NVIDIA RTX A6000

NVIDIA’s RTX A6000 is a workstation-class GPU built on the Ampere architecture, utilizing the GA102 chip fabricated on Samsung’s 8 nm process. With 28,300 million transistors on a 628 mm² die, this card targets professional visualization and compute workloads. Released on October 4, 2020, it is now designated as end-of-life, with its successor being the Workstation Ada generation. The card carries a launch MSRP of 4,649 USD. This analysis is based on the available benchmark data, which shows a mixed but generally competitive performance profile against its nearest rivals.

Benchmark Performance

The RTX A6000’s average benchmark score across all recorded tests is 42,653, placing it in the 84th percentile of all GPUs. This is a strong overall showing, but the data reveals a nuanced picture when compared directly to its nearest rivals. The card's performance is virtually identical to the AMD Radeon RX 7650 GRE, with a negligible delta of -0.2% (the RX 7650 GRE scores 42,723, slightly higher). Similarly, it is just 0.2% ahead of the NVIDIA GeForce RTX 4070 SUPER, which scores 42,576. These are effectively statistical ties, indicating that the A6000’s aggregate performance is on par with these mainstream and enthusiast-class cards despite its workstation pedigree.

The comparison with older professional hardware is more decisive. The RTX A6000 outperforms the NVIDIA Quadro M6000 by 1.5% (43,313 vs. 42,653), a modest generational gain that underscores the diminishing returns in raw aggregate compute for professional cards over time. Against the AMD Radeon Pro 580X, the A6000 holds a 1.6% advantage (41,991 vs. 42,653). These slim margins in the average score, however, belie significant differences in specific workloads. The benchmark suite shows a wide variance in performance: the card achieves a strong Passmark G3D score of 22,577 and a Geekbench OpenCL score of 180,149, while its Passmark DirectX 12 score is a surprisingly low 87. This suggests that the A6000’s architecture is heavily optimized for compute-centric tasks rather than legacy or specific graphics API paths.

In compute-oriented tests, the GPU demonstrates its professional focus. The Passmark GPU Compute score of 14,110 is robust, and the Geekbench Vulkan score of 165,451 indicates solid performance in modern, low-overhead APIs. The raw specifications support this: the card delivers 38.71 TFLOPS of FP32 and FP16 performance (at a 1:1 ratio), with a texture fill rate of 604.8 GTexel/s and a pixel rate of 201.6 GPixel/s. These figures show that while the A6000 may not lead in every synthetic graphics test, its compute throughput is the primary driver of its high average score and its position in the 84th percentile. The data indicates a card that trades blows with newer consumer GPUs on aggregate but provides a specialized performance profile aimed at professional applications.

Ray Tracing and Feature Set

The RTX A6000 is equipped with a full complement of Ampere’s specialized hardware. It includes 84 RT cores and 336 Tensor cores, which are essential for ray-traced workloads and AI-accelerated tasks. The presence of these cores is reflected in the API support: the card is compliant with DirectX 12 Ultimate (12_2), which mandates hardware ray tracing and variable rate shading. This ensures compatibility with the latest gaming and professional rendering engines that leverage these features. The card also supports OpenGL 4.6 and Vulkan 1.4, providing broad cross-platform API coverage for professional visualization suites.

The feature set extends to memory and connectivity. With 48 GB of GDDR6 memory on a 384-bit bus, the A6000 offers 768.0 GB/s of bandwidth, a configuration designed for massive datasets in scientific computing, AI training, and high-resolution rendering. The PCIe 4.0 x16 interface ensures adequate data transfer speeds for host-to-device communication. Display output is handled by four DisplayPort 1.4a connectors, supporting high-resolution multi-monitor setups. The high transistor density of 45.1M / mm² and the 8 nm process node from Samsung are architectural details that define the card's power envelope and performance ceiling, but the benchmark data focuses on its computational capabilities, which are substantial in ray tracing and compute.

Power and Cooling

The RTX A6000 has a thermal design power (TDP) of 300 W, a figure that dictates its cooling and power supply requirements. The card is a dual-slot design, measuring 267 mm (10.5 inches) in length and 112 mm (4.4 inches) in height, making it a sizable but standard professional form factor. Power is delivered via a single 8-pin EPS connector, which is a notable choice, as EPS connectors are typically used for CPU power but are employed here for the GPU’s high draw. The manufacturer recommends a 700 W power supply for systems using this card.

This power requirement is consistent with the card’s high-end specifications, including 10,752 shading units, 336 TMUs, and 112 ROPs. The 300 W TDP is a substantial heat load that requires an effective cooling solution, and the dual-slot design indicates a robust heatsink and fan assembly. The data does not provide specific thermal performance figures, but the power connector and PSU recommendation are clear indicators of the system-level requirements. Users must ensure their power supply has the necessary 8-pin EPS cable and sufficient wattage headroom, as the 700 W suggestion is a baseline for the entire system, not just the GPU.

FAQ

Q: What is the average benchmark score of the RTX A6000?

A: The RTX A6000 has an average benchmark score of 42,653, which places it in the 84th percentile of all GPUs.

Q: How does the RTX A6000 compare to the GeForce RTX 4070 SUPER?

A: The RTX A6000 is 0.2% ahead of the RTX 4070 SUPER in average benchmark score (42,653 vs. 42,576), making the two cards effectively equal in aggregate performance.

Q: What is the memory configuration of the RTX A6000?

A: The card features 48 GB of GDDR6 memory with a 384-bit bus width, providing a memory bandwidth of 768.0 GB/s.

Q: What power supply is recommended for the RTX A6000?

A: The manufacturer recommends a 700 W power supply, and the card requires a single 8-pin EPS power connector for operation.

Q: Does the RTX A6000 support hardware ray tracing?

A: Yes, it includes 84 RT cores and is compliant with DirectX 12 Ultimate (12_2), which includes hardware ray tracing support.

Q: What is the production status and launch price of the RTX A6000?

A: The RTX A6000 is end-of-life, and its launch MSRP was 4,649 USD.

How It Compares

AMD Radeon RX 7650 GRE: The data shows a near-exact tie between these two cards. The RX 7650 GRE has an average score of 42,723, which is 0.2% higher than the A6000’s 42,653. In terms of raw aggregate benchmarks, there is no meaningful winner, suggesting that the A6000’s professional features and memory capacity are not reflected in these particular tests, but its compute performance is competitive with this Radeon offering.

NVIDIA GeForce RTX 4070 SUPER: The A6000 edges out the RTX 4070 SUPER by a slim 0.2% margin (42,653 vs. 42,576). This is a significant finding, as it shows the older workstation card keeping pace with a newer, consumer-focused Ada Lovelace GPU in overall benchmark averages. The performance is essentially identical, highlighting that the A6000’s Ampere architecture remains relevant in compute-heavy scenarios.

NVIDIA Quadro M6000: The A6000 leads its professional predecessor, the Quadro M6000, by 1.5% in average score (42,653 vs. 43,313). Wait, the data indicates the M6000 actually scores higher at 43,313, meaning the A6000 is 1.5% behind the M6000. This is an unexpected result in the aggregate, showing that generational improvements in this professional line did not translate to a higher average score in this benchmark set, likely due to the M6000’s strength in specific legacy tests.

AMD Radeon Pro 580X: The A6000 outperforms the Radeon Pro 580X by 1.6% in average score (42,653 vs. 41,991). This is a modest but consistent advantage, reinforcing the A6000’s position as a more capable compute and graphics solution for professional workloads. The performance gap is narrow, but the A6000’s modern architecture and feature set provide a clear edge over this older competitor.