Intel Arc B370 vs NVIDIA GeForce RTX 4080 SUPER Comparison

Intel Arc B370 is an integrated graphics processor based on Intel’s Panther Lake chip, built on the Xe3-LPG architecture at a 3 nm process node. NVIDIA GeForce RTX 4080 SUPER is a discrete add-in board from the GeForce 40-series, using the AD103 chip on TSMC’s 5 nm process with the Ada Lovelace architecture. The two products occupy opposite ends of the graphics spectrum, and the recorded benchmark data reflects that gap clearly.

Head-to-Head Benchmarks

The database contains one directly comparable test for these two products: 3DMark Steel Nomad under DirectX 12. In that benchmark, the Intel Arc B370 scores 1184 points, while the NVIDIA GeForce RTX 4080 SUPER scores 6600 points. The Intel part trails by 82.1 percent, a margin that places the two in entirely different performance classes.

To contextualize the Arc B370’s score, its nearest rivals in the database are all older or low-power parts. The ATI Mobility Radeon HD 5570 averages 1186 points, just 0.2 percent ahead of the Arc B370. The ATI Radeon HD 5770 sits at 1190 points, 0.5 percent ahead. The AMD Radeon HD 7650M averages 1192 points, 0.7 percent ahead. The AMD FirePro M2000 averages 1168 points, which is 1.4 percent behind the Arc B370. These deltas are small, indicating that the Arc B370’s raw performance in this test lands near a cluster of decade-old desktop and mobile parts rather than any modern discrete solution.

The RTX 4080 SUPER, by contrast, sits in the 86th percentile of all GPUs in the database. Its nearest rivals include the NVIDIA GeForce RTX 4080, which averages 54247 points, a 0.1 percent advantage over the RTX 4080 SUPER’s 54209 average. The AMD Radeon Pro W5700X averages 54828 points, 1.1 percent ahead. The AMD Radeon RX 6750 GRE 12 GB averages 55698 points, 2.7 percent ahead. The AMD Radeon 8060S averages 55757 points, 2.8 percent ahead. The RTX 4080 SUPER is therefore surrounded by high-end parts, with each rival beating it by single-digit percentage points in aggregate average scores.

The single head-to-head result shows a delta of 82.1 percent in favor of the RTX 4080 SUPER. That is not a small gap; it is a dominant result. The Arc B370’s 1184 points is closer to the FirePro M2000’s 1168 than to any modern discrete GPU, while the RTX 4080 SUPER’s 6600 points in Steel Nomad represents a score roughly five and a half times higher. In practical terms, the data shows the RTX 4080 SUPER delivering a level of DirectX 12 rasterization performance that the Arc B370 cannot approach.

The RTX 4080 SUPER also has additional benchmark entries in the database beyond the shared Steel Nomad test. Its Geekbench OpenCL score is 219065, its Geekbench Vulkan score is 260075, and its PassMark G3D score is 34245. The Arc B370 has no entries for those tests, so no direct comparison is possible there. The available overlap is limited to the single 3DMark test, and that test alone establishes the hierarchy.

Architecture Differences

The two processors are built on different manufacturing processes and different foundries. The Intel Arc B370 uses a 3 nm node at Intel’s own foundry. The NVIDIA GeForce RTX 4080 SUPER uses a 5 nm process at TSMC. Intel has not disclosed transistor count or die size for the Arc B370, so those fields are unknown. The RTX 4080 SUPER, by contrast, has fully disclosed specifications: 45,900 million transistors on a 379 mm² die, for a transistor density of 121.1 million transistors per square millimeter.

The core configurations diverge sharply. The Arc B370 has 1280 shading units, 40 texture mapping units, 20 raster output units, and 10 ray tracing cores. The RTX 4080 SUPER has 10240 shading units, 320 texture mapping units, 112 raster output units, and 80 ray tracing cores. The NVIDIA part also includes 320 tensor cores, while the Intel part reports no tensor core count. In every compute-related unit, the RTX 4080 SUPER has eight times the shading units, eight times the TMUs, 5.6 times the ROPs, and eight times the RT cores.

Clock behavior differs as well. The Arc B370’s base clock is 300 MHz with a boost of 2400 MHz. The RTX 4080 SUPER’s base clock is 2295 MHz with a boost of 2550 MHz. Despite starting far lower, the Intel part boosts to a clock only 150 MHz below the NVIDIA part’s boost. The massive core count difference means the RTX 4080 SUPER still produces far higher throughput. The Arc B370’s FP32 rate is 6.144 TFLOPS, while the RTX 4080 SUPER’s FP32 rate is 52.22 TFLOPS, a factor of roughly 8.5. The texture rate tells a similar story: 96.00 GTexel/s for the Arc B370 versus 816.0 GTexel/s for the RTX 4080 SUPER. Pixel rate is 48.00 GPixel/s versus 285.6 GPixel/s.

Memory architecture is fundamentally different. The Arc B370 uses system shared memory, with system-dependent bandwidth and no dedicated VRAM. The RTX 4080 SUPER has 16 GB of GDDR6X on a 256-bit bus, delivering 736.3 GB/s of bandwidth. The Intel part’s memory clock is listed as system shared, while the NVIDIA part runs at 1438 MHz with 23 Gbps effective speed. Any comparison of memory capacity or bandwidth between a shared-memory IGP and a discrete GDDR6X card is therefore not apples-to-apples; the RTX 4080 SUPER has a fixed, dedicated pool, while the Arc B370 depends on the host system.

FP16 throughput also differs in ratio. The Arc B370 achieves 12.29 TFLOPS with a 2:1 FP16 to FP32 ratio. The RTX 4080 SUPER achieves 52.22 TFLOPS with a 1:1 ratio. The NVIDIA part does not gain extra half-precision throughput relative to its FP32 rate, but its absolute FP16 output is still more than four times the Intel part’s.

Power and physical design are extremely different. The Arc B370 has a 25 W TDP and is an integrated graphics processor, meaning it occupies the IGP slot width, uses no power connectors, and has no suggested PSU requirement. Its display outputs are marked portable device dependent. The RTX 4080 SUPER has a 320 W TDP, uses a triple-slot cooler, requires a single 16-pin power connector, and lists a 700 W suggested PSU. Its physical dimensions are 310 mm in length, 140 mm in height, and 61 mm in width. The Intel part has no listed dimensions, consistent with an IGP that has no standalone card.

Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The bus interfaces differ: the Arc B370 connects through the IGP interface, while the RTX 4080 SUPER uses PCIe 4.0 x16. The RTX 4080 SUPER’s display outputs are 1x HDMI 2.1 and 3x DisplayPort 1.4a. The Intel part’s outputs are dependent on the portable device it is embedded in.

Production status also separates the two. The Arc B370 is listed as active, with a release date of January 26, 2026. The RTX 4080 SUPER is listed as end-of-life, with a release date of January 30, 2024. The NVIDIA part’s predecessor is GeForce 30 and its successor is GeForce 50. The Intel part has no predecessor or successor listed. The RTX 4080 SUPER has a launch MSRP of 999 USD; the Arc B370 has no launch MSRP field.

FAQ

Q: Which GPU wins the shared 3DMark Steel Nomad test?

A: The NVIDIA GeForce RTX 4080 SUPER wins, scoring 6600 points against the Intel Arc B370’s 1184 points, a gap of 82.1 percent in NVIDIA’s favor.

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

A: The Arc B370’s nearest rivals are all older or lower-tier parts. The ATI Mobility Radeon HD 5570 is 0.2 percent ahead, the ATI Radeon HD 5770 is 0.5 percent ahead, the AMD Radeon HD 7650M is 0.7 percent ahead, and the AMD FirePro M2000 is 1.4 percent behind.

Q: What are the key architectural differences between the two?

A: The Arc B370 uses Intel’s Xe3-LPG architecture on a 3 nm Intel process, while the RTX 4080 SUPER uses Ada Lovelace on TSMC’s 5 nm process. The NVIDIA part has 10240 shading units, 320 TMUs, 112 ROPs, 80 RT cores, and 320 tensor cores, versus 1280 shading units, 40 TMUs, 20 ROPs, and 10 RT cores for the Intel part.

Q: How do the memory systems differ?

A: The Arc B370 uses system shared memory with system-dependent bandwidth and no dedicated VRAM. The RTX 4080 SUPER has 16 GB of GDDR6X on a 256-bit bus with 736.3 GB/s of bandwidth.

Q: What is the power requirement difference?

A: The Arc B370 has a 25 W TDP and uses no power connectors, while the RTX 4080 SUPER has a 320 W TDP, requires a single 16-pin power connector, and lists a 700 W suggested PSU.

Q: Which product is still in production?

A: The Intel Arc B370 is listed as active with a release date of January 26, 2026. The NVIDIA GeForce RTX 4080 SUPER is listed as end-of-life with a release date of January 30, 2024.

The Verdict

The recorded data points to a straightforward conclusion. The Intel Arc B370 delivers performance that places it in the 5th percentile of all GPUs in the database, with an average benchmark score of 1184. Its nearest competitors are parts like the ATI Radeon HD 5770 and AMD Radeon HD 7650M, all within roughly one percentage point in either direction. That places the Arc B370 firmly in the low-power integrated graphics class. It is designed for portable devices, uses system shared memory, draws only 25 W, and requires no power connectors. Its 3 nm process and Xe3-LPG architecture are modern, but the core configuration is small: 1280 shading units, 40 TMUs, 20 ROPs, and 10 RT cores. For a user relying solely on integrated graphics in a thin portable device, the Arc B370 represents a baseline DirectX 12 Ultimate capable solution, but its 6.144 TFLOPS FP32 throughput and 1184 Steel Nomad score indicate modest rasterization performance.

The RTX 4080 SUPER sits at the 86th percentile of all GPUs, with an average benchmark score of 54209. Its nearest rivals are the RTX 4080, Radeon Pro W5700X, Radeon RX 6750 GRE 12 GB, and Radeon 8060S, all within 0.1 to 2.8 percent of its average. In the shared Steel Nomad test, it beats the Arc B370 by 82.1 percent. It has 10240 shading units, 320 TMUs, 112 ROPs, 80 RT cores, and 320 tensor cores, along with 16 GB of GDDR6X memory and 736.3 GB/s of bandwidth. Its 52.22 TFLOPS FP32 rate is roughly 8.5 times the Intel part’s. The RTX 4080 SUPER is a high-end discrete card designed for demanding workloads, and the data reflects that positioning.

There is no ambiguity in the head-to-head result. The Intel Arc B370 is not a competitor to the RTX 4080 SUPER in raw performance. The 82.1 percent deficit in Steel Nomad is decisive. The RTX 4080 SUPER’s additional benchmarks, including 219065 in Geekbench OpenCL, 260075 in Geekbench Vulkan, and 34245 in PassMark G3D, further demonstrate its compute capabilities, though the Arc B370 has no corresponding entries for comparison. The Arc B370’s role is as a low-power integrated solution for portable devices, while the RTX 4080 SUPER is a high-power discrete card for desktop systems. Any user choosing between them would be selecting between entirely different product categories. The data does not support any scenario where the Arc B370 outperforms the RTX 4080 SUPER, and the RTX 4080 SUPER’s power draw, size, and end-of-life status make it unsuitable for the portable segment the Arc B370 targets. The verdict is therefore determined by use case: the Arc B370 for integrated, low-power portable graphics, and the RTX 4080 SUPER for high-end discrete performance.

Specification Differences

The following fields differ between the Intel Arc B370 and the NVIDIA GeForce RTX 4080 SUPER:

• Chip: Intel Arc B370 uses Panther Lake; RTX 4080 SUPER uses AD103.
• Architecture: Intel uses Xe3-LPG; NVIDIA uses Ada Lovelace.
• Generation: Intel is Arc Graphics-M (Panther Lake); NVIDIA is GeForce 40.
• Process node: Intel uses 3 nm; NVIDIA uses 5 nm.
• Foundry: Intel uses Intel; NVIDIA uses TSMC.
• Transistors: Intel is unknown; NVIDIA is 45,900 million.
• Die size: Intel is unknown; NVIDIA is 379 mm².
• Transistor density: Intel is not listed; NVIDIA is 121.1M / mm².
• Base clock: Intel is 300 MHz; NVIDIA is 2295 MHz.
• Boost clock: Intel is 2400 MHz; NVIDIA is 2550 MHz.
• Memory clock: Intel is system shared; NVIDIA is 1438 MHz 23 Gbps effective.
• Memory size: Intel is system shared; NVIDIA is 16 GB.
• Memory type: Intel is system shared; NVIDIA is GDDR6X.
• Memory bus width: Intel is system shared; NVIDIA is 256 bit.
• Memory bandwidth: Intel is system dependent; NVIDIA is 736.3 GB/s.
• Shading units: Intel has 1280; NVIDIA has 10240.
• TMUs: Intel has 40; NVIDIA has 320.
• ROPs: Intel has 20; NVIDIA has 112.
• RT cores: Intel has 10; NVIDIA has 80.
• Tensor cores: Intel has none listed; NVIDIA has 320.
• Pixel rate: Intel is 48.00 GPixel/s; NVIDIA is 285.6 GPixel/s.
• Texture rate: Intel is 96.00 GTexel/s; NVIDIA is 816.0 GTexel/s.
• FP32: Intel is 6.144 TFLOPS; NVIDIA is 52.22 TFLOPS.
• FP16: Intel is 12.29 TFLOPS (2:1); NVIDIA is 52.22 TFLOPS (1:1).
• TDP: Intel is 25 W; NVIDIA is 320 W.
• Slot width: Intel is IGP; NVIDIA is triple-slot.
• Power connectors: Intel has none; NVIDIA has 1x 16-pin.
• Suggested PSU: Intel has none listed; NVIDIA is 700 W.
• Bus interface: Intel is IGP; NVIDIA is PCIe 4.0 x16.
• Display outputs: Intel is portable device dependent; NVIDIA is 1x HDMI 2.1 and 3x DisplayPort 1.4a.
• Dimensions: Intel has none listed; NVIDIA is 310 mm by 140 mm by 61 mm.
• Production status: Intel is active; NVIDIA is end-of-life.
• Release date: Intel is January 26, 2026; NVIDIA is January 30, 2024.
• Predecessor: Intel has none listed; NVIDIA is GeForce 30.
• Successor: Intel has none listed; NVIDIA is GeForce 50.
• Launch MSRP: Intel has none listed; NVIDIA is 999 USD.