GPU Comparison

The Intel Arc A380 and NVIDIA GeForce GTX 980 represent two very different generations of GPU technology that happen to land at nearly the same overall performance level. The data shows the Arc A380 wins the modern API tests decisively, while the GTX 980 dominates legacy DirectX workloads and raw rasterization. The A380 takes 3 of 10 head-to-head benchmarks, while the GTX 980 wins 7. However, the Arc A380 achieves this with a 75 W TDP versus 165 W for the GTX 980, and the A380's average benchmark score of 8558 edges out the GTX 980's 8167, placing them at the 44th and 43rd percentiles of all GPUs respectively. This is a classic case of architectural age versus modern feature support.

The Verdict

For users running modern DirectX 12 Ultimate titles, the Intel Arc A380 is the clear choice, as it leads the GTX 980 by 70.5% in 3DMark Steel Nomad DX12 and by 63% in Geekbench Vulkan. The A380 also offers hardware ray tracing, which the GTX 980 lacks entirely. However, for anyone with a library of older games or applications that rely on DirectX 9, 10, or 11, the GTX 980 is dramatically faster, winning those tests by margins ranging from 30.2% to 55.5%. The GTX 980 also holds a 43.7% lead in Passmark G3D and a 41.9% lead in Passmark GPU Compute, indicating substantially higher raw shader throughput for traditional workloads. The A380's 6 GB of GDDR6 memory provides more capacity than the GTX 980's 4 GB, which can matter for modern textures, but the GTX 980 counters with a wider 256-bit bus and higher 224.4 GB/s bandwidth versus 186.0 GB/s for the A380. Ultimately, the choice depends entirely on which API generation the user's software targets; the data shows no single card is comprehensively superior.

Architecture Differences

The Intel Arc A380 is built on the Xe-HPG architecture using the DG2-128 chip, fabricated on a 6 nm process at TSMC. This is a stark contrast to the GTX 980's Maxwell 2.0 architecture on the GM204 chip, which uses a 28 nm process. The A380 packs 7,200 million transistors into a 157 mm² die, yielding a transistor density of 45.9M per mm², while the GTX 980 has 5,200 million transistors on a much larger 398 mm² die, with a density of just 13.1M per mm². This process advantage explains why the A380 achieves its performance at 75 W versus the GTX 980's 165 W TDP.

The A380 features 1,024 shading units, 64 TMUs, and 32 ROPs, alongside 8 dedicated ray tracing cores. The GTX 980 has double the shading units at 2,048, double the TMUs at 128, and double the ROPs at 64, but has no ray tracing hardware. The A380 supports DirectX 12 Ultimate (12_2), while the GTX 980 is limited to DirectX 12 (12_1). Both cards support OpenGL 4.6 and Vulkan 1.4. The A380's memory configuration uses 6 GB of GDDR6 on a 96-bit bus, while the GTX 980 uses 4 GB of GDDR5 on a 256-bit bus. The A380 also features a PCIe 4.0 x8 interface, whereas the GTX 980 uses PCIe 3.0 x16.

Where Each One Wins

The Arc A380's wins are concentrated in modern, low-level APIs. Its 70.5% lead in 3DMark Steel Nomad DX12 demonstrates a massive advantage in contemporary DirectX 12 rendering. The 63% lead in Geekbench Vulkan shows similar strength in Vulkan workloads. The A380 also wins Geekbench OpenCL by 10.2%, indicating better general-purpose compute performance in that specific test. These results suggest the A380 is optimized for forward-looking game engines that leverage DirectX 12 Ultimate features and ray tracing.

The GTX 980 wins every legacy DirectX test. Its 55.5% lead in Passmark DirectX 9 is the largest margin, followed by 54.2% in DirectX 11, 30.2% in DirectX 10, and 23.9% in DirectX 12. The GTX 980 also leads in Passmark G2D by 23%, in Passmark G3D by 43.7%, and in Passmark GPU Compute by 41.9%. These results indicate that for traditional rasterization pipelines and older game engines, the GTX 980's higher shading unit count and wider memory bus provide significantly more throughput. The GTX 980's 77.82 GPixel/s pixel rate and 155.6 GTexel/s texture rate also exceed the A380's 65.60 GPixel/s and 131.2 GTexel/s.

FAQ

Q: Which card has better DirectX 12 performance?

A: The Intel Arc A380 wins 3DMark Steel Nomad DX12 with a score of 808 versus 474 for the GTX 980, a 70.5% advantage. However, in Passmark DirectX 12, the GTX 980 wins 46 to 35, a 23.9% lead. The A380 supports DirectX 12 Ultimate (12_2), while the GTX 980 supports DirectX 12 (12_1).

Q: Does the GTX 980 support ray tracing?

A: No. The GTX 980 has no ray tracing cores listed in the data. The Intel Arc A380 has 8 ray tracing cores, which is a key architectural advantage for modern titles.

Q: Which card has more memory bandwidth?

A: The GTX 980 has higher memory bandwidth at 224.4 GB/s, achieved with 4 GB of GDDR5 on a 256-bit bus. The Arc A380 has 186.0 GB/s from 6 GB of GDDR6 on a 96-bit bus.

Q: What are the power requirements for each card?

A: The Arc A380 has a 75 W TDP and requires a suggested 250 W power supply with a single 8-pin connector. The GTX 980 has a 165 W TDP and requires a suggested 450 W power supply with two 6-pin connectors.

Q: Which card has a higher overall average benchmark score?

A: The Intel Arc A380 has an average benchmark score of 8558, which is slightly higher than the GTX 980's 8167. The A380 sits at the 44th percentile of all GPUs, while the GTX 980 sits at the 43rd percentile.

Q: Which card is better for older DirectX 9 games?

A: The GTX 980 is overwhelmingly better, scoring 164 in Passmark DirectX 9 versus 73 for the Arc A380. This represents a 55.5% lead for the GTX 980.

Head-to-Head Benchmarks

The most striking result in the data is the 3DMark Steel Nomad DX12 test, where the Arc A380 scores 808 against the GTX 980's 474. This 70.5% margin is the largest win for either card and highlights the A380's modern architecture advantage in DirectX 12 Ultimate workloads. Similarly, in Geekbench Vulkan, the A380 posts 36,736 versus 22,543 for the GTX 980, a 63% lead. The A380 also wins Geekbench OpenCL with 38,224 points, a 10.2% advantage over the GTX 980's 34,676.

The GTX 980's counterarguments are equally decisive in legacy APIs. In Passmark DirectX 9, the GTX 980 scores 164 versus 73 for the A380, a 55.5% difference. Passmark DirectX 11 shows an 83 to 38 result, a 54.2% lead for the GTX 980. The GTX 980 also wins Passmark DirectX 10 at 53 versus 37, and Passmark DirectX 12 at 46 versus 35. In Passmark G3D, the GTX 980 more than doubles the A380's score with 11,095 versus 6,252, a 43.7% margin. Passmark GPU Compute follows a similar pattern, with the GTX 980 winning 4,753 to 2,762, a 41.9% lead. Even in Passmark G2D, the GTX 980 leads 792 to 610, a 23% difference. These results paint a clear picture: the A380 is specialized for future-facing APIs, while the GTX 980 is a brute-force rasterizer that excels in traditional and older workloads.

Specification Differences

The two cards differ fundamentally in process technology, with the Arc A380 using a 6 nm process and the GTX 980 using 28 nm. The A380 has 7,200 million transistors on a 157 mm² die, while the GTX 980 has 5,200 million on a 398 mm² die. Transistor density is 45.9M per mm² for the A380 versus 13.1M for the GTX 980. Clock speeds differ, with the A380 running at 2000 MHz base and 2050 MHz boost, while the GTX 980 runs at 1127 MHz base and 1216 MHz boost. Memory configurations differ: the A380 has 6 GB GDDR6 at 1937 MHz with 15.5 Gbps effective speed on a 96-bit bus, while the GTX 980 has 4 GB GDDR5 at 1753 MHz with 7 Gbps effective speed on a 256-bit bus. Bandwidth is 186.0 GB/s for the A380 versus 224.4 GB/s for the GTX 980. The A380 has 1,024 shading units, 64 TMUs, and 32 ROPs; the GTX 980 has 2,048 shading units, 128 TMUs, and 64 ROPs. The A380 has 8 ray tracing cores, while the GTX 980 has none. Pixel rate is 65.60 GPixel/s for the A380 versus 77.82 GPixel/s for the GTX 980. Texture rate is 131.2 GTexel/s for the A380 versus 155.6 GTexel/s for the GTX 980. FP32 performance is 4.198 TFLOPS for the A380 versus 4.981 TFLOPS for the GTX 980. The A380 supports FP16 at 8.397 TFLOPS, while the GTX 980 has no listed FP16 capability. TDP is 75 W for the A380 versus 165 W for the GTX 980. Power connections are 1x 8-pin for the A380 versus 2x 6-pin for the GTX 980. Suggested PSU is 250 W for the A380 versus 450 W for the GTX 980. The A380 uses PCIe 4.0 x8; the GTX 980 uses PCIe 3.0 x16. Display outputs are 1x HDMI 2.1 and 3x DisplayPort 2.0 for the A380, while the GTX 980 has 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.2. The A380 measures 222 mm in length, while the GTX 980 is longer at 267 mm.