Intel Arc A350M vs NVIDIA GeForce GTX 780 Ti Comparison

The Intel Arc A350M and the NVIDIA GeForce GTX 780 Ti represent two very different eras of GPU design, and the benchmark data shows a clear, if not overwhelming, winner in raw compute. The GeForce GTX 780 Ti, a desktop card from 2013, holds a decisive edge in both available compute tests, though the Arc A350M is a mobile part with a fraction of the power draw. The GTX 780 Ti wins the OpenCL test with a score of 27326 against the Arc A350M’s 24546, a 10.2% advantage. In Vulkan, the lead narrows slightly to 9.1%, with the NVIDIA card scoring 27238 versus the Intel part’s 24747. Despite these losses, the Arc A350M’s average benchmark score of 24647 is actually higher than the GTX 780 Ti’s 24236 average, a quirk driven by the GTX 780 Ti’s additional Metal benchmark result of 18144, which drags its average down. The data shows that while the older card wins the head-to-head compute tests, the newer mobile GPU is more consistent across different API workloads.

Head-to-Head Benchmarks

The two benchmark tests included in the comparison are Geekbench OpenCL and Geekbench Vulkan. The NVIDIA GeForce GTX 780 Ti emerges as the winner in both, but the margin is not overwhelming. In the OpenCL test, the GTX 780 Ti scores 27326, beating the Intel Arc A350M’s 24546 by a delta of -10.2% from the Intel card’s perspective. This means the Intel part trails by about ten percent, a significant gap for a GPU that is nearly a decade newer. The Vulkan result tells a similar story: the GTX 780 Ti posts 27238, while the Arc A350M manages 24747, a 9.1% deficit. The consistency of the NVIDIA card’s performance across both APIs suggests that its older Kepler architecture still has substantial raw compute muscle, particularly in its 2880 shading units and 5.345 TFLOPS of FP32 throughput.

However, the average benchmark scores paint a more nuanced picture. The Arc A350M has an average benchmark score of 24647, which is 1.7% higher than the GTX 780 Ti’s 24236. This is because the NVIDIA card also includes a Geekbench Metal test, where it scores 18144, significantly lower than its OpenCL and Vulkan results. That Metal score pulls its average down, while the Intel card only has OpenCL and Vulkan results, both of which are closely clustered. When placing these GPUs against their nearest rivals, the Arc A350M sits at a 70th percentile versus all GPUs, and its nearest competitor is the AMD Radeon RX 590, which has an average score of 24744, putting the Intel card just 0.4% behind. The GTX 780 Ti also sits at the 70th percentile, but its nearest rival is the NVIDIA GeForce GTX 1630, which averages 24277, making the GTX 780 Ti just 0.2% slower.

The head-to-head data shows that the GTX 780 Ti’s wins are not dominant, but they are consistent. The Intel card’s best performance relative to the NVIDIA card comes in Vulkan, where the gap is 9.1% rather than 10.2%. This could indicate that the Arc A350M’s Xe-HPG architecture, which supports Vulkan 1.4, is better optimized for modern APIs than the Kepler architecture’s Vulkan 1.2.175 implementation. Yet, the raw numbers still favor the older card, which has a memory bandwidth of 336.6 GB/s over a 384-bit bus, compared to the Intel card’s 112.0 GB/s over a 64-bit bus. That six-fold difference in bandwidth is likely a major factor in the NVIDIA card’s compute lead.

The Verdict

The data is clear: the NVIDIA GeForce GTX 780 Ti is the faster GPU in raw compute benchmarks. It wins both head-to-head tests, and its FP32 throughput of 5.345 TFLOPS is significantly higher than the Intel Arc A350M’s 3.379 TFLOPS. For users who need maximum compute performance in OpenCL or Vulkan workloads, the GTX 780 Ti is the pick. Its texture rate of 222.7 GTexel/s also dwarfs the Intel card’s 105.6 GTexel/s, and its pixel rate of 55.68 GPixel/s edges out the Arc A350M’s 52.80 GPixel/s. The NVIDIA card’s 3 GB of GDDR5 memory on a 384-bit bus provides 336.6 GB/s of bandwidth, which is three times the Intel card’s bandwidth, making it better suited for memory-heavy tasks.

However, the Intel Arc A350M is not without its merits. It has a higher average benchmark score of 24647 compared to the GTX 780 Ti’s 24236, driven by the NVIDIA card’s weaker Metal performance. The Arc A350M also has a much lower TDP of 25 W, versus the GTX 780 Ti’s 250 W, making it a far more efficient part. The Intel card is an IGP (integrated graphics processor), meaning it is designed for portable devices, while the GTX 780 Ti is a dual-slot desktop card requiring a 600 W power supply. If power efficiency, modern API support, or a compact form factor are the priorities, the Arc A350M is the sensible choice. It supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the GTX 780 Ti is limited to DirectX 12 (11_1) and Vulkan 1.2.175. The Arc A350M also has 6 ray tracing cores, a feature the GTX 780 Ti lacks entirely. For gamers or professionals who prioritize ray tracing and modern feature sets, the Intel card is the better option, despite its lower raw compute.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The Intel Arc A350M has a higher average benchmark score of 24647, compared to the NVIDIA GeForce GTX 780 Ti’s 24236. This is partly because the NVIDIA card’s Metal score of 18144 lowers its average.

Q: Does the NVIDIA GeForce GTX 780 Ti support ray tracing?

A: No. The GTX 780 Ti has no ray tracing cores, while the Intel Arc A350M has 6 RT cores. The Intel card also supports DirectX 12 Ultimate, which includes ray tracing features.

Q: What is the memory bandwidth difference between the two cards?

A: The NVIDIA GeForce GTX 780 Ti has a memory bandwidth of 336.6 GB/s, while the Intel Arc A350M has 112.0 GB/s. The NVIDIA card uses a 384-bit bus with 3 GB of GDDR5 memory, whereas the Intel card uses a 64-bit bus with 4 GB of GDDR6 memory.

Q: How do the two GPUs compare in the Geekbench Vulkan test?

A: The NVIDIA GeForce GTX 780 Ti scores 27238 in Vulkan, beating the Intel Arc A350M’s 24747 by 9.1%. This is the smaller of the two head-to-head margins, suggesting the Intel card is more competitive in modern API workloads.

Q: Which card has a higher transistor density?

A: The Intel Arc A350M has a transistor density of 45.9M / mm², while the NVIDIA GeForce GTX 780 Ti has 12.6M / mm². This is due to the Intel card’s 6 nm process node versus the NVIDIA card’s 28 nm node.

Q: What is the power consumption of each GPU?

A: The Intel Arc A350M has a TDP of 25 W, while the NVIDIA GeForce GTX 780 Ti has a TDP of 250 W. The NVIDIA card also requires a 600 W suggested power supply and uses 1x 6-pin + 1x 8-pin power connectors.

Specification Differences

The two GPUs differ dramatically in nearly every specification. The Intel Arc A350M is built on a 6 nm process at TSMC, with a die size of 157 mm², whereas the NVIDIA GeForce GTX 780 Ti uses a 28 nm process and has a die size of 561 mm². The Intel card packs 7,200 million transistors into its smaller die, while the NVIDIA card has 7,080 million transistors on a much larger die. This leads to a transistor density of 45.9M / mm² for the Intel card versus 12.6M / mm² for the NVIDIA card. The memory subsystems are also completely different: the Arc A350M has 4 GB of GDDR6 on a 64-bit bus, while the GTX 780 Ti has 3 GB of GDDR5 on a 384-bit bus. The clocks differ as well, with the Intel card boosting to 2200 MHz and the NVIDIA card boosting to 928 MHz. The GTX 780 Ti has a base clock of 875 MHz, while the Intel card’s base is 1150 MHz. The NVIDIA card has 2880 shading units, 240 TMUs, and 48 ROPs, while the Intel card has 768 shading units, 48 TMUs, and 24 ROPs. Power consumption is a stark contrast: the Arc A350M sips at 25 W, while the GTX 780 Ti draws 250 W. The form factors also differ, with the Intel card being an IGP and the NVIDIA card being a dual-slot card with dimensions of 267 mm x 111 mm x 38 mm. The bus interface is PCIe 4.0 x8 for the Intel card and PCIe 3.0 x16 for the NVIDIA card.

Architecture Differences

The Intel Arc A350M uses the Xe-HPG architecture, specifically the DG2-128 chip, and belongs to the Alchemist generation (Arc 3 Mobile). It is fabricated by TSMC on a 6 nm node. The NVIDIA GeForce GTX 780 Ti uses the Kepler architecture, based on the GK110B chip, and belongs to the GeForce 700 generation. It is also fabricated by TSMC, but on a 28 nm node. The Intel card supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, and it includes 6 RT cores for ray tracing. The NVIDIA card supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.175, but has no RT cores. The Intel card’s FP32 performance is 3.379 TFLOPS, and it has a FP16 performance of 6.758 TFLOPS (2:1). The NVIDIA card has an FP32 performance of 5.345 TFLOPS and no listed FP16 capability. The Intel card has a pixel rate of 52.80 GPixel/s and a texture rate of 105.6 GTexel/s, while the NVIDIA card has a pixel rate of 55.68 GPixel/s and a texture rate of 222.7 GTexel/s. The NVIDIA card’s display outputs include 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2, whereas the Intel card’s outputs are described as portable device dependent. The GTX 780 Ti was released on 2013-11-06, while the Arc A350M was released on 2022-03-29. The NVIDIA card’s predecessor is the GeForce 600 series and its successor is the GeForce 900 series; the Intel card has no predecessor or successor listed.