GPU Comparison

The Intel Arc A310 and the NVIDIA GeForce GTX 460 are separated by over a decade of GPU architecture, yet benchmark data reveals a surprisingly narrow overall performance gap. While the Arc A310 dominates in raw compute and modern API support, the GTX 460’s legacy driver optimizations and specific workload characteristics keep it competitive in certain legacy scenarios.

Head-to-Head Benchmarks

The only direct head-to-head benchmark available is Geekbench OpenCL, and the result is a decisive victory for the Intel Arc A310. The Arc A310 scores 30,607 points, while the GTX 460 manages just 7,925 points. That represents a 74.1% deficit for the older NVIDIA card. This is not a marginal difference; the Arc A310 delivers nearly four times the compute throughput in this OpenCL workload, reflecting its modern 6 nm process node, 2.688 TFLOPS FP32 performance, and 768 shading units versus the GTX 460’s 907.2 GFLOPS and 336 shading units.

However, the broader benchmark picture is more nuanced. The GTX 460’s average benchmark score is 7,925, placing it in the 41st percentile of all GPUs. The Arc A310’s average across all its tested workloads is 7,550, placing it in the 40th percentile. Despite the OpenCL blowout, the two cards sit within 4.7% of each other in average score, with the GTX 460 actually holding a slight edge in overall average. This discrepancy suggests that the Arc A310’s OpenCL result is an outlier, while its real-world performance across a wider set of tests is far less impressive.

Looking at the Arc A310’s individual results, the pattern is clear: it excels in compute-oriented tasks but struggles in rasterization-heavy workloads. Its PassMark G3D score is 5,433, which is respectable, but its DirectX 9 score is just 69, its DirectX 10 score is 31, its DirectX 11 score is 33, and its DirectX 12 score is 29. These extremely low DirectX scores indicate significant driver overhead or architectural inefficiencies in traditional gaming APIs. The GTX 460, by contrast, has no such DirectX benchmark data available, but its Fermi architecture was designed for the DirectX 11 era and is known to perform consistently in that generation’s titles.

The nearest rivals for each card further illustrate the performance context. The GTX 460’s closest competitor is the NVIDIA Quadro K4100M, which averages 7,906 points, a mere 0.2% difference. The Arc A310’s nearest rival is the AMD Radeon R7 250, averaging 7,557 points, just 0.1% behind. Interestingly, the GTX 460 is 1.4% ahead of the NVIDIA Quadro P5000 and GTX 880M, while the Arc A310 is 1% behind the GeForce GTX 1650. These comparisons show that both cards are clustered in a narrow performance band, despite their vastly different architectures and release dates.

The Verdict

From the data, the Intel Arc A310 is the clear winner for any workload that leverages modern compute APIs like OpenCL or Vulkan. Its 30,607 OpenCL score versus the GTX 460’s 7,925 is not just a win; it is a generational leap. The Arc A310 also offers hardware ray tracing (6 RT cores), DirectX 12 Ultimate support, and Vulkan 1.4, features the GTX 460 completely lacks. For users running modern games or GPU-accelerated applications that utilize these APIs, the choice is unambiguous.

However, the GTX 460 remains the more consistent performer in legacy DirectX titles. Its Fermi architecture was built for DirectX 11, and while it only supports DirectX 12 (11_0), the baseline feature level, it was optimized for that era’s games. The Arc A310’s PassMark DirectX scores, ranging from 29 to 69, are catastrophic for any DirectX-based gaming. If the workload is primarily older DirectX 9, 10, or 11 games, the GTX 460 is likely to deliver playable frame rates where the Arc A310 may struggle or fail outright.

The average benchmark scores tell the real story: the GTX 460 averages 7,925 points (41st percentile) versus the Arc A310’s 7,550 points (40th percentile). These are statistically equivalent performance levels, differing by less than 5%. This means that for a general-purpose GPU, neither card offers a significant overall advantage. The choice comes down to workload: modern compute and ray tracing favor the Arc A310; legacy DirectX gaming favors the GTX 460.

Architecture Differences

The architectural gap between these two GPUs is vast, reflecting a 12-year evolution in GPU design. The GTX 460 uses the GF104 chip based on NVIDIA’s Fermi architecture, manufactured on a 40 nm process at TSMC. It contains 1,950 million transistors on a 332 mm² die, yielding a transistor density of 5.9 million per mm². The Arc A310 uses the DG2-128 chip based on Intel’s Xe-HPG architecture, also fabricated by TSMC but on a 6 nm process. It packs 7,200 million transistors into a much smaller 157 mm² die, achieving a transistor density of 45.9 million per mm², nearly eight times denser.

Memory configurations differ significantly. The GTX 460 has 768 MB of GDDR5 memory on a 192-bit bus, running at 3.6 Gbps effective, providing 86.40 GB/s of bandwidth. The Arc A310 has 4 GB of GDDR6 memory on a narrower 64-bit bus, running at 15.5 Gbps effective, yielding 124.0 GB/s of bandwidth. Despite the Arc’s higher total bandwidth, the GTX 460’s wider bus may offer lower latency in certain access patterns.

Compute resources are heavily skewed toward the Arc A310. It has 768 shading units, 32 TMUs, and 16 ROPs, compared to the GTX 460’s 336 shading units, 56 TMUs, and 24 ROPs. The Arc’s pixel rate is 28.00 GPixel/s versus 9.450 GPixel/s for the GTX 460, and its texture rate is 56.00 GTexel/s versus 37.80 GTexel/s. The FP32 throughput is 2.688 TFLOPS for the Arc versus 907.2 GFLOPS for the GTX 460, nearly triple. The Arc also supports FP16 at 5.376 TFLOPS (2:1 ratio), which the GTX 460 lacks entirely.

Power and physical characteristics diverge dramatically. The GTX 460 has a 160 W TDP, requires dual-slot cooling, two 6-pin power connectors, and a 450 W suggested PSU. The Arc A310 sips power at just 30 W, uses a single-slot design, requires no power connectors, and works with a 200 W suggested PSU. The GTX 460 is 210 mm long, while the Arc A310’s dimensions are not listed. The Arc also supports PCIe 4.0 x8, while the GTX 460 uses PCIe 2.0 x16.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 460 has an average benchmark score of 7,925, while the Intel Arc A310 averages 7,550. The GTX 460 is 4.7% higher, though both fall within the 40-41st percentile of all GPUs.

Q: How much faster is the Arc A310 in OpenCL compute?

A: The Arc A310 scores 30,607 in Geekbench OpenCL, while the GTX 460 scores 7,925. The Arc A310 is 74.1% faster, representing a near fourfold increase in raw compute throughput.

Q: Does the Arc A310 support hardware ray tracing?

A: Yes, the Intel Arc A310 includes 6 ray tracing cores and supports DirectX 12 Ultimate (12_2). The NVIDIA GeForce GTX 460 has no ray tracing cores and only supports DirectX 12 (11_0).

Q: What are the power requirements for each card?

A: The GTX 460 has a 160 W TDP, requires two 6-pin power connectors, and a 450 W suggested PSU. The Arc A310 has a 30 W TDP, needs no power connectors, and works with a 200 W suggested PSU.

Q: Which card has more memory bandwidth?

A: The Intel Arc A310 has 124.0 GB/s of bandwidth using 4 GB of GDDR6 on a 64-bit bus. The GTX 460 has 86.40 GB/s using 768 MB of GDDR5 on a 192-bit bus.

Q: How do the cards compare in legacy DirectX performance?

A: The Arc A310 scores 69 in PassMark DirectX 9, 31 in DirectX 10, 33 in DirectX 11, and 29 in DirectX 12. No equivalent data exists for the GTX 460, but its Fermi architecture was designed for DirectX 11-era titles.

Where Each One Wins

The Intel Arc A310 wins decisively in every modern compute and API-forward scenario. Its 2.688 TFLOPS FP32 performance, 6 RT cores, and support for DirectX 12 Ultimate and Vulkan 1.4 make it the only viable choice for ray-traced games, Vulkan-based titles, and GPU compute workloads like OpenCL. The 74.1% OpenCL victory over the GTX 460 is the single largest performance gap in the data, and its 4 GB memory buffer is four times larger, benefiting texture-heavy applications. The Arc A310’s 30 W TDP also makes it ideal for low-power or small-form-factor systems where the GTX 460’s 160 W requirement and dual-slot cooler would be prohibitive.

The NVIDIA GeForce GTX 460 wins in the field of legacy gaming and consistent average performance. Its average benchmark score of 7,925 is 4.7% higher than the Arc A310’s 7,550, and it holds a higher percentile ranking (41st versus 40th). The GTX 460’s 192-bit memory bus, while providing less total bandwidth, may offer better latency characteristics for certain access patterns. More importantly, the Arc A310’s abysmal PassMark DirectX scores (29-69) suggest it is nearly unusable for DirectX 9-12 games, while the GTX 460 was purpose-built for that era. For users with a library of older DirectX titles, the GTX 460 is the safer bet.

The GTX 460 also wins on raw pixel fill rate relative to its compute deficit, with 24 ROPs versus the Arc’s 16, though the Arc’s higher clock speeds mitigate this. In terms of connectivity, the Arc A310 offers four mini-DisplayPort 2.0 outputs, while the GTX 460 has two DVI and one mini-HDMI 1.3a, modern monitors will favor the Arc. The GTX 460’s launch MSRP was 199 USD, a fact that is historical only, as both cards are end-of-life products. The verdict is workload-dependent: modern and compute-heavy tasks go to the Arc A310, while legacy DirectX gaming stays with the GTX 460.