Intel Arc A310 vs NVIDIA GeForce GTX 650 Ti Boost Comparison

# Head-to-Head Benchmarks

The data reveals a decisive generational gap between these two GPUs, with the Intel Arc A310 dominating both shared benchmark tests. In Geekbench OpenCL, the Arc A310 scores 30,607 against the GTX 650 Ti Boost's 9,302, a 69.6% margin in Intel's favor. The Vulkan test tells a similar story: the Arc A310 reaches 28,964 while the NVIDIA card manages 10,041, leaving the older GPU 65.3% behind. These are not close contests; the Intel part more than triples the NVIDIA card's compute output in both APIs.

The average benchmark scores reflect this overall trend, though the gap narrows considerably. The GTX 650 Ti Boost posts an average score of 8,067 across all its benchmark entries, while the Arc A310 averages 7,550. That places the NVIDIA card in the 42nd percentile of all GPUs, slightly ahead of the Arc A310's 40th percentile. The head-to-head results show Intel winning 2 tests to 0, but the broader benchmark suite suggests the NVIDIA card holds its own in some workloads that favor its architecture.

Looking at the nearest rivals provides additional context. The GTX 650 Ti Boost sits within 0.3% of the NVIDIA GRID K2, GTX 650 Ti, GTX 880M, and Quadro P5000, all clustered around the 8,040–8,080 average score range. The Arc A310's closest competitors include the AMD Radeon R7 250 at 7,557 (0.1% ahead), the Radeon Pro WX 3100 at 7,580 (0.4% ahead), and the NVIDIA GeForce GTX 1650 at 7,472 (1% behind). This positioning suggests the Arc A310 competes with entry-level discrete GPUs from several generations ago, while the GTX 650 Ti Boost punches near mid-range territory of its era.

The benchmark data implies that raw compute performance tells only part of the story. The Arc A310's OpenCL and Vulkan scores are remarkable for a 30 W card, but the GTX 650 Ti Boost's higher average across mixed workloads hints at strengths in legacy or driver-optimized paths that the head-to-head tests do not capture.

# FAQ

Q: Which GPU wins the direct head-to-head benchmark comparison?

A: The Intel Arc A310 wins both shared tests. It leads by 69.6% in Geekbench OpenCL (30,607 vs 9,302) and by 65.3% in Geekbench Vulkan (28,964 vs 10,041).

Q: How do the average benchmark scores compare between the two cards?

A: The NVIDIA GeForce GTX 650 Ti Boost has a higher average benchmark score of 8,067, compared to the Intel Arc A310's 7,550. This places the NVIDIA card at the 42nd percentile of all GPUs, versus the Arc A310's 40th percentile.

Q: What are the closest rivals for each GPU based on average score?

A: For the GTX 650 Ti Boost, the nearest rival is the NVIDIA GRID K2 at 8,080 (0.2% higher), followed by the GTX 650 Ti at 8,053 (0.2% lower). For the Arc A310, the AMD Radeon R7 250 at 7,557 is 0.1% higher, and the NVIDIA GeForce GTX 1650 at 7,472 is 1% lower.

Q: Does the Intel Arc A310 support ray tracing?

A: Yes, the Arc A310 includes 6 dedicated ray tracing cores, a feature the GTX 650 Ti Boost lacks entirely. This is part of the Xe-HPG architecture's DirectX 12 Ultimate support.

Q: Which card has higher memory bandwidth despite smaller bus width?

A: The GTX 650 Ti Boost has higher bandwidth at 144.2 GB/s over a 192-bit bus, while the Arc A310 delivers 124.0 GB/s over a 64-bit bus. The Intel card compensates with faster GDDR6 memory at 15.5 Gbps effective, versus the NVIDIA card's GDDR5 at 6 Gbps effective.

Q: How significant is the power consumption difference?

A: The Arc A310 draws 30 W compared to the GTX 650 Ti Boost's 134 W, making Intel's card roughly four times more power-efficient. The suggested PSU requirement drops from 300 W to 200 W, and the Arc A310 needs no external power connectors while the NVIDIA card requires a single 6-pin connector.

# Architecture Differences

The architectural chasm between these two GPUs spans nearly a decade of semiconductor evolution. The NVIDIA GeForce GTX 650 Ti Boost uses the GK106 chip built on Kepler architecture, fabricated on TSMC's 28 nm process. It packs 2,540 million transistors into a 221 mm² die, yielding a transistor density of 11.5 million per square millimeter. The Intel Arc A310, by contrast, employs the DG2-128 chip under the Xe-HPG architecture, also on TSMC silicon but at 6 nm. This newer node crams 7,200 million transistors into a smaller 157 mm² die, achieving 45.9 million transistors per square millimeter — a fourfold density improvement.

The compute layouts differ substantially. Both cards feature 768 shading units, but the NVIDIA card pairs them with 64 texture mapping units and 24 raster operation pipelines, while the Arc A310 uses 32 TMUs and 16 ROPs. The Intel card compensates with 6 ray tracing cores, a feature absent from the Kepler design. Pixel and texture throughput reflect these configurations: the GTX 650 Ti Boost hits 16.51 GPixel/s and 66.05 GTexel/s, while the Arc A310 reaches 28.00 GPixel/s and 56.00 GTexel/s.

The API support reveals the generational shift. The GTX 650 Ti Boost lists DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The Arc A310 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, adding hardware ray tracing and modern feature sets. The Arc also offers FP16 compute at 5.376 TFLOPS via a 2:1 ratio, while the NVIDIA card lists no FP16 capability. The Intel card's FP32 throughput of 2.688 TFLOPS nearly doubles the GTX 650 Ti Boost's 1.585 TFLOPS, explaining its dominant OpenCL and Vulkan results.

# Specification Differences

The two cards diverge across nearly every specification category. The GTX 650 Ti Boost operates at a 980 MHz base clock with a 1032 MHz boost, while the Arc A310 runs at a flat 1750 MHz for both base and boost. Memory configurations contrast sharply: the NVIDIA card offers 2 GB of GDDR5 on a 192-bit bus with 144.2 GB/s bandwidth, whereas the Intel card provides 4 GB of GDDR6 on a 64-bit bus with 124.0 GB/s bandwidth. Memory clocks differ as well, with the GTX 650 Ti Boost at 1502 MHz (6 Gbps effective) and the Arc A310 at 1937 MHz (15.5 Gbps effective).

Physical and power characteristics could hardly be more different. The GTX 650 Ti Boost is a dual-slot card measuring 241 mm (9.5 inches) in length, requiring a 300 W suggested PSU and a single 6-pin power connector. The Arc A310 is single-slot with no external power connectors and a 200 W suggested PSU. Display outputs also diverge: the NVIDIA card offers 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2, while the Intel card provides 4x mini-DisplayPort 2.0. The bus interface upgrades from PCIe 3.0 x16 on the NVIDIA card to PCIe 4.0 x8 on the Intel part.

Release timing spans nearly a decade: the GTX 650 Ti Boost launched March 25, 2013, while the Arc A310 arrived October 11, 2022. Both are now end-of-life products. The NVIDIA card had a launch MSRP of 169 USD; the Arc A310 has no recorded launch MSRP in the data. The NVIDIA card's predecessor is the GeForce 500 series and successor the GeForce 700 series, while the Arc A310 follows Xe Graphics and precedes Battlemage.

# The Verdict

The benchmark data paints a nuanced picture. In raw compute performance, the Intel Arc A310 is the clear winner, tripling the GTX 650 Ti Boost's scores in OpenCL and Vulkan. Its modern architecture delivers ray tracing, DirectX 12 Ultimate, and dramatically better power efficiency — 30 W versus 134 W. Anyone prioritizing compute workloads, modern API support, or energy-conscious operation should choose the Arc A310 without hesitation.

However, the average benchmark score tells a different story. The GTX 650 Ti Boost's 8,067 average edges out the Arc A310's 7,550, and its 42nd percentile ranking versus 40th suggests the older card maintains competitiveness in certain legacy workloads. The NVIDIA card also offers higher memory bandwidth (144.2 GB/s vs 124.0 GB/s) and more TMUs and ROPs, which may benefit texture-heavy or rasterization-focused tasks from its era.

The verdict depends on context. For modern applications leveraging Vulkan or OpenCL, the Arc A310's advantages are overwhelming. For older DirectX 9/10/11 titles or workloads optimized for Kepler's specific strengths, the GTX 650 Ti Boost's higher average score and established driver maturity may prove more reliable. The data does not support a universal winner — it supports two different winners for two different use cases.

# Where Each One Wins

The Intel Arc A310 dominates in compute-oriented scenarios. Its Geekbench OpenCL score of 30,607 versus 9,302 demonstrates a 69.6% advantage, making it the obvious choice for general-purpose GPU compute, machine learning inference, or any OpenCL-accelerated workload. The Vulkan test shows a similar 65.3% lead, suggesting the Arc A310 excels in modern gaming APIs and Vulkan-based applications. Its 6 ray tracing cores and DirectX 12 Ultimate support make it the only option for hardware-accelerated ray tracing. The 4 GB GDDR6 memory capacity doubles the NVIDIA card's 2 GB, aiding larger textures and datasets. The 30 W power draw and single-slot design make it suitable for low-power or space-constrained systems, with no external power connectors required.

The NVIDIA GeForce GTX 650 Ti Boost wins in legacy compatibility and specific hardware metrics. Its higher average benchmark score of 8,067 versus 7,550 indicates broader consistency across mixed workloads, even if its peak scores are lower. The 192-bit memory bus and 144.2 GB/s bandwidth exceed the Arc A310's 124.0 GB/s, potentially benefiting memory-bandwidth-sensitive applications. With 64 TMUs versus 32, the NVIDIA card delivers 66.05 GTexel/s against the Arc A310's 56.00 GTexel/s, suggesting superior texture fill rates. Its launch date in 2013 means it has over nine years of driver optimization for older DirectX 9/10/11 titles, which the Arc A310's PassMark scores (69 in DirectX 9, 33 in DirectX 11) suggest may still be relevant. The 2 GB memory footprint may be limiting, but for period-appropriate games and applications, the GTX 650 Ti Boost's established compatibility and higher average score make it the safer pick.