Intel Arc A380 vs NVIDIA GeForce GTX 870M Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 870M has an average benchmark score of 9959, which places it in the 48th percentile of all GPUs. The Intel Arc A380 has an average score of 8558, placing it in the 44th percentile.

Q: How do the two GPUs compare in the only head-to-head benchmark test?

A: In the Geekbench OpenCL test, the Intel Arc A380 scored 38224, while the NVIDIA GeForce GTX 870M scored 12630. The Intel card won this test with a delta percentage of -67 (meaning the NVIDIA result is 67% lower).

Q: What are the nearest rivals for each card according to the database?

A: For the GTX 870M, the closest rival is the AMD Radeon Pro 5300M with an average score of 10013 (delta -0.5%). For the Arc A380, the closest rival is the AMD FirePro W5170M with an average score of 8595 (delta -0.4%).

Q: What is the difference in transistor density between the two chips?

A: The Intel Arc A380 uses a 6 nm process at TSMC and has a transistor density of 45.9M per mm². The NVIDIA GTX 870M uses a 28 nm process and has a transistor density of only 12.0M per mm².

Q: Which GPU has more shading units?

A: The NVIDIA GeForce GTX 870M has 1344 shading units, while the Intel Arc A380 has 1024 shading units. However, the Intel card has a higher FP32 compute throughput at 4.198 TFLOPS versus 2.599 TFLOPS for the NVIDIA card.

Q: What is the difference in memory capacity and type?

A: The NVIDIA GTX 870M comes with 3 GB of GDDR5 memory on a 192-bit bus, delivering 120.0 GB/s of bandwidth. The Intel Arc A380 has 6 GB of GDDR6 memory on a 96-bit bus, delivering 186.0 GB/s of bandwidth.

Architecture Differences

The two GPUs come from fundamentally different architectural eras and design philosophies. The NVIDIA GeForce GTX 870M is built on the Kepler architecture, specifically the GK104 chip, manufactured on a 28 nm process at TSMC. This chip contains 3,540 million transistors on a die size of 294 mm², resulting in a transistor density of 12.0M per mm². It belongs to the GeForce 800M generation, which was released in 2014, and its production status is marked as end-of-life.

The Intel Arc A380 is built on the Xe-HPG architecture, specifically the DG2-128 chip, manufactured on a 6 nm process at TSMC. This chip packs 7,200 million transistors into a die size of just 157 mm², achieving a transistor density of 45.9M per mm². It belongs to the Alchemist (Arc 3) generation and was released in 2022, with its production status also marked as end-of-life. The difference in process node and transistor density is stark: the Intel chip crams more than twice the transistors into roughly half the die area.

The memory subsystems also differ significantly. The GTX 870M uses 3 GB of GDDR5 memory clocked at 1250 MHz (5 Gbps effective) across a 192-bit bus, yielding 120.0 GB/s of bandwidth. The Arc A380 uses 6 GB of GDDR6 memory clocked at 1937 MHz (15.5 Gbps effective) across a 96-bit bus, yielding 186.0 GB/s of bandwidth. Despite having a narrower bus, the Intel card achieves 55% more memory bandwidth thanks to the faster GDDR6 standard.

In terms of compute resources, the GTX 870M has 1344 shading units, 112 texture mapping units, and 24 ROPs. The Arc A380 has 1024 shading units, 64 TMUs, and 32 ROPs. Notably, the Intel card includes 8 ray tracing cores, a feature entirely absent from the older NVIDIA Kepler design. The GTX 870M also lacks tensor cores, as does the Intel card.

The API support tells a story of generational progress. The GTX 870M supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The Arc A380 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The Intel card also supports FP16 compute at 8.397 TFLOPS (2:1 ratio), while the NVIDIA card has no FP16 capability listed.

Physical form factors differ dramatically. The GTX 870M is an MXM Module with no power connectors and a 100 W TDP. The Arc A380 is a dual-slot card measuring 222 mm in length, 114 mm in height, and 42 mm in width, with a single 8-pin power connector and a 75 W TDP. The Intel card uses a PCIe 4.0 x8 bus interface, while the NVIDIA card uses MXM-B (3.0). The Arc A380 offers fixed display outputs (1x HDMI 2.1 and 3x DisplayPort 2.0), whereas the GTX 870M's outputs are portable device dependent.

Head-to-Head Benchmarks

The recorded data shows only one direct head-to-head benchmark between these two GPUs: Geekbench OpenCL. In this test, the Intel Arc A380 scored 38224, while the NVIDIA GeForce GTX 870M scored 12630. The Intel card won decisively with a delta percentage of -67, meaning the NVIDIA result is 67% lower than the Intel score. This is a substantial margin, indicating that in raw compute workloads, the Arc A380 holds a clear advantage.

To contextualize this result, consider the average benchmark scores from the broader database. The GTX 870M's average score is 9959, while the Arc A380's average is 8558. This creates an interesting discrepancy: the Intel card wins the OpenCL test by a wide margin, yet its overall average benchmark score is lower. The explanation lies in the benchmark suite composition. The Arc A380 has multiple recorded scores, including a low PassMark DirectX 10 score of 37, a DirectX 11 score of 38, and a DirectX 12 score of 35. These low scores pull down its average. The GTX 870M has only two recorded benchmarks: Geekbench Metal (7288) and Geekbench OpenCL (12630).

Looking at the nearest rivals reinforces the picture. The GTX 870M's closest competitor is the AMD Radeon Pro 5300M with an average score of 10013, which is only 0.5% higher. The Intel Arc A380's closest rival is the AMD FirePro W5170M with an average score of 8595, which is 0.4% higher. In both cases, the rivals are extremely close, suggesting these cards sit in a tightly contested performance tier, even though the silicon generations are eight years apart.

The Arc A380 also shows a strong Vulkan result of 36736 in Geekbench Vulkan, though this test was not run on the GTX 870M. Its PassMark G3D score is 6252, and its GPU compute score is 2762. These additional data points indicate that the Intel card performs better in modern API workloads, while the older NVIDIA card has no recorded results for these tests.

Specification Differences

The following specifications differ between the two cards:

• Chip: GK104 (NVIDIA) versus DG2-128 (Intel)
• Architecture: Kepler versus Xe-HPG
• Generation: GeForce 800M versus Alchemist (Arc 3)
• Process Node: 28 nm versus 6 nm
• Transistors: 3,540 million versus 7,200 million
• Die Size: 294 mm² versus 157 mm²
• Transistor Density: 12.0M / mm² versus 45.9M / mm²
• Base Clock: 941 MHz versus 2000 MHz
• Boost Clock: 967 MHz versus 2050 MHz
• Memory Clock: 1250 MHz (5 Gbps effective) versus 1937 MHz (15.5 Gbps effective)
• Memory Size: 3 GB versus 6 GB
• Memory Type: GDDR5 versus GDDR6
• Memory Bus Width: 192 bit versus 96 bit
• Memory Bandwidth: 120.0 GB/s versus 186.0 GB/s
• Shading Units: 1344 versus 1024
• TMUs: 112 versus 64
• ROPs: 24 versus 32
• RT Cores: None versus 8
• Pixel Rate: 27.08 GPixel/s versus 65.60 GPixel/s
• Texture Rate: 108.3 GTexel/s versus 131.2 GTexel/s
• FP32: 2.599 TFLOPS versus 4.198 TFLOPS
• FP16: Not available versus 8.397 TFLOPS (2:1)
• TDP: 100 W versus 75 W
• Slot Width: MXM Module versus Dual-slot
• Power Connectors: None versus 1x 8-pin
• Suggested PSU: Not listed versus 250 W
• Bus Interface: MXM-B (3.0) versus PCIe 4.0 x8
• Display Outputs: Portable Device Dependent versus 1x HDMI 2.1, 3x DisplayPort 2.0
• DirectX Support: 12 (11_0) versus 12 Ultimate (12_2)
• Vulkan Support: 1.2.175 versus 1.4
• Release Date: 2014-03-11 versus 2022-06-13
• Predecessor: GeForce 700M versus Xe Graphics
• Successor: GeForce 900M versus Battlemage
• Dimensions: Not listed versus 222 mm x 114 mm x 42 mm
• Launch MSRP: Not listed versus 149 USD

The Verdict

The data paints a clear picture of generational progression, but with a twist in the average scores. The Intel Arc A380 wins the only direct head-to-head benchmark by a 67% margin in Geekbench OpenCL. It also offers more than double the memory capacity (6 GB versus 3 GB), higher memory bandwidth (186.0 GB/s versus 120.0 GB/s), and substantially higher compute throughput (4.198 TFLOPS versus 2.599 TFLOPS). The Intel card also brings modern features like ray tracing cores, DirectX 12 Ultimate support, and Vulkan 1.4.

However, the NVIDIA GeForce GTX 870M holds a higher average benchmark score (9959 versus 8558) and a higher percentile ranking (48th versus 44th). Its nearest rivals are all within 1.1% of its average score, indicating it sits in a well-populated performance tier. The GTX 870M also has more shading units (1344 versus 1024) and more TMUs (112 versus 64), though these advantages do not translate into higher raw throughput.

The launch MSRP for the Arc A380 is 149 USD, while the GTX 870M has no recorded launch MSRP. The data suggests that the Intel card is the more capable performer in compute-heavy tasks, while the NVIDIA card's average score reflects its stronger showing in the limited set of benchmarks recorded for it.

Where Each One Wins

Intel Arc A380 wins in:

• Geekbench OpenCL performance, where it scores 38224 versus 12630, a 67% advantage
• Memory capacity, with 6 GB versus 3 GB
• Memory bandwidth, with 186.0 GB/s versus 120.0 GB/s
• FP32 compute, with 4.198 TFLOPS versus 2.599 TFLOPS
• Pixel rate, with 65.60 GPixel/s versus 27.08 GPixel/s
• Texture rate, with 131.2 GTexel/s versus 108.3 GTexel/s
• Ray tracing, with 8 dedicated RT cores versus none
• API support, with DirectX 12 Ultimate and Vulkan 1.4
• Power efficiency, with a 75 W TDP versus 100 W
• Modern form factor, with PCIe 4.0 x8 and fixed display outputs

NVIDIA GeForce GTX 870M wins in:

• Average benchmark score, with 9959 versus 8558
• Percentile ranking, 48th versus 44th
• Shading unit count, 1344 versus 1024
• TMU count, 112 versus 64
• Memory bus width, 192-bit versus 96-bit
• MXM form factor, which suits portable device integration

For users prioritizing modern API support, compute throughput, and memory bandwidth, the Intel Arc A380 is the clear choice based on the recorded data. For users who value the higher average benchmark score and the mobile MXM form factor, the NVIDIA GeForce GTX 870M holds an edge in the database's aggregated metrics. The choice ultimately depends on whether the workload favors the Intel card's raw compute wins or the NVIDIA card's more consistent average performance profile.