Intel Arc B370 vs NVIDIA GeForce RTX 5070 Comparison

Head-to-Head Benchmarks

The sole recorded 3DMark Steel Nomad DX12 benchmark in the database delivers a decisive result: the NVIDIA GeForce RTX 5070 scores 5077 points, while the Intel Arc B370 scores 1184 points. This gives NVIDIA a victory with a delta of -76.7 percent for the Intel part, meaning the RTX 5070 outperforms the Arc B370 by a factor of roughly 4.3 in this specific test. The margin is not close; it is a categorical difference in raw graphics throughput.

However, the database also contains a broader set of benchmark results for the RTX 5070, including Geekbench OpenCL and Vulkan scores of 172660 and 178923, respectively, alongside PassMark tests covering DirectX 9, 10, 11, 12, G2D, G3D, and GPU compute. The Arc B370 only has the single 3DMark result, so direct comparisons across those other workloads are not possible from recorded data. The RTX 5070's PassMark G3D score of 29137 and its GPU compute score of 15787 further illustrate its substantial compute capability, but no corresponding figures exist for the Arc B370 to establish a head-to-head delta.

What the data does allow is a percentile comparison. The RTX 5070 sits at the 82nd percentile among all GPUs in the database, while the Arc B370 sits at the 5th percentile. That gap of 77 percentile points is the clearest single indicator of their relative standing. In the nearest rival analysis, the Arc B370's average score of 1184 places it within 0.2 percent of the ATI Mobility Radeon HD 5570 (1186), 0.5 percent behind the ATI Radeon HD 5770 (1190), and 0.7 percent behind the AMD Radeon HD 7650M (1192). It does lead the AMD FirePro M2000 (1168) by 1.4 percent. The RTX 5070's average score of 40377, meanwhile, is 0.1 percent ahead of the AMD Radeon Pro 580 (40318), 0.8 percent ahead of the AMD Radeon Pro WX 7100 (40063), and 2 percent ahead of the NVIDIA RTX A500 Mobile (39568), while trailing the AMD Radeon Pro 5300 (40870) by 1.2 percent.

Where Each One Wins

The Intel Arc B370 wins in no recorded benchmark test. The head-to-head table shows zero victories for the Intel part, with the single 3DMark Steel Nomad test going to NVIDIA. This is not a nuanced split; the database contains no workload where the Arc B370 outperforms the RTX 5070.

The RTX 5070, by contrast, wins the only direct comparison and also posts strong absolute scores across multiple test suites. Its Geekbench Vulkan score of 178923 is higher than its OpenCL score of 172660, indicating solid compute performance under both APIs. The PassMark suite shows a peak DirectX 9 score of 320, a DirectX 11 score of 277, and a DirectX 10 score of 180, with DirectX 12 at 108. The G2D score of 1305 is modest relative to the G3D score of 29137, which is expected for a discrete graphics card focused on 3D rendering rather than 2D desktop operations.

For the Arc B370, the only available benchmark score of 1184 in 3DMark Steel Nomad places it firmly in entry-level territory, as its 5th percentile rank confirms. The nearest rivals, all older AMD and ATI parts from the HD 5000 and HD 7000 series, indicate that the Arc B370 competes with legacy integrated and low-end discrete solutions rather than modern mid-range or high-end cards.

Architecture Differences

The two GPUs come from different manufacturers, fabs, and process nodes. The Intel Arc B370 uses the Panther Lake chip with the Xe3-LPG architecture, built on Intel's 3 nm process. The NVIDIA GeForce RTX 5070 uses the GB205 chip with the Blackwell 2.0 architecture, built on TSMC's 5 nm process. The RTX 5070 has a transistor count of 31,100 million on a 263 mm² die, yielding a density of 118.3M per mm². The Arc B370's transistor count and die size are listed as unknown in the database, so no direct density comparison is possible.

Compute resources differ sharply. The RTX 5070 has 6144 shading units, 192 texture mapping units, 80 raster output units, 48 ray tracing cores, and 192 tensor cores. The Arc B370 has 1280 shading units, 40 TMUs, 20 ROPs, and 10 ray tracing cores, with tensor cores listed as null. The pixel rate for the RTX 5070 is 201.0 GPixel/s versus 48.00 GPixel/s for the Arc B370. Texture rate is 482.3 GTexel/s versus 96.00 GTexel/s. FP32 throughput is 30.87 TFLOPS for NVIDIA versus 6.144 TFLOPS for Intel. FP16 performance is 30.87 TFLOPS (1:1) for NVIDIA versus 12.29 TFLOPS (2:1) for Intel.

Memory architecture is fundamentally different. The RTX 5070 uses 12 GB of GDDR7 on a 192-bit bus with 672.0 GB/s bandwidth. The Arc B370 uses system shared memory with a system shared type, bus width, and bandwidth that is system dependent. Clock speeds also differ: the RTX 5070 has a base clock of 2325 MHz and a boost of 2512 MHz, with memory at 1750 MHz (28 Gbps effective), while the Arc B370 has a base of 300 MHz and a boost of 2400 MHz.

Power and physical design diverge completely. The RTX 5070 has a 250 W TDP, is a dual-slot card with a 1x 16-pin power connector, requires a 600 W suggested PSU, and uses a PCIe 5.0 x16 interface. The Arc B370 has a 25 W TDP, is an IGP (integrated graphics processor) with no slot width, no power connectors, and an IGP bus interface. Display outputs for the RTX 5070 are 1x HDMI 2.1b and 3x DisplayPort 2.1b, while the Arc B370's outputs are portable device dependent. The RTX 5070 has physical dimensions of 245 mm length, 115 mm height, and 40 mm width; the Arc B370 has no recorded dimensions.

Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API feature parity exists at the specification level. The release dates differ: the Arc B370 launched on 2026-01-26, while the RTX 5070 launched earlier on 2025-03-03. The RTX 5070 has a recorded predecessor (GeForce 40) and successor (GeForce 60), while the Arc B370 has neither in the database.

FAQ

Q: Which GPU has the higher 3DMark Steel Nomad DX12 score?

A: The NVIDIA GeForce RTX 5070 scores 5077, which is 76.7 percent higher than the Intel Arc B370's 1184.

Q: What is the percentile rank of each GPU among all GPUs in the database?

A: The RTX 5070 is at the 82nd percentile, while the Arc B370 is at the 5th percentile.

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

A: Its average score of 1184 is 0.2 percent below the ATI Mobility Radeon HD 5570, 0.5 percent below the ATI Radeon HD 5770, 0.7 percent below the AMD Radeon HD 7650M, and 1.4 percent above the AMD FirePro M2000.

Q: What memory configuration does each GPU use?

A: The RTX 5070 uses 12 GB of GDDR7 on a 192-bit bus with 672.0 GB/s bandwidth. The Arc B370 uses system shared memory with system dependent bandwidth.

Q: What is the TDP difference between the two?

A: The RTX 5070 has a TDP of 250 W, while the Arc B370 has a TDP of 25 W, a tenfold difference.

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.

Specification Differences

| Specification | Intel Arc B370 | NVIDIA GeForce RTX 5070 |

|---------------|----------------|--------------------------|

| Chip | Panther Lake | GB205 |

| Architecture | Xe3-LPG | Blackwell 2.0 |

| Process Node | 3 nm | 5 nm |

| Foundry | Intel | TSMC |

| Transistors | unknown | 31,100 million |

| Die Size | unknown | 263 mm² |

| Base Clock | 300 MHz | 2325 MHz |

| Boost Clock | 2400 MHz | 2512 MHz |

| Memory Size | System Shared | 12 GB |

| Memory Type | System Shared | GDDR7 |

| Memory Bus Width | System Shared | 192 bit |

| Memory Bandwidth | System Dependent | 672.0 GB/s |

| Shading Units | 1280 | 6144 |

| TMUs | 40 | 192 |

| ROPs | 20 | 80 |

| RT Cores | 10 | 48 |

| Tensor Cores | null | 192 |

| Pixel Rate | 48.00 GPixel/s | 201.0 GPixel/s |

| Texture Rate | 96.00 GTexel/s | 482.3 GTexel/s |

| FP32 | 6.144 TFLOPS | 30.87 TFLOPS |

| FP16 | 12.29 TFLOPS (2:1) | 30.87 TFLOPS (1:1) |

| TDP | 25 W | 250 W |

| Slot Width | IGP | Dual-slot |

| Power Connectors | None | 1x 16-pin |

| Suggested PSU | null | 600 W |

| Bus Interface | IGP | PCIe 5.0 x16 |

| Display Outputs | Portable Device Dependent | 1x HDMI 2.1b, 3x DisplayPort 2.1b |

| Dimensions | null | 245 mm x 115 mm x 40 mm |

| Release Date | 2026-01-26 | 2025-03-03 |

| Predecessor | null | GeForce 40 |

| Successor | null | GeForce 60 |

| Launch MSRP | null | 549 USD |

The Verdict

The recorded data separates these two GPUs into distinct categories. The Intel Arc B370 is an integrated graphics solution with a 25 W TDP, system shared memory, and a single benchmark score of 1184 that places it at the 5th percentile, alongside legacy ATI and AMD parts from over a decade ago. Its nearest rivals, all scoring within 0.2 to 0.7 percent of each other, confirm that it delivers performance comparable to old entry-level discrete GPUs rather than modern gaming hardware.

The NVIDIA GeForce RTX 5070 is a discrete, dual-slot, 250 W card with 12 GB of dedicated GDDR7 memory, 6144 shading units, 48 RT cores, and a 672.0 GB/s memory bandwidth. Its 3DMark score of 5077 is 76.7 percent higher than the Arc B370's, and its 82nd percentile rank places it well above the median GPU. The PassMark G3D score of 29137 and GPU compute score of 15787 further demonstrate its multi-faceted strength across compute and rendering workloads.

For users requiring a discrete graphics card with high throughput, ray tracing capability, and dedicated memory, the RTX 5070 is the only viable option in this comparison. Its launch MSRP is 549 USD. For users relying on integrated graphics in a power-constrained system, the Arc B370 provides basic rendering capability at a fraction of the power draw, but the benchmark data shows no workload where it competes with the RTX 5070. The verdict from the database is unambiguous: the RTX 5070 wins the only head-to-head test, holds a far higher percentile rank, and offers dramatically more compute resources across every measured specification.