GPU Comparison

The NVIDIA GeForce GTX 980 and the Intel Arc A310 represent two very different approaches to graphics hardware, separated by eight years of architectural evolution. The benchmark data shows a decisive overall victory for the older NVIDIA card, but the Intel part claims a notable win in one modern API test. The GTX 980 secures 8 of 9 head-to-head victories, with its most dramatic margin coming in Passmark’s DirectX 11 test, where it scores 83 against the Arc A310’s 33, a 151.5% advantage. The NVIDIA card also more than doubles the Intel part in Passmark G3D, posting 11095 versus 5433, a 104.2% lead. The only test the Arc A310 wins is Geekbench Vulkan, where it scores 28964 against the GTX 980’s 22543, a 22.2% margin in favor of Intel.

Head-to-Head Benchmarks

The Geekbench OpenCL results show the GTX 980 ahead by a solid margin, scoring 34676 compared to the Arc A310’s 30607, a 13.3% difference. This is the closest contest between the two cards in any benchmark, suggesting that raw compute workloads are relatively balanced. The Vulkan result flips the script entirely. Intel’s Arc A310 delivers 28964 points, which is 22.2% higher than the GTX 980’s 22543. This is the only instance where the newer architecture’s API support translates into a measurable performance advantage.

DirectX legacy tests heavily favor the NVIDIA card. In Passmark DirectX 10, the GTX 980 scores 53 against 31, a 71% lead. The DirectX 11 test is even more lopsided: 83 versus 33, a 151.5% advantage for NVIDIA. DirectX 12 results show a similar pattern, with the GTX 980 at 46 and the Arc A310 at 29, a 58.6% gap. The DirectX 9 benchmark is also one-sided, with the GTX 980 scoring 164 versus 69, a 137.7% difference. These results indicate that the older Maxwell architecture retains a significant edge in legacy DirectX workloads.

The Passmark G2D test measures 2D graphics performance, and here the GTX 980 wins 792 to 625, a 26.7% margin. The G3D test, which is more representative of overall 3D rendering, shows the GTX 980 at 11095, which is more than double the Arc A310’s 5433, a 104.2% lead. In GPU compute, the GTX 980 again dominates, scoring 4753 against 2157, a 120.4% advantage. The average benchmark score aggregates all tests, giving the GTX 980 an 8167 average versus 7550 for the Arc A310, a difference that places the NVIDIA card at the 43rd percentile of all GPUs, while the Intel card sits at the 40th percentile.

Where Each One Wins

The NVIDIA GeForce GTX 980 is the clear winner for anyone running DirectX-based games or applications, particularly those that rely on older API versions. Its Passmark DirectX 9 score of 164 is more than double the Arc A310’s 69, and its DirectX 11 score of 83 is two and a half times higher. For users with a library of legacy titles, the GTX 980 offers substantially better compatibility and performance. The card also wins decisively in 3D rendering workloads, as shown by its 104.2% lead in Passmark G3D, and in compute tasks, where its 120.4% advantage in Passmark GPU compute is substantial.

The Intel Arc A310’s single victory comes in Geekbench Vulkan, where it outperforms the GTX 980 by 22.2%. This suggests that for applications built around the Vulkan API, the newer Intel architecture handles the workload more efficiently. The Arc A310 also benefits from a much lower power envelope, with a 30 W TDP compared to the GTX 980’s 165 W, though this is a specification difference rather than a performance result. The Intel card’s average benchmark score of 7550 is close to its nearest rival, the AMD Radeon R7 250, which averages 7557, a delta of just -0.1%. In contrast, the GTX 980’s nearest rival is the NVIDIA GeForce GTX 950M, which averages 8135, a delta of 0.4%, showing the GTX 980 is positioned in a slightly higher performance tier.

FAQ

Q: Which card has the higher average benchmark score?

A: The NVIDIA GeForce GTX 980 has an average benchmark score of 8167, while the Intel Arc A310 averages 7550. This places the GTX 980 at the 43rd percentile of all GPUs and the Arc A310 at the 40th percentile.

Q: In which benchmark does the Intel Arc A310 outperform the GTX 980?

A: The Arc A310 wins the Geekbench Vulkan test, scoring 28964 against the GTX 980’s 22543, a 22.2% margin in Intel’s favor. This is the only head-to-head test where the Intel card comes out ahead.

Q: How large is the GTX 980’s lead in DirectX 11 performance?

A: In Passmark DirectX 11, the GTX 980 scores 83, while the Arc A310 scores 33. This represents a 151.5% advantage for the NVIDIA card, the largest performance gap in any benchmark between the two.

Q: What are the respective TDP values for these cards?

A: The NVIDIA GeForce GTX 980 has a TDP of 165 W and requires a suggested power supply of 450 W. The Intel Arc A310 has a TDP of 30 W and a suggested power supply of 200 W.

Q: How do the two cards compare in memory bandwidth?

A: The GTX 980 has a 256-bit memory bus with GDDR5 memory and a bandwidth of 224.4 GB/s. The Arc A310 has a 64-bit bus with GDDR6 memory and a bandwidth of 124.0 GB/s.

Q: Which card has more shading units?

A: The GTX 980 has 2048 shading units, while the Arc A310 has 768. The GTX 980 also has 128 texture mapping units and 64 ROPs, versus 32 TMUs and 16 ROPs for the Arc A310.

Specification Differences

The two cards differ significantly in nearly every core specification. The GTX 980 operates at a base clock of 1127 MHz with a boost clock of 1216 MHz, while the Arc A310 runs at a flat 1750 MHz for both base and boost. Memory configuration also diverges: the GTX 980 uses 4 GB of GDDR5 on a 256-bit bus, yielding 224.4 GB/s of bandwidth, whereas the Arc A310 uses 4 GB of GDDR6 on a 64-bit bus, providing 124.0 GB/s. The pixel rate for the GTX 980 is 77.82 GPixel/s, compared to 28.00 GPixel/s for the Arc A310. Texture rate follows the same pattern, with the GTX 980 at 155.6 GTexel/s and the Arc A310 at 56.00 GTexel/s.

The FP32 compute throughput is another major differentiator. The GTX 980 delivers 4.981 TFLOPS, while the Arc A310 delivers 2.688 TFLOPS. The Intel card does support FP16 at 5.376 TFLOPS with a 2:1 ratio, a feature the GTX 980 does not list. The power requirements are starkly different: the GTX 980 draws 165 W and needs two 6-pin power connectors, while the Arc A310 draws only 30 W and requires no external power connectors. The suggested power supply is 450 W for the GTX 980 and 200 W for the Arc A310. Physical dimensions also differ, with the GTX 980 being a dual-slot card measuring 267 mm in length, while the Arc A310 is a single-slot card with no listed dimensions. The bus interface is PCIe 3.0 x16 for NVIDIA and PCIe 4.0 x8 for Intel.

Architecture Differences

The architectural gap between these two GPUs is substantial. The GTX 980 uses the GM204 chip built on NVIDIA’s Maxwell 2.0 architecture, fabricated on a 28 nm process at TSMC. The Arc A310 uses the DG2-128 chip based on Intel’s Xe-HPG architecture, manufactured on a 6 nm process, also at TSMC. The transistor counts reflect the process node difference: the GTX 980 contains 5,200 million transistors on a 398 mm² die, while the Arc A310 packs 7,200 million transistors on a much smaller 157 mm² die. This results in a transistor density of 13.1M per mm² for the GTX 980 and 45.9M per mm² for the Arc A310.

The Arc A310 features 6 dedicated ray tracing cores, a capability entirely absent from the GTX 980. The Intel card also 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 Arc A310 is part of the Alchemist generation (Arc 3) and succeeds Xe Graphics, with Battlemage as its successor. The GTX 980 belongs to the GeForce 900 generation, succeeding the GeForce 700 series and preceding the GeForce 10 series. The NVIDIA card offers display outputs of 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.2, while the Arc A310 provides 4x mini-DisplayPort 2.0.

The Verdict

The data clearly favors the NVIDIA GeForce GTX 980 for users prioritizing raw performance across most workloads. Its wins in 8 of 9 head-to-head benchmarks, including a 151.5% lead in DirectX 11 and a 104.2% lead in Passmark G3D, make it the stronger choice for gaming and general 3D rendering. The GTX 980 also holds a 13.3% advantage in OpenCL compute and a 120.4% lead in GPU compute, making it better suited for compute-heavy tasks. Its higher average benchmark score of 8167, compared to 7550 for the Arc A310, places it in a higher performance percentile.

The Intel Arc A310 is the better option only for workloads that leverage the Vulkan API, where it leads by 22.2% in Geekbench Vulkan. It also offers modern features like ray tracing and DirectX 12 Ultimate support, which the GTX 980 lacks. The Arc A310’s dramatically lower power draw of 30 W versus 165 W makes it far more efficient, though efficiency is not a benchmark score. For users building a system around Vulkan-based software or requiring the latest API features, the Arc A310 is the appropriate pick. For everyone else, the GTX 980’s benchmark results make it the superior performer despite its older architecture. The GTX 980 launched with an MSRP of 549 USD, a figure that reflects its original high-end positioning, while the Arc A310 has no listed launch MSRP.