Intel Arc A770M vs NVIDIA GeForce GTX 690 Comparison

Intel Arc A770M and NVIDIA GeForce GTX 690 represent two very different eras of GPU design, and the benchmark data reflects that gap starkly. The Arc A770M, built on a modern 6 nm process with 21,700 million transistors, completely outclasses the older 28 nm GTX 690, which packs only 3,540 million transistors. In the two shared benchmarks, the Intel part wins decisively, with a 414.4% lead in Geekbench OpenCL and a 346.3% lead in Geekbench Vulkan. The GTX 690, a dual-GPU card from 2012, still holds a respectable 60th percentile ranking among all GPUs, but the Arc A770M sits slightly higher at the 62nd percentile, suggesting that while the Intel part is faster, the gap in overall standings is narrower than the raw compute scores imply.

The Verdict

The data presents a clear choice for different user profiles. The Intel Arc A770M is the superior performer in every measurable benchmark shared between the two cards. It wins both head-to-head tests, with Geekbench OpenCL scores of 89,494 versus 17,399, and Geekbench Vulkan scores of 74,422 versus 16,675. For anyone prioritizing raw compute or modern API support, the Arc A770M is the only rational pick. It also offers 16 GB of GDDR6 memory with 512.0 GB/s bandwidth, compared to the GTX 690's 2 GB GDDR5 at 192.3 GB/s, making it vastly more capable for modern workloads that demand large memory pools.

However, the GTX 690 is not without its niche. Its average benchmark score of 17,037 places it within 1% of the Arc A770M's nearest rivals, including the AMD Radeon RX 460 (18,373) and the NVIDIA GeForce RTX 3060 Mobile (18,159). The GTX 690 also holds a 60th percentile ranking, only two points below the Arc A770M's 62nd. For a card released on May 2, 2012, with a 999 USD launch MSRP, this longevity is noteworthy. Yet, the GTX 690's 300 W TDP and dual-slot design with two 8-pin power connectors demand a 700 W PSU, whereas the Arc A770M is an IGP with a 120 W TDP. The verdict is straightforward: the Arc A770M wins on performance, efficiency, and modern features; the GTX 690 is a historical artifact that only makes sense for legacy system collectors, not for practical use.

Architecture Differences

The architectural divide between these two GPUs is generational. The Intel Arc A770M uses the DG2-512 chip based on the Xe-HPG architecture, part of the Alchemist generation for Arc 7 Mobile. It is fabricated on a 6 nm process at TSMC, with a die size of 406 mm² and a transistor density of 53.4 million per mm². In contrast, the GTX 690 uses the GK104 chip with the Kepler architecture, part of the GeForce 600 generation, built on a 28 nm process at the same foundry, with a 294 mm² die and a transistor density of just 12.0 million per mm². The Arc A770M's transistor count of 21,700 million is over six times higher than the GTX 690's 3,540 million, which explains the massive compute advantage.

Core configurations differ equally. The Arc A770M features 4,096 shading units, 256 TMUs, and 128 ROPs, along with 32 dedicated ray tracing cores. The GTX 690 has 1,536 shading units, 128 TMUs, and just 32 ROPs, with no ray tracing support. Clock speeds favor the older card in raw frequency—GTX 690 runs at 915 MHz base and 1019 MHz boost, while the Arc A770M runs at 1650 MHz base and 2050 MHz boost—but the Intel part's higher clocks and massively higher core count deliver a 16.79 TFLOPS FP32 rate versus the GTX 690's 3.130 TFLOPS. Memory technology also diverges: the Arc A770M uses GDDR6 at 2000 MHz (16 Gbps effective) with a 256-bit bus, while the GTX 690 uses GDDR5 at 1502 MHz (6 Gbps effective) with the same 256-bit bus width, resulting in 512.0 GB/s versus 192.3 GB/s bandwidth.

API support further separates them. The Arc A770M supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, while the GTX 690 only reaches DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The Arc A770M also supports PCIe 4.0 x16, whereas the GTX 690 is limited to PCIe 3.0 x16. Physically, the GTX 690 is a dual-slot card measuring 279 mm in length, 111 mm in height, and 38 mm in width, with three DVI outputs and one mini-DisplayPort 1.2. The Arc A770M is an IGP with portable-device-dependent display outputs, meaning it is designed for laptops rather than desktop expansion.

Where Each One Wins

The Intel Arc A770M wins decisively in every benchmark category where both cards have data. In Geekbench OpenCL, it scores 89,494 against the GTX 690's 17,399, a 414.4% advantage. In Geekbench Vulkan, it scores 74,422 against 16,675, a 346.3% lead. These results indicate the Arc A770M is the clear choice for compute-heavy tasks like OpenCL-based rendering, machine learning inference, or Vulkan gaming. Its 16 GB memory and 512.0 GB/s bandwidth also make it suitable for large datasets or high-resolution textures, where the GTX 690's 2 GB capacity would be a hard bottleneck.

The GTX 690, however, wins in the context of legacy compatibility and niche scenarios. Its average benchmark score of 17,037 places it within 0.1% of the AMD Radeon HD 7970M (17,019) and within 1% of the NVIDIA GeForce RTX 3070 (17,208), showing it still competes with mid-range cards from later generations in aggregate benchmarks. It also has a higher average score than the AMD Radeon RX 7600 XT (17,083) by 0.3%, per the deltaPct of -0.3% in that rival's comparison. For users running older DirectX 11 or Vulkan 1.2 applications, the GTX 690's drivers and architecture may be more stable, though the data does not quantify this. The GTX 690's dual-GPU design was originally flagship-class, and its 60th percentile ranking indicates it still outperforms 60% of all GPUs in the database.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The Intel Arc A770M has an average benchmark score of 18,383, while the NVIDIA GeForce GTX 690 has 17,037. The Arc A770M is 7.9% higher.

Q: How does the GTX 690 compare to its nearest rivals?

A: The GTX 690's average score of 17,037 is within 0.1% of the AMD Radeon HD 7970M (17,019) and 0.6% of the NVIDIA Tesla M4 (16,932). It is 0.3% behind the AMD Radeon RX 7600 XT (17,083) and 1% behind the NVIDIA GeForce RTX 3070 (17,208).

Q: What are the memory specifications for each card?

A: The Arc A770M has 16 GB of GDDR6 with a 256-bit bus and 512.0 GB/s bandwidth. The GTX 690 has 2 GB of GDDR5 with a 256-bit bus and 192.3 GB/s bandwidth.

Q: Does the GTX 690 support ray tracing?

A: No, the GTX 690 has no ray tracing cores listed. The Arc A770M has 32 RT cores.

Q: What is the power consumption difference?

A: The Arc A770M has a TDP of 120 W, while the GTX 690 has a TDP of 300 W. The GTX 690 also requires a 700 W suggested PSU and dual 8-pin connectors.

Q: Which GPU has better API support?

A: The Arc A770M supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The GTX 690 supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175.

Head-to-Head Benchmarks

The two shared benchmarks tell a story of total dominance. In Geekbench OpenCL, the Intel Arc A770M scores 89,494, which is 414.4% higher than the GTX 690's 17,399. This is not a marginal win; it is a five-fold performance advantage. The OpenCL test typically stresses raw compute throughput, and the Arc A770M's 16.79 TFLOPS FP32 rate, combined with its 4,096 shading units, crushes the GTX 690's 3.130 TFLOPS and 1,536 shading units. Even accounting for the GTX 690's dual-GPU design, which historically had scaling inefficiencies, the modern Intel part's single-die architecture delivers more consistent performance.

In Geekbench Vulkan, the gap narrows slightly but remains enormous. The Arc A770M scores 74,422 versus the GTX 690's 16,675, a 346.3% lead. Vulkan is a low-overhead API that favors modern hardware with better driver optimization. The GTX 690's Vulkan support tops out at version 1.2.175, while the Arc A770M supports Vulkan 1.4, allowing it to leverage newer features and better scheduling. The GTX 690's older Kepler architecture also lacks the dedicated RT cores found in the Arc A770M, though Vulkan benchmarks do not directly measure ray tracing. The data shows that in any modern API workload, the Arc A770M is the only viable option.

Beyond these two tests, the Arc A770M has a broader benchmark portfolio, including 3DMark Steel Nomad DX12 (score 2,278), PassMark G3D (11,774), and PassMark GPU Compute (4,778). The GTX 690 has no data in these tests, so direct comparisons cannot be made. However, the Arc A770M's PassMark DirectX 12 score of 70 and DirectX 11 score of 69, alongside its 62nd percentile ranking, suggest it is a well-rounded performer. The GTX 690's 60th percentile, despite its age, is impressive but ultimately insufficient to close the gap. The head-to-head results are unambiguous: the Arc A770M wins 2-0, with wins in both Geekbench tests, and the victory margins are not close.