Intel Arc B370 vs NVIDIA RTX 6000D Comparison

Head-to-Head Benchmarks

The single recorded head-to-head benchmark in the database is 3DMark Steel Nomad DX12. The NVIDIA RTX 6000D scores 3522, while the Intel Arc B370 scores 1184. This produces a delta of -66.4% for the Intel part, meaning the NVIDIA card delivers nearly three times the performance in this test. The RTX 6000D wins the only direct comparison, and the data records zero wins for the Intel Arc B370 against one win for NVIDIA.

The magnitude of this gap is substantial. In absolute terms, the RTX 6000D posts a score that is 2338 points higher than the Arc B370. The Intel part sits at the 5th percentile of all GPUs in the database, while the RTX 6000D sits at the 98th percentile. That places the two products at opposite ends of the performance spectrum. The Arc B370's average benchmark score is 1184, which matches its single Steel Nomad result. The RTX 6000D's average benchmark score is 195964, a figure heavily influenced by its Geekbench OpenCL result of 388405 alongside its Steel Nomad score of 3522.

The RTX 6000D's nearest rivals in the database are all professional or data-center class accelerators. It sits 0.8% ahead of the NVIDIA Tesla V100S PCIe 32 GB (194415), 4.7% ahead of the NVIDIA A100 SXM4 40 GB (187147), and 6.1% ahead of the NVIDIA RTX 5000 Ada Generation (184664). The only rival that scores higher is the NVIDIA A100 PCIe 80 GB at 207124, which puts the RTX 6000D 5.4% behind that part. These comparisons show that the RTX 6000D is positioned among the highest-performing accelerators in the database, not merely among consumer graphics cards.

The Intel Arc B370's nearest rivals are all older, low-end parts. It sits 0.2% behind the ATI Mobility Radeon HD 5570 (1186), 0.5% behind the ATI Radeon HD 5770 (1190), and 0.7% behind the AMD Radeon HD 7650M (1192). It beats the AMD FirePro M2000 (1168) by 1.4%. These delta values are small, indicating that the Arc B370 performs at a level comparable to graphics hardware from roughly a decade older. The 5th percentile ranking confirms that this is one of the lowest-performing GPUs in the database.

Architecture Differences

The two products use fundamentally different silicon. The Intel Arc B370 is built on the Panther Lake chip using the Xe3-LPG architecture, belonging to the Arc Graphics-M (Panther Lake) generation. The process node is 3 nm with Intel as the foundry. The NVIDIA RTX 6000D uses the GB202 chip with Blackwell 2.0 architecture, belonging to the Blackwell PRO W (x000) generation. Its process node is 5 nm with TSMC as the foundry. The transistor counts diverge sharply: the RTX 6000D has 92,200 million transistors on a 750 mm² die with a transistor density of 122.9M per mm², while the Arc B370's transistor count and die size are listed as unknown in the database.

The compute resources are vastly different. The Intel Arc B370 has 1280 shading units, 40 texture mapping units, 20 ROPs, and 10 ray tracing cores. The NVIDIA RTX 6000D has 19968 shading units, 624 texture mapping units, 192 ROPs, and 156 ray tracing cores. The RTX 6000D also includes 624 tensor cores, while the Arc B370 has no tensor core count listed in the database. The NVIDIA part's shading unit count is roughly 15.6 times higher than Intel's, and its RT core count is 15.6 times higher as well.

Clock behavior differs as well. The Arc B370 runs at a base clock of 300 MHz and boosts to 2400 MHz. The RTX 6000D runs at a base clock of 1992 MHz and boosts to 2430 MHz. The NVIDIA part's base clock is significantly higher, though the boost clocks are closer. The memory subsystems are entirely different. The Arc B370 uses system-shared memory, meaning its memory size, type, bus width, and bandwidth are all dependent on the host system. The RTX 6000D has dedicated 84 GB of GDDR7 memory on a 448-bit bus with 1.40 TB/s bandwidth and a memory clock of 1560 MHz (25 Gbps effective).

Rates and throughput figures confirm the performance gap. The Arc B370 delivers a pixel rate of 48.00 GPixel/s, a texture rate of 96.00 GTexel/s, FP32 performance of 6.144 TFLOPS, and FP16 performance of 12.29 TFLOPS (2:1). The RTX 6000D delivers a pixel rate of 466.6 GPixel/s, a texture rate of 1,516.3 GTexel/s, FP32 performance of 97.04 TFLOPS, and FP16 performance of 97.04 TFLOPS (1:1). The NVIDIA part's FP32 throughput is roughly 15.8 times higher, matching the shading unit ratio.

Power and physical characteristics also differ. The Arc B370 has a TDP of 25 W, uses no power connectors, and is an integrated graphics processor (IGP) with no slot width. The RTX 6000D has a TDP of 600 W, uses a single 16-pin power connector, requires a 1000 W suggested power supply, and occupies a dual-slot form factor. The RTX 6000D measures 304 mm in length, 137 mm in height, and 40 mm in width. The Arc B370's dimensions are not listed. The bus interface for the Arc B370 is IGP, while the RTX 6000D uses PCIe 5.0 x16. Display outputs for the Arc B370 are listed as portable device dependent, while the RTX 6000D provides 4x DisplayPort 2.1b.

Both parts support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The Arc B370's production status is Active with a release date of 2026-01-26. The RTX 6000D's production status is Active with a release date of 2025-07-13. The RTX 6000D's predecessor is listed as Workstation Ada. The Arc B370 has no predecessor listed.

Where Each One Wins

The data records a single benchmark category, 3DMark Steel Nomad DX12, and the NVIDIA RTX 6000D wins it outright. There are no benchmark categories in which the Intel Arc B370 records a win. The RTX 6000D also holds the Geekbench OpenCL result of 388405, a test that the Arc B370 has no recorded entry for. This means the database shows the NVIDIA part winning both recorded benchmark tests, while the Intel part has only one recorded benchmark result.

The use cases follow directly from the hardware specifications. The RTX 6000D is built for workloads that demand massive parallel compute. Its 19968 shading units, 624 tensor cores, and 97.04 TFLOPS FP32 throughput position it for professional rendering, AI inference, and compute-heavy tasks. The 84 GB GDDR7 memory with 1.40 TB/s bandwidth supports large datasets that would exceed the memory capacity of most GPUs. The 4x DisplayPort 2.1b outputs and PCIe 5.0 x16 interface suit a workstation environment with multiple high-resolution displays.

The Arc B370 is an integrated GPU with 25 W TDP and system-shared memory. Its 1280 shading units and 6.144 TFLOPS FP32 performance place it at the low end of the database. The 3 nm process node and 300 MHz base clock indicate a power-efficient design intended for portable devices, as the display outputs are listed as portable device dependent. The 5th percentile ranking among all GPUs confirms that it is not designed for demanding graphics workloads. Its nearest rivals, all with scores within 1.4% of the Arc B370, are from the ATI and AMD HD 5000/7000 series, which were entry-level discrete parts from an older generation.

The RTX 6000D's nearest rivals reinforce its positioning. The Tesla V100S, A100 variants, and RTX 5000 Ada Generation are all accelerators designed for data centers or professional workstations. The RTX 6000D sits within a few percent of those parts in average benchmark score. The 98th percentile ranking places it among the top 2% of all GPUs in the database. This is not a consumer gaming card by the numbers; it is a professional compute device.

The Arc B370's role is the opposite. It sits at the 5th percentile, among the bottom 5% of all GPUs. Its performance is comparable to mobile and entry-level desktop parts from over a decade ago, based on the rival scores. The system-shared memory means its effective bandwidth depends entirely on the host platform. The 2:1 FP16 ratio on the Arc B370 indicates that half-precision throughput is achieved by pairing FP32 units, whereas the RTX 6000D's 1:1 FP16 ratio means it has dedicated half-precision throughput equal to its FP32 rate.

The Verdict

The benchmark data is unambiguous. The NVIDIA RTX 6000D is the higher-performing product by a wide margin, winning the only head-to-head test with a 66.4% delta in its favor. The RTX 6000D's 98th percentile ranking and its position among Tesla and A100-class accelerators in the nearest rivals list confirm that it belongs in the highest tier of GPU performance. The Intel Arc B370's 5th percentile ranking and its position among ATI HD 5000-series parts confirm that it belongs in the lowest tier.

The RTX 6000D is the appropriate choice for workloads that require maximum compute throughput, large memory capacity, and professional-grade output support. Its 84 GB of GDDR7 memory, 97.04 TFLOPS FP32, and 624 tensor cores provide the resources needed for large-scale rendering, AI training, or scientific computing. The 600 W TDP and 1000 W suggested power supply indicate a system designed for sustained heavy load. The launch MSRP is 8,565 USD.

The Arc B370 is an integrated solution for portable devices, with a 25 W TDP and no external power connectors. Its 1280 shading units and 6.144 TFLOPS FP32 are sufficient for basic graphics output but not for demanding compute. The 3 nm process and 300 MHz base clock prioritize efficiency over performance. The system-shared memory architecture means there is no dedicated VRAM allocation; the memory size, type, bus width, and bandwidth are all system dependent. This makes the Arc B370 suitable for lightweight mobile workloads where discrete graphics are not available.

The choice between these two products depends entirely on the intended use case. A system requiring the RTX 6000D's level of compute, memory, and output capabilities has no alternative in this comparison. A system requiring the Arc B370's power efficiency and integrated form factor similarly has no alternative in this comparison. The data does not support any scenario where the Arc B370 outperforms the RTX 6000D in a recorded benchmark.

FAQ

Q: Which GPU wins the 3DMark Steel Nomad DX12 benchmark?

A: The NVIDIA RTX 6000D wins with a score of 3522, while the Intel Arc B370 scores 1184. The delta is -66.4% for the Intel part.

Q: What is the percentile ranking of each GPU?

A: The Intel Arc B370 is at the 5th percentile of all GPUs in the database. The NVIDIA RTX 6000D is at the 98th percentile.

Q: How much memory does each GPU have?

A: The Intel Arc B370 uses system-shared memory, so its memory size, type, bus width, and bandwidth are system dependent. The NVIDIA RTX 6000D has 84 GB of GDDR7 memory on a 448-bit bus with 1.40 TB/s bandwidth.

Q: What are the FP32 performance figures?

A: The Intel Arc B370 delivers 6.144 TFLOPS FP32. The NVIDIA RTX 6000D delivers 97.04 TFLOPS FP32.

Q: What process nodes are used?

A: The Intel Arc B370 uses a 3 nm process with Intel as the foundry. The NVIDIA RTX 6000D uses a 5 nm process with TSMC as the foundry.

Q: What are the TDP figures?

A: The Intel Arc B370 has a TDP of 25 W. The NVIDIA RTX 6000D has a TDP of 600 W and requires a 1000 W suggested power supply.

Specification Differences

| Specification | Intel Arc B370 | NVIDIA RTX 6000D |

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

| Chip | Panther Lake | GB202 |

| Architecture | Xe3-LPG | Blackwell 2.0 |

| Generation | Arc Graphics-M (Panther Lake) | Blackwell PRO W (x000) |

| Process Node | 3 nm | 5 nm |

| Foundry | Intel | TSMC |

| Transistors | unknown | 92,200 million |

| Die Size | unknown | 750 mm² |

| Transistor Density | null | 122.9M / mm² |

| Base Clock | 300 MHz | 1992 MHz |

| Boost Clock | 2400 MHz | 2430 MHz |

| Memory Size | System Shared | 84 GB |

| Memory Type | System Shared | GDDR7 |

| Memory Bus Width | System Shared | 448 bit |

| Memory Bandwidth | System Dependent | 1.40 TB/s |

| Shading Units | 1280 | 19968 |

| TMUs | 40 | 624 |

| ROPs | 20 | 192 |

| RT Cores | 10 | 156 |

| Tensor Cores | null | 624 |

| Pixel Rate | 48.00 GPixel/s | 466.6 GPixel/s |

| Texture Rate | 96.00 GTexel/s | 1,516.3 GTexel/s |

| FP32 | 6.144 TFLOPS | 97.04 TFLOPS |

| FP16 | 12.29 TFLOPS (2:1) | 97.04 TFLOPS (1:1) |

| TDP | 25 W | 600 W |

| Slot Width | IGP | Dual-slot |

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

| Suggested PSU | null | 1000 W |

| Bus Interface | IGP | PCIe 5.0 x16 |

| Display Outputs | Portable Device Dependent | 4x DisplayPort 2.1b |

| Length | null | 304 mm (12 inches) |

| Height | null | 137 mm (5.4 inches) |

| Width | null | 40 mm (1.6 inches) |

| Release Date | 2026-01-26 | 2025-07-13 |

| Predecessor | null | Workstation Ada |

| Launch MSRP | null | 8,565 USD |