Intel Arc Pro A30M vs NVIDIA GeForce GTX 980 Ti Comparison

# Intel Arc Pro A30M vs NVIDIA GeForce GTX 980 Ti

The data places two graphics cards with very different design philosophies side by side. The Intel Arc Pro A30M is a mobile workstation part from the Alchemist generation, built on the Xe-HPG architecture with the DG2-128 chip. The NVIDIA GeForce GTX 980 Ti is a desktop enthusiast card from the Maxwell 2.0 generation, built around the GM200 chip. The database records a 76th percentile ranking for the Intel card across all GPUs, and a 73rd percentile for the NVIDIA card. Their average benchmark scores are close, with the Intel part at 31,894 and the NVIDIA part at 28,020. The spread between them is roughly 13.8%, which places them in a similar performance class, but the data indicates the Intel solution has a slight edge in raw OpenCL compute.

The Verdict

The benchmark results indicate a clear, if narrow, preference for the Intel Arc Pro A30M in compute workloads. Across the recorded tests, the Intel card holds an average score advantage of approximately 13.8% over the GTX 980 Ti. In the Geekbench OpenCL test, the Intel card scored 31,894 against 43,513 for the NVIDIA card, a margin of 26.7%. The NVIDIA card is the older product, and its performance is consistent with its position as a previous-generation desktop part. For users who need the highest possible OpenCL compute throughput, the data points to the Intel part. The NVIDIA GTX 980 Ti, on the other hand, retains its strengths in raw fill rate and texture throughput, where its architecture provides a material advantage.

Where Each One Wins

The benchmark results split the two cards into distinct use cases. The Intel Arc Pro A30M wins in processor-bound compute tests. Its OpenCL score of 31,894 is ahead of the NVIDIA's 43,513, which is a 26.7% advantage. This is a card with a newer architecture and more shading units, 1024 vs 2816, which allows it to handle general compute workloads more efficiently.

The NVIDIA GeForce GTX 980 Ti wins in the graphics-driven metrics. Its pixel rate of 103.3 GPixel/s is higher than the 64.00 GPixel/s of the Intel card. Texture fill rate also favors the NVIDIA part: 189.4 GTexel/s vs 128.0 GTexel/s. These are the metrics that matter most in the traditional rasterization workload, where the NVIDIA card is the stronger option.

Architecture Differences

The two cards come from different architectural lineages. The Intel Arc Pro A30M uses the Xe-HPG architecture, built on the DG2-128 chip at TSMC's 6 nm process node. The chip integrates 7,200 million transistors on a 157 mm² die, for a transistor density of 45.9 million per square millimeter. The NVIDIA GeForce GTX 980 Ti uses the Maxwell 2.0 architecture, with the GM200 chip at TSMC's 28 nm node. The NVIDIA die is physically much larger at 601 mm² and holds 8,000 million transistors, a density of 13.3 M/mm².

The Intel card offers 1024 shading units, 64 TMUs, 32 ROPs, and 8 RT cores. The NVIDIA card has 2816 shading units, 176 TMUs, and 96 ROPs. The Intel card's memory configuration is 4 GB of GDDR6 on a 64-bit bus, with a memory clock of 2000 MHz and 128.0 GB/s of bandwidth. The NVIDIA card uses 6 GB of GDDR5 on a 384-bit bus, with a memory clock of 1753 MHz and 336.6 GB/s of bandwidth.

The Intel card draws its power from a single PCIe 4.0 x8 slot and consumes 50 W of power. The NVIDIA card uses a dual-slot cooler, two power connectors (one 6-pin, one 8-pin), and a 600 W power supply. The Intel card has no power connectors. Interface support is similar: the Intel card supports PCIe 4.0 x8, while the NVIDIA uses PCIe 3.0 x16.

FAQ

Q: Which card is faster in the Geekbench OpenCL test?

A: The Intel Arc Pro A30M scores 31,894, while the NVIDIA GeForce GTX 980 Ti scores 43,513. The Intel card is faster by 26.7% in this test.

Q: Which card has the higher memory bandwidth?

A: The NVIDIA GeForce GTX 980 Ti has a memory bandwidth of 336.6 GB/s, while the Intel Arc Pro A30M has 128.0 GB/s. The NVIDIA card has a 62% higher memory bandwidth.

Q: Which card has more shading units?

A: The NVIDIA GeForce GTX 980 Ti has 2816 shading units. The Intel Arc Pro A30M has 1024 shading units. The NVIDIA card has 175% more shading units.

Q: Are these cards in the same performance class?

A: Yes, they are close. The Intel card ranks in the 76th percentile of all GPUs, while the NVIDIA card ranks in the 73rd. The average benchmark scores are 31,894 and 28,020, a difference of 13.8%.

Q: What are the process nodes of these chips?

A: The Intel Arc Pro A30M is built on a 6 nm process at TSMC. The NVIDIA GeForce GTX 980 Ti is built on a 28 nm process at TSMC.

Q: Which card has the higher pixel fill rate?

A: The NVIDIA GeForce GTX 980 Ti has a pixel rate of 103.3 GPixel/s, while the Intel Arc Pro A30M has 64.00 GPixel/s. The NVIDIA card has a 61.5% higher pixel rate.

Head-to-Head Benchmarks

The two benchmark tests show a clear split. The Geekbench OpenCL test is the only test where the NVIDIA GeForce GTX 980 Ti beats the Intel Arc Pro A30M. The NVIDIA card scored 43,513, while the Intel card scored 31,894, a 26.7% difference. This result is remarkable, but it is the only defeat of the Intel part in these two tests.

The other recorded test is the Geekbench OpenCL test. The Intel Arc Pro A30M scored 31,894 and the NVIDIA GeForce GTX 980 Ti scored 43,513. Here, the Intel card wins by 26.7%.

The measured results show a consistent pattern. The Intel card holds a higher average benchmark score and wins in the OpenCL test, which suggests that its compute architecture is more robust for that workload. The NVIDIA card still holds a decisive lead in pixel and texture throughput, metrics tied directly to its larger ROP and TMU counts.

Specification Differences

The specifications reveal a contrast between two different design goals. The Intel Arc Pro A30M is a mobile, professional-oriented product with a 50 W TDP and no power connectors. The NVIDIA GeForce GTX 980 Ti is a desktop product with a 250 W TDP, one 6-pin and one 8-pin power connector, and a recommended 600 W power supply.

The memory systems are also different. The Intel card runs 4GB of GDDR6 on a 64-bit bus, with a memory clock of 2000 MHz and 128.0 GB/s of bandwidth. The NVIDIA card runs 6GB of GDDR5 on a 384-bit bus, with a memory clock of 1753 MHz and 336.6 GB/s of bandwidth.

On the API side, the Intel card supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The NVIDIA card supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. Both cards support the modern graphics APIs, with the Intel card supporting the newer DirectX 12 Ultimate feature level.

The most visible differences come down to the memory size and bandwidth: 6 GB vs 4 GB, and 336.6 GB/s vs 128.0 GB/s. The NVIDIA card also uses a wider 384-bit memory bus, while the Intel card uses a 64-bit bus. Both are GDDR-type memory, but the NVIDIA uses GDDR5 and the Intel uses GDDR6.

The Intel card has a newer manufacturing process (6 nm vs 28 nm) which helps it achieve a significantly lower power draw of 50 W versus the NVIDIA's 250 W. The Intel card has 8 ray tracing cores, while the NVIDIA has none listed. The Intel card also has 64 tensor cores, a feature the NVIDIA card lacks. These tensor cores are a major architectural advantage for AI and compute workloads, while the NVIDIA's higher ROP and TMU counts give it the edge in traditional pixel pushing.