock exceeds the Arc A310E's 2000 MHz fixed clock by 20%. Power efficiency favors the Arc B370 as well, with a 25 W TDP versus 75 W, meaning it delivers higher throughput at one-third the power draw. The Arc B370 is the only one of the two with a recorded benchmark score, and that score, 1184, places it in the 5th percentile of all GPUs, with a 1.4% lead over the AMD FirePro M2000.

The Arc A310E wins in memory architecture and platform independence. Its 4 GB of dedicated GDDR6 memory with 124.0 GB/s bandwidth provides a fixed resource that does not compete with the CPU for bandwidth. The Arc B370's system shared memory has system-dependent bandwidth, meaning its memory performance is not guaranteed and will vary by host platform. The Arc A310E also offers a discrete card form factor with a PCIe 4.0 x8 interface, while the Arc B370 is an IGP with no bus interface beyond the integrated path. Display outputs are another clear win: the Arc A310E provides 4x mini-DisplayPort 2.0, while the Arc B370's display outputs are listed as portable device dependent. The Arc A310E has a fixed 2000 MHz clock with no boost variation, which may offer more consistent frame pacing in sustained workloads compared to the Arc B370's variable 300 MHz to 2400 MHz clock range. For systems with weak or slow system memory, the Arc A310E's dedicated GDDR6 will deliver more predictable memory performance than the Arc B370's shared memory approach.