Intel Arc B370 vs Intel Arc B580 Comparison

Intel Arc B370 is an integrated graphics processor built for Panther Lake mobile platforms, while Intel Arc B580 is a discrete, dual-slot add-in card from the Battlemage generation. The recorded benchmark data shows one head-to-head result, and the two parts are separated by a wide performance gap that aligns with their different roles and physical designs.

FAQ

Q: Which GPU is faster in the recorded 3DMark Steel Nomad DX12 test?

A: Intel Arc B580 scores 3068, which is 61.4% higher than Intel Arc B370’s 1184. The B580 wins the only shared benchmark in the database.

Q: How does each GPU rank against all other GPUs in the database?

A: Intel Arc B370 sits at the 5th percentile, while Intel Arc B580 sits at the 68th percentile. This indicates the B580 is positioned far higher in the overall performance distribution.

Q: What are the nearest rivals for each GPU based on average benchmark score?

A: Intel Arc B370’s nearest rival is ATI Mobility Radeon HD 5570 with an average score of 1186, a 0.2% difference. Intel Arc B580’s nearest rival is AMD Radeon RX 580 2048SP with an average score of 23061, a 0.2% difference.

Q: What memory configuration does each GPU use?

A: Intel Arc B370 uses system shared memory with system dependent bandwidth and no dedicated VRAM. Intel Arc B580 uses 12 GB of GDDR6 memory on a 192-bit bus with 456.0 GB/s bandwidth.

Q: What is the power draw of each GPU?

A: Intel Arc B370 has a 25 W TDP and requires no power connectors, functioning as an integrated GPU. Intel Arc B580 has a 190 W TDP, uses one 8-pin power connector, and recommends a 450 W power supply.

Q: Do both GPUs support the same graphics APIs?

A: Yes, both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Architecture Differences

Intel Arc B370 uses the Panther Lake chip built on Intel’s 3 nm process, employing the Xe3-LPG architecture. It belongs to the Arc Graphics-M (Panther Lake) generation. Intel Arc B580 uses the BMG-G21 chip built on TSMC’s 5 nm process, employing the Xe2-HPG architecture from the Battlemage (Arc 5) generation. The transistor counts differ substantially: B580 packs 19,600 million transistors on a 272 mm² die with a density of 72.1M per mm², while B370’s transistor count and die size are not recorded.

The shading engine scales directly with these architectural differences. B370 has 1280 shading units, 40 texture mapping units, 20 ROPs, and 10 ray tracing cores. B580 has 2560 shading units, 160 TMUs, 80 ROPs, and 20 ray tracing cores, exactly doubling each of those resources. The B580 also runs at a much higher clock: 2670 MHz base and boost versus 300 MHz base and 2400 MHz boost for the B370.

The memory architecture is a fundamental split. B370 shares system memory with the host processor, with no dedicated VRAM, no fixed bus width, and bandwidth that depends on the system. B580 uses 12 GB of GDDR6 on a 192-bit interface with 456.0 GB/s of dedicated bandwidth. This gives the discrete card predictable, high-bandwidth memory access that the integrated part cannot match.

Head-to-Head Benchmarks

The database contains a single direct comparison between these two GPUs: 3DMark Steel Nomad DX12. Intel Arc B580 scores 3068, while Intel Arc B370 scores 1184. The delta is 61.4% in favor of the B580. In absolute terms, the B580 delivers roughly 2.6 times the score of the B370 in this test.

The B580’s advantage in this benchmark aligns with its larger compute resource pool. The B580 has twice the shading units, TMUs, ROPs, and ray tracing cores of the B370. It also runs at a higher boost clock, 2670 MHz versus 2400 MHz. The combination of double the execution units and a higher clock produces the large score gap observed.

The B370’s nearest rivals in the database, ATI Mobility Radeon HD 5570 and ATI Radeon HD 5770, score 1186 and 1190 respectively, both within 0.5% of the B370’s 1184. This places the integrated GPU at roughly the level of a decade-old discrete mobile part. The B580’s nearest rivals, AMD Radeon RX 580 2048SP and NVIDIA GeForce RTX 2080, score 23061 and 22895 respectively, with the B580 sitting between them. The B580 also has additional benchmark entries, including Geekbench OpenCL at 92821, Geekbench Vulkan at 109672, and Passmark G3D at 15748, which the B370 lacks.

Specification Differences

The two GPUs differ across nearly every measurable specification. The process node is 3 nm for B370 versus 5 nm for B580, with different foundries: Intel for B370, TSMC for B580. The chip designations differ: Panther Lake for B370, BMG-G21 for B580. The B580 has a recorded transistor count of 19,600 million and die size of 272 mm²; these figures are unknown for the B370.

Clock speeds differ significantly. B370 has a 300 MHz base and 2400 MHz boost. B580 has a 2670 MHz base and 2670 MHz boost, meaning the boost clock is effectively the sustained clock. Memory clocks also differ: B370 uses system shared memory with no dedicated clock, while B580 runs at 2375 MHz with 19 Gbps effective.

Compute throughput shows the scale of the gap. B370 delivers 6.144 TFLOPS FP32 and 12.29 TFLOPS FP16 (2:1). B580 delivers 13.67 TFLOPS FP32 and 27.34 TFLOPS FP16 (2:1). Pixel rate is 48.00 GPixel/s for B370 versus 213.6 GPixel/s for B580. Texture rate is 96.00 GTexel/s for B370 versus 427.2 GTexel/s for B580.

Physical and power characteristics diverge completely. B370 is an IGP with a 25 W TDP, no power connectors, and no slot width requirement. B580 is a dual-slot card measuring 272 mm in length, 115 mm in height, and 45 mm in width, with a 190 W TDP, one 8-pin power connector, and a suggested 450 W power supply. The B580 uses a PCIe 4.0 x8 bus interface, while the B370 uses an integrated graphics path with no external bus. Display outputs also differ: B370 depends on the portable device, while B580 provides 1x HDMI 2.1a and 3x DisplayPort 2.1.

Release timing differs by over a year. B580 launched on 2024-12-12 with a launch MSRP of 249 USD. B370 launched on 2026-01-26 with no recorded MSRP. The B580’s predecessor is recorded as Alchemist; the B370 has no recorded predecessor.

Where Each One Wins

Intel Arc B580 wins the only direct benchmark, 3DMark Steel Nomad DX12, by a 61.4% margin. The B580 also holds a substantial resource advantage: double the shading units, TMUs, ROPs, and ray tracing cores, plus 12 GB of dedicated GDDR6 with 456.0 GB/s bandwidth. The B580’s 68th percentile ranking versus the B370’s 5th percentile confirms a wide overall performance gap. For any workload that stresses raw graphics throughput, dedicated memory bandwidth, or ray tracing resource counts, the B580 is the clear winner.

Intel Arc B370 wins in efficiency and integration. Its 25 W TDP is dramatically lower than the B580’s 190 W. It requires no power connectors, no expansion slot, and no separate power supply recommendation. It uses system shared memory, which removes the need for dedicated VRAM allocation. Its 3 nm process node and integrated design make it suitable for compact portable devices where the B580’s dual-slot 272 mm card cannot physically fit. The B370’s performance level, near that of ATI Mobility Radeon HD 5570 and ATI Radeon HD 5770, is sufficient for basic graphics tasks in a low-power platform.

The benchmark record shows no test where the B370 wins against the B580. The wins count is 0 for B370 and 1 for B580. The B370’s only recorded benchmark score, 1184, falls below every B580 benchmark entry, including the lowest B580 result in Passmark DirectX 10 at 76, though that test measures different workloads and is not directly comparable.

The Verdict

The data indicates that Intel Arc B580 is the substantially faster GPU for 3D rendering workloads. Its 3DMark Steel Nomad score of 3068 versus 1184, its 68th percentile standing, and its doubled compute resources all point to a part designed for discrete graphics performance. The B580’s nearest rivals include the NVIDIA GeForce RTX 2080 and RTX 3080, which places it in the upper-midrange tier of the database’s performance distribution. The recorded 61.4% delta in the shared benchmark is decisive.

Intel Arc B370, by contrast, is a low-power integrated solution. Its 5th percentile ranking and proximity to ATI Mobility Radeon HD 5570 and AMD FirePro M2000 indicate it belongs to the entry-level segment. The 25 W TDP and IGP form factor make it a fit for thin-and-light systems where discrete graphics are not an option. Its performance is adequate for basic acceleration but not competitive with the B580 in any recorded metric.

For a builder selecting between these two, the choice depends entirely on the platform. A portable device with an integrated Panther Lake processor gets the B370 by design. A desktop or larger laptop that can accommodate a dual-slot, 190 W card with a 450 W power supply gets the B580. The benchmark data shows no scenario where the B370 outperforms the B580, so the B580 is the correct pick whenever the physical and power constraints allow it.