Intel Arc A310 vs NVIDIA T600 Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The Intel Arc A310 has a higher average benchmark score of 7550 compared to the NVIDIA T600’s 7035. This places the A310 in the 40th percentile of all GPUs, while the T600 sits at the 39th percentile.

Q: How does the Intel Arc A310 compare to the NVIDIA GeForce GTX 1650?

A: The Arc A310’s average score of 7550 is 1% higher than the GTX 1650’s average of 7472. In the nearestRivals data, the A310 also edges out the AMD Radeon R7 250 by 0.1% and the AMD Radeon Pro WX 3100 by 0.4%.

Q: What is the NVIDIA T600’s closest rival in performance?

A: The T600’s average score of 7035 is just 0.2% higher than the NVIDIA GeForce GTX 680M’s 7023. It also leads the AMD Radeon R5 M240 by 0.9% and the NVIDIA GeForce GTX 675M by 1.3%, while trailing the NVIDIA GeForce GTX 970 by 1.7%.

Q: Which card wins in DirectX 12 benchmarks?

A: The Intel Arc A310 wins the Passmark DirectX 12 test with a score of 29, which is 16% higher than the NVIDIA T600’s 25. This is one of the A310’s three head-to-head wins.

Q: Which card has the higher memory bandwidth?

A: The NVIDIA T600 has a higher memory bandwidth of 160.0 GB/s, compared to the Intel Arc A310’s 124.0 GB/s. The T600 also has a wider 128-bit memory bus versus the A310’s 64-bit bus.

Q: Are both cards single-slot designs with no power connectors?

A: Yes, both the Intel Arc A310 and the NVIDIA T600 are single-slot cards that require no additional power connectors. Both also have a suggested PSU of 200 W.

---

Architecture Differences

The Intel Arc A310 is built on Intel’s Xe-HPG architecture, specifically the DG2-128 chip, and belongs to the Alchemist (Arc 3) generation. It is fabricated on a 6 nm process at TSMC, packing 7,200 million transistors into a 157 mm² die, resulting in a transistor density of 45.9M per mm². The NVIDIA T600, in contrast, uses the Turing architecture with the TU117 chip, part of the Quadro Turing (Tx000) generation. It is built on a 12 nm process, also at TSMC, with 4,700 million transistors on a larger 200 mm² die, giving a lower transistor density of 23.5M per mm².

The A310 features 768 shading units, 32 texture mapping units, and 16 ROPs, along with 6 dedicated ray tracing cores. The T600 has fewer shading units at 640, but more TMUs (40) and twice the ROPs (32). Notably, the T600 has no dedicated ray tracing cores, while the A310 includes them. Neither card lists tensor cores in the provided data.

The memory subsystems differ significantly. Both cards have 4 GB of GDDR6 memory, but the A310 uses a 64-bit bus with a bandwidth of 124.0 GB/s, while the T600 uses a 128-bit bus with a bandwidth of 160.0 GB/s. The A310’s memory runs at an effective 15.5 Gbps, while the T600’s runs at 10 Gbps effective.

Clock speeds also diverge. The A310 has a base and boost clock of 1750 MHz, while the T600 has a base clock of 735 MHz and a boost clock of 1335 MHz. This leads to different compute rates: the A310 delivers 2.688 TFLOPS FP32 and 5.376 TFLOPS FP16 (2:1), whereas the T600 delivers 1.709 TFLOPS FP32 and 3.418 TFLOPS FP16 (2:1). Pixel and texture rates favor each card differently: the A310 has a pixel rate of 28.00 GPixel/s and a texture rate of 56.00 GTexel/s, while the T600 has a higher pixel rate of 42.72 GPixel/s but a slightly lower texture rate of 53.40 GTexel/s.

The A310 supports PCIe 4.0 x8, while the T600 uses PCIe 3.0 x16. Display outputs also differ: the A310 has 4x mini-DisplayPort 2.0, whereas the T600 has 4x mini-DisplayPort 1.4a. In terms of API support, the A310 supports DirectX 12 Ultimate (12_2), while the T600 supports DirectX 12 (12_1); both support OpenGL 4.6 and Vulkan 1.4. Power consumption is low for both, with the A310 at 30 W and the T600 at 40 W.

---

Head-to-Head Benchmarks

The benchmark data shows a clear split between the two cards. The Intel Arc A310 wins 3 of the 9 head-to-head tests, while the NVIDIA T600 wins 6. The A310’s biggest victory comes in the Geekbench Vulkan test, where it scores 28964 against the T600’s 25580, a 13.2% advantage. In Geekbench OpenCL, the A310 scores 30607 versus 27875, a 9.8% lead. The A310 also wins the Passmark DirectX 12 test with a score of 29 versus 25, a 16% margin.

The NVIDIA T600 dominates the remaining tests. Its largest win is in Passmark DirectX 9, where it scores 114 against the A310’s 69, a 39.5% gap. In Passmark DirectX 11, the T600 scores 49 versus 33, a 32.7% lead. The T600 also wins Passmark G3D with 6479 versus 5433 (a 16.1% difference) and Passmark G2D with 756 versus 625 (a 17.3% difference). In Passmark GPU Compute, the T600 scores 2402 against the A310’s 2157, a 10.2% margin. The narrowest T600 win is in Passmark DirectX 10, where it scores 32 versus 31, a 3.1% edge.

These results indicate the A310 excels in modern, compute-oriented and Vulkan workloads, while the T600 is stronger in legacy DirectX 9/10/11 titles and general 2D/3D rasterization. The A310’s higher FP32 throughput (2.688 TFLOPS vs 1.709 TFLOPS) likely contributes to its Geekbench wins, while the T600’s higher pixel rate (42.72 GPixel/s vs 28.00 GPixel/s) helps in DirectX 9 and G3D tests.

---

Specification Differences

The two cards differ in several key specification fields. The most notable differences are:

• Chip and Architecture: Intel DG2-128 (Xe-HPG) vs NVIDIA TU117 (Turing)
• Process Node: 6 nm vs 12 nm
• Transistors: 7,200 million vs 4,700 million
• Die Size: 157 mm² vs 200 mm²
• Transistor Density: 45.9M / mm² vs 23.5M / mm²
• Base Clock: 1750 MHz vs 735 MHz
• Boost Clock: 1750 MHz vs 1335 MHz
• Memory Clock: 1937 MHz (15.5 Gbps effective) vs 1250 MHz (10 Gbps effective)
• Memory Bus Width: 64 bit vs 128 bit
• Memory Bandwidth: 124.0 GB/s vs 160.0 GB/s
• Shading Units: 768 vs 640
• TMUs: 32 vs 40
• ROPs: 16 vs 32
• Ray Tracing Cores: 6 vs None
• Pixel Rate: 28.00 GPixel/s vs 42.72 GPixel/s
• Texture Rate: 56.00 GTexel/s vs 53.40 GTexel/s
• FP32 Performance: 2.688 TFLOPS vs 1.709 TFLOPS
• FP16 Performance: 5.376 TFLOPS vs 3.418 TFLOPS
• TDP: 30 W vs 40 W
• Bus Interface: PCIe 4.0 x8 vs PCIe 3.0 x16
• Display Outputs: 4x mini-DisplayPort 2.0 vs 4x mini-DisplayPort 1.4a
• DirectX Support: 12 Ultimate (12_2) vs 12 (12_1)
• Release Date: 2022-10-11 vs 2021-04-11
• Predecessor: Xe Graphics vs Quadro Volta
• Successor: Battlemage vs Workstation Ampere

Both cards share the same memory size (4 GB GDDR6), single-slot design, no power connectors, 200 W suggested PSU, OpenGL 4.6, Vulkan 1.4, and end-of-life production status.

---

Where Each One Wins

Intel Arc A310 wins in:

• Modern API performance: It leads in Geekbench Vulkan (13.2%) and OpenCL (9.8%), plus Passmark DirectX 12 (16%). This makes it the better pick for workloads that leverage Vulkan, compute shaders, or DirectX 12 Ultimate features.
• Raw compute throughput: With 2.688 TFLOPS FP32 and 5.376 TFLOPS FP16, the A310 offers significantly higher theoretical compute than the T600, which explains its Geekbench leads.
• Ray tracing support: The A310 includes 6 ray tracing cores, which the T600 lacks entirely. This is a functional advantage for any application that can use hardware ray tracing.
• Efficiency per watt: The A310 achieves its higher average score (7550) at a lower TDP of 30 W, compared to the T600’s 40 W.

NVIDIA T600 wins in:

• Legacy DirectX performance: The T600 dominates in Passmark DirectX 9 (39.5% lead) and DirectX 11 (32.7% lead), along with a smaller win in DirectX 10 (3.1%). This makes it the stronger choice for older games or applications that rely on these APIs.
• Rasterization and 2D/3D graphics: The T600 wins Passmark G3D (16.1%) and G2D (17.3%), indicating better overall rasterization and desktop/graphics compositing performance.
• Compute in specific tests: The T600 also wins Passmark GPU Compute (10.2%), despite lower theoretical FP32, suggesting better optimization in that particular benchmark.
• Memory bandwidth: The T600’s 160.0 GB/s bandwidth and 128-bit bus provide more headroom for bandwidth-sensitive tasks, even though the A310 has faster effective memory clock (15.5 Gbps vs 10 Gbps).

---

The Verdict

The data paints a clear picture: the Intel Arc A310 is the compute-focused modern card, while the NVIDIA T600 is the rasterization-focused legacy card. The A310’s higher average benchmark score (7550 vs 7035) and wins in Vulkan, OpenCL, and DirectX 12 make it the better choice for users running current-generation games, compute-heavy workloads, or any application that benefits from hardware ray tracing. Its 6 nm process and 7,200 million transistors also suggest a more advanced design.

However, the NVIDIA T600 is not without merit. Its wins in DirectX 9, 10, and 11, plus G2D and G3D, mean it is better suited for older software libraries, 2D-heavy professional environments, or scenarios where the T600’s higher pixel rate (42.72 GPixel/s) and wider memory bus (128-bit) matter more than raw FP32 compute. The T600’s lead in Passmark GPU Compute (10.2%) also shows that theoretical specs don’t always translate to real-world wins.

For a builder deciding between the two, the choice comes down to workload. If the priority is modern APIs, compute throughput, and ray tracing, the Intel Arc A310 is the data-backed pick. If the priority is legacy DirectX compatibility, 2D/3D rasterization, and higher memory bandwidth, the NVIDIA T600 is the safer bet. Both cards are end-of-life, single-slot, and require no power connectors, so installation is straightforward for either. The A310’s 30 W TDP is lower, but the T600’s 40 W is still modest. Ultimately, the A310 wins on average performance and future-facing features, while the T600 wins on compatibility and specific legacy benchmarks.